From bbaetz at acm.org Sat Mar 1 00:25:26 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Sat, 1 Mar 2003 11:25:26 +1100 Subject: Support for multiple "Locations" in bugzilla In-Reply-To: <3E5FE71E.6090700@mozilla.org> References: <20030228160833.GH16925@somanetworks.com> <20030228122106.G29082@sigkill.com> <20030228204414.GJ29540@somanetworks.com> <3E5FCC97.1000700@mozilla.org> <20030228210919.GK29540@somanetworks.com> <3E5FD267.8040306@mozilla.org> <20030228213030.GN29540@somanetworks.com> <3E5FDA74.4040407@mozilla.org> <20030228222205.GA1614@mango.home> <3E5FE71E.6090700@mozilla.org> Message-ID: <20030301002525.GA2019@mango.home> On Fri, Feb 28, 2003 at 10:47:58PM +0000, Gervase Markham wrote: > > A Google for 'database "mapping tables"' doesn't define the term. Do you > mean lots of tables like: No, I mean tables like: bug_id, field_id, value We will need several of those, however - one for string types, and one for number types, so that indexes work optimally. For stuff like os's, (field_id, value) is an FK [Foreign Key] to another table which has the name<-> number mapping. (If we ever support email types that way, then wed need a third one, to fk to the profiles table) > with slight variations depending on whether it's a 1->1 or 1->many or > enum-style mapping? And doing joins everywhere to connect it all up? 1->1 vs 1->many is only how many entries there are. We may want (another) separate table for that to add the UNIQUE constraint - I haven't thought too much about that. > > >Hows postgres support coming along? :) > > However it's coming along, MySQL support isn't going away... so any > solution has to work sensibly in both. > I want mysql to go away :) But yeah, its not happening in the short term. > Gerv > Bradley From bbaetz at acm.org Sat Mar 1 01:47:17 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Sat, 1 Mar 2003 12:47:17 +1100 Subject: Support for multiple "Locations" in bugzilla In-Reply-To: <20030301002525.GA2019@mango.home> References: <20030228122106.G29082@sigkill.com> <20030228204414.GJ29540@somanetworks.com> <3E5FCC97.1000700@mozilla.org> <20030228210919.GK29540@somanetworks.com> <3E5FD267.8040306@mozilla.org> <20030228213030.GN29540@somanetworks.com> <3E5FDA74.4040407@mozilla.org> <20030228222205.GA1614@mango.home> <3E5FE71E.6090700@mozilla.org> <20030301002525.GA2019@mango.home> Message-ID: <20030301014717.GA2465@mango.home> On Sat, Mar 01, 2003 at 11:25:26AM +1100, Bradley Baetz wrote: > > I want mysql to go away :) But yeah, its not happening in the short > term. ... and theres a reason for this, BTW. Consider a query to find all the bugs where I am the assignee, or have added a comment, using postgresql-7.4cvs (ie not released), and mysql-3.23.54a For 20,000 users, 100,000 bugs, and 1,000,000 comments randomly distributed: > select distinct bugs.bug_id FROM bugs LEFT JOIN longdescs USING(bug_id) WHERE (bugs.assigned_to=86 OR longdescs.who=86); mysql: 3.33 sec postgres: 14423.07 msec [I can get this down to 11758.74 msec by changing a config option - I should probably file a bug on that] (returning 46 rows) Now, that looks pretty bad for pg. _BUT_, that query isn't what we really wanted to know. > SELECT bugs.bug_id FROM bugs WHERE bugs.assigned_to=86 OR bugs.bug_id IN (SELECT bug_id FROM longdescs WHERE longdescs.who=86) mysql: error postgres: 137.66 msec (The rather large disclaimer here is that that last result needs cvs-postgres to be smart with the subselect. And before anyone asks, the first query isn't the same as the second without the distinct, and pg doesn't pass the distinct from the top down to lower layers, AFAIK, so this can't be easily done automatically by the optimiser) I can get it down to 0.65 msec (no, that is not a typo) if I use a union query, but thats not the sort of thing we can generate from Bugzilla::Search ourselves, its more a job for the optimiser, which probably isn't useful in most queries - its just because this one has no other conditions. Now, heres something even more fun. With innodb tables, the first query takes 111.62 sec, +/- 10sec. 4.0 may be faster I guess, and I haven't played with any of the (many) tuning paramaters. I also may have just found a particularly bad test case or something. All times are from the 2nd run through, plus after a VACUUM ANALYZE for pg. Bradley From gerv at mozilla.org Sun Mar 2 12:34:06 2003 From: gerv at mozilla.org (Gervase Markham) Date: Sun, 02 Mar 2003 12:34:06 +0000 Subject: bug life cycle in Bugzilla In-Reply-To: <000d01c2d7fa$7bb2b3a0$800310ac@tripathi.bhartitelesoft.com> References: <002301c2d745$6587a960$800310ac@tripathi.bhartitelesoft.com> <20030227040954.B23574@sigkill.com> <000d01c2d7fa$7bb2b3a0$800310ac@tripathi.bhartitelesoft.com> Message-ID: <3E61FA3E.9020403@mozilla.org> geeta wrote: > Hello All, > > Now there are queries regarding Bugzilla which needs to be addressed very > soon, please let me know incase any code changes required: geeta: It's very difficult to keep track of your messages, because the clock on your computer or mailserver is set incorrectly, and so they always appear to be nine days old. Please could you fix this? Thanks :-) Gerv From gerv at mozilla.org Sun Mar 2 12:35:04 2003 From: gerv at mozilla.org (Gervase Markham) Date: Sun, 02 Mar 2003 12:35:04 +0000 Subject: Fw: bug life cycle in Bugzilla In-Reply-To: <001001c2d7ff$971715a0$800310ac@tripathi.bhartitelesoft.com> References: <001001c2d7ff$971715a0$800310ac@tripathi.bhartitelesoft.com> Message-ID: <3E61FA78.5060101@mozilla.org> > Sorry for bugging you again and again................. > Further to this, I need to add a new field in my bug report and query and > also a timestamp for how much time it took to resolve a bug.Although I do > see in the query form the field for the Bug Changes. If you get 2.17.3 you can turn on the timetracking feature, which lets you do this. Gerv From gerv at mozilla.org Sun Mar 2 12:35:35 2003 From: gerv at mozilla.org (Gervase Markham) Date: Sun, 02 Mar 2003 12:35:35 +0000 Subject: bug life cycle in Bugzilla In-Reply-To: <000d01c2d7fa$7bb2b3a0$800310ac@tripathi.bhartitelesoft.com> References: <002301c2d745$6587a960$800310ac@tripathi.bhartitelesoft.com> <20030227040954.B23574@sigkill.com> <000d01c2d7fa$7bb2b3a0$800310ac@tripathi.bhartitelesoft.com> Message-ID: <3E61FA97.3010301@mozilla.org> geeta wrote: > 1. Suppose while identifying the bug , my source of bug changes , do I have > to consider the components as a source can a bug id be tracked accordingly. > I see a Bug Activity Log > > 2. When I close and fix a bug, I run a query against that bug , I don't get > the results . I'm afraid I don't understand the question. Could you explain a bit further? Gerv From gerv at mozilla.org Sun Mar 2 17:53:54 2003 From: gerv at mozilla.org (Gervase Markham) Date: Sun, 02 Mar 2003 17:53:54 +0000 Subject: Living without Custom Fields Message-ID: <3E624532.5030203@mozilla.org> As the whole custom field thing is turning into a bit of a FAQ, I wrote this: http://www.gerv.net/hacking/custom-fields.html I plan to post the URL to the newsgroup in a few days, and then use it to answer anyone who comes in asking about custom fields. Comments and suggestions on the text are welcome. Gerv From preed at sigkill.com Sun Mar 2 19:47:25 2003 From: preed at sigkill.com (J. Paul Reed) Date: Sun, 2 Mar 2003 11:47:25 -0800 Subject: Living without Custom Fields In-Reply-To: <3E624532.5030203@mozilla.org>; from gerv@mozilla.org on Sun, Mar 02, 2003 at 05:53:54PM +0000 References: <3E624532.5030203@mozilla.org> Message-ID: <20030302114725.A25322@sigkill.com> On 02 Mar 2003 at 17:53:54, Gervase Markham moved bits on my disk to say: > As the whole custom field thing is turning into a bit of a FAQ, I wrote > this: > > http://www.gerv.net/hacking/custom-fields.html > > I plan to post the URL to the newsgroup in a few days, and then use it to > answer anyone who comes in asking about custom fields. Comments and > suggestions on the text are welcome. If this is going to the project's "official answer" to this feature, then I would change the following: Option 0: "It points out that several of the reasons often given for adding a custom field are bogus <, including the main one: you often don't need them. ... or, whatever his main reason was. I haven't read that piece for awhile >" Option 1: "does Bugzilla already support what you want and you just need to turn the feature on?" Option 2: "Your organisation or project may have **inexperienced** people filing bugs, " You spelled organization wrong, too. ;-) Anyway, I think it's a good document, but if we're going to point people at it, I think it needs to be a bit... friendlier. We shouldn't act like people are stupid for wanting custom fields, which I think some of this text borders on (mostly what I noted in 1 and 2). Later, Paul ------------------------------------------------------------------------ J. Paul Reed -- 0xDF8708F8 || preed at sigkill.com || web.sigkill.com/preed To hold on to sanity too tight is insane. -- Nick Falzone, Pushing Tin I use PGP; you should use PGP too... if only to piss off John Ashcroft From geeta.tripathi at bhartitelesoft.com Mon Mar 3 05:36:15 2003 From: geeta.tripathi at bhartitelesoft.com (geeta) Date: Mon, 3 Mar 2003 11:06:15 +0530 Subject: bug life cycle in Bugzilla References: <002301c2d745$6587a960$800310ac@tripathi.bhartitelesoft.com> <20030227040954.B23574@sigkill.com> <000d01c2d7fa$7bb2b3a0$800310ac@tripathi.bhartitelesoft.com> <3E61FA97.3010301@mozilla.org> Message-ID: <001701c2e146$d3bc79a0$800310ac@tripathi.bhartitelesoft.com> Hello Gervase, I We have implemented 2.16.1 in our organisation. There are few concerns.. 1. What features an be used to know the efforts that takes to resolve a bug. & 2. Suppose in a buglife cycle it is identified that the bug is not in component1 and component2 , is there any feature in Bugzilla can be utilised for the defect leakage per component basis as in the summary reports we don't have these details. We would be needing these details for our quality processes and I have to convince my Quality Guys here........sorry to bug you. If these features are not supported do I need to add some new fields and do some customisations. Or do I have to upgrade 2.16.1 to anyother versions that supports these features. Looking forward to hear from you. I would be grateful if you could give me your ideas on the same Thanks and regards, Geeta ----- Original Message ----- From: "Gervase Markham" To: Sent: Sunday, March 02, 2003 6:05 PM Subject: Re: bug life cycle in Bugzilla > geeta wrote: > > 1. Suppose while identifying the bug , my source of bug changes , do I have > > to consider the components as a source can a bug id be tracked accordingly. > > I see a Bug Activity Log > > > > 2. When I close and fix a bug, I run a query against that bug , I don't get > > the results . > > I'm afraid I don't understand the question. Could you explain a bit further? > > Gerv > > ---- > To view or change your list settings, click here: > > From gerv at mozilla.org Mon Mar 3 08:43:28 2003 From: gerv at mozilla.org (Gervase Markham) Date: Mon, 03 Mar 2003 08:43:28 +0000 Subject: Living without Custom Fields In-Reply-To: <20030302114725.A25322@sigkill.com> References: <3E624532.5030203@mozilla.org> <20030302114725.A25322@sigkill.com> Message-ID: <3E6315B0.505@mozilla.org> J. Paul Reed wrote: > If this is going to the project's "official answer" to this feature, then I > would change the following: Well, currently it's not official in any way. > Anyway, I think it's a good document, but if we're going to point people at > it, I think it needs to be a bit... friendlier. We shouldn't act like > people are stupid for wanting custom fields, which I think some of this > text borders on (mostly what I noted in 1 and 2). It's a fair point. You know me: I speak my mind :-) But I'll tone it down a bit. Gerv From gerv at mozilla.org Mon Mar 3 08:58:24 2003 From: gerv at mozilla.org (Gervase Markham) Date: Mon, 03 Mar 2003 08:58:24 +0000 Subject: bug life cycle in Bugzilla In-Reply-To: <001701c2e146$d3bc79a0$800310ac@tripathi.bhartitelesoft.com> References: <002301c2d745$6587a960$800310ac@tripathi.bhartitelesoft.com> <20030227040954.B23574@sigkill.com> <000d01c2d7fa$7bb2b3a0$800310ac@tripathi.bhartitelesoft.com> <3E61FA97.3010301@mozilla.org> <001701c2e146$d3bc79a0$800310ac@tripathi.bhartitelesoft.com> Message-ID: <3E631930.6000100@mozilla.org> > 1. What features an be used to know the efforts that takes to resolve a > bug. Turn on Time Tracking. You will need 2.17.3 for this. > & > 2. Suppose in a buglife cycle it is identified that the bug is not in > component1 and component2 , is there any feature in Bugzilla can be > utilised for the defect leakage per component basis as in the summary > reports we don't have these details. You could construct a (fairly complicated) search against the Bugs Activity table to tell you how many bugs had been moved from one component to another over a certain amount of time, yes. But there's no pretty report which tells you. Gerv From geeta.tripathi at bhartitelesoft.com Mon Mar 3 09:12:23 2003 From: geeta.tripathi at bhartitelesoft.com (geeta) Date: Mon, 3 Mar 2003 14:42:23 +0530 Subject: bug life cycle in Bugzilla References: <002301c2d745$6587a960$800310ac@tripathi.bhartitelesoft.com> <20030227040954.B23574@sigkill.com> <000d01c2d7fa$7bb2b3a0$800310ac@tripathi.bhartitelesoft.com> <3E61FA97.3010301@mozilla.org> <001701c2e146$d3bc79a0$800310ac@tripathi.bhartitelesoft.com> <3E631930.6000100@mozilla.org> Message-ID: <003901c2e165$0443e4a0$800310ac@tripathi.bhartitelesoft.com> Thanks again Gervase. regards, Geeta ----- Original Message ----- From: "Gervase Markham" To: Sent: Monday, March 03, 2003 2:28 PM Subject: Re: bug life cycle in Bugzilla > > 1. What features an be used to know the efforts that takes to resolve a > > bug. > > Turn on Time Tracking. You will need 2.17.3 for this. > > > & > > 2. Suppose in a buglife cycle it is identified that the bug is not in > > component1 and component2 , is there any feature in Bugzilla can be > > utilised for the defect leakage per component basis as in the summary > > reports we don't have these details. > > You could construct a (fairly complicated) search against the Bugs > Activity table to tell you how many bugs had been moved from one > component to another over a certain amount of time, yes. But there's no > pretty report which tells you. > > Gerv > > ---- > To view or change your list settings, click here: > > From geeta.tripathi at bhartitelesoft.com Tue Mar 4 07:19:13 2003 From: geeta.tripathi at bhartitelesoft.com (geeta) Date: Tue, 4 Mar 2003 12:49:13 +0530 Subject: bug life cycle in Bugzilla References: <002301c2d745$6587a960$800310ac@tripathi.bhartitelesoft.com> <20030227040954.B23574@sigkill.com> <000d01c2d7fa$7bb2b3a0$800310ac@tripathi.bhartitelesoft.com> <3E61FA97.3010301@mozilla.org> <001701c2e146$d3bc79a0$800310ac@tripathi.bhartitelesoft.com> <3E631930.6000100@mozilla.org> Message-ID: <002501c2e21e$60248ea0$800310ac@tripathi.bhartitelesoft.com> Hello Gervase, Is 2.17.3 stable. Can we go ahead with its implmentation in our organisation. Regards, Geeta ----- Original Message ----- From: "Gervase Markham" To: Sent: Monday, March 03, 2003 2:28 PM Subject: Re: bug life cycle in Bugzilla > > 1. What features an be used to know the efforts that takes to resolve a > > bug. > > Turn on Time Tracking. You will need 2.17.3 for this. > > > & > > 2. Suppose in a buglife cycle it is identified that the bug is not in > > component1 and component2 , is there any feature in Bugzilla can be > > utilised for the defect leakage per component basis as in the summary > > reports we don't have these details. > > You could construct a (fairly complicated) search against the Bugs > Activity table to tell you how many bugs had been moved from one > component to another over a certain amount of time, yes. But there's no > pretty report which tells you. > > Gerv > > ---- > To view or change your list settings, click here: > > From gerv at mozilla.org Tue Mar 4 07:30:38 2003 From: gerv at mozilla.org (Gervase Markham) Date: Tue, 04 Mar 2003 07:30:38 +0000 Subject: bug life cycle in Bugzilla In-Reply-To: <002501c2e21e$60248ea0$800310ac@tripathi.bhartitelesoft.com> References: <002301c2d745$6587a960$800310ac@tripathi.bhartitelesoft.com> <20030227040954.B23574@sigkill.com> <000d01c2d7fa$7bb2b3a0$800310ac@tripathi.bhartitelesoft.com> <3E61FA97.3010301@mozilla.org> <001701c2e146$d3bc79a0$800310ac@tripathi.bhartitelesoft.com> <3E631930.6000100@mozilla.org> <002501c2e21e$60248ea0$800310ac@tripathi.bhartitelesoft.com> Message-ID: <3E64561E.3030207@mozilla.org> geeta wrote: > Hello Gervase, > > Is 2.17.3 stable. Can we go ahead with its implmentation in our > organisation. Well, I can't answer that question for you! :-) Gerv From jon at vmware.com Wed Mar 5 01:01:00 2003 From: jon at vmware.com (Jonathan Schatz) Date: 04 Mar 2003 17:01:00 -0800 Subject: problem with reports.cgi Message-ID: <1046826060.32243.23.camel@jonschatz-lx.vmware.com> when i try to use reports.cgi, i get a broken image link. opening the broken image link in a new window, i get: Can't locate Bugzilla/Template/Plugin/GD/Graph/bars.pm in @INC (@INC contains: . /usr/lib/perl5/5.6.1/i386-linux /usr/lib/perl5/5.6.1 /usr/lib/perl5/site_perl/5.6.1/i386-linux /usr/lib/perl5/site_perl/5.6.1 /usr/lib/perl5/site_perl/5.6.0 /usr/lib/perl5/site_perl /usr/lib/perl5/vendor_perl/5.6.1/i386-linux /usr/lib/perl5/vendor_perl/5.6.1 /usr/lib/perl5/vendor_perl) at /usr/lib/perl5/site_perl/5.6.1/i386-linux/Template/Plugins.pm line 194. where does the problem lie here? i found the GD::Graph installation location on my machine and symlinked it into my bugzilla tree. i then got a new error: Can't locate object method "load" via package "Bugzilla::Template::Plugin::GD::Graph::bars" (perhaps you forgot to load "Bugzilla::Template::Plugin::GD::Graph::bars"?) at /usr/lib/perl5/site_perl/5.6.1/i386-linux/Template/Plugins.pm line 206. there doesn't seem to be a method named load in any of these packages, and even if there was, it's not in the Bugzilla::Template::Plugin namespace anyway. any ideas? -jon -- Jonathan Schatz Engineering System Administrator VMware, Inc "Te occidere possunt sed te edere non possunt nefas est." From bbaetz at acm.org Wed Mar 5 07:26:30 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Wed, 5 Mar 2003 18:26:30 +1100 Subject: problem with reports.cgi In-Reply-To: <1046826060.32243.23.camel@jonschatz-lx.vmware.com> References: <1046826060.32243.23.camel@jonschatz-lx.vmware.com> Message-ID: <20030305072630.GA1348@mango.home> On Tue, Mar 04, 2003 at 05:01:00PM -0800, Jonathan Schatz wrote: > when i try to use reports.cgi, i get a broken image link. > opening the broken image link in a new window, i get: > > where does the problem lie here? i found the GD::Graph installation location > on my machine and symlinked it into my bugzilla tree. i then got a new error: GD::Graph should be installed separately, not into the Bugzilla tree. The template plugin should be a separate thing - what TT version do you have? And are you sure that its installed correctly? Bradley From geeta.tripathi at bhartitelesoft.com Wed Mar 5 07:22:16 2003 From: geeta.tripathi at bhartitelesoft.com (geeta) Date: Wed, 5 Mar 2003 12:52:16 +0530 Subject: bug life cycle in Bugzilla References: <002301c2d745$6587a960$800310ac@tripathi.bhartitelesoft.com> <20030227040954.B23574@sigkill.com> <000d01c2d7fa$7bb2b3a0$800310ac@tripathi.bhartitelesoft.com> <3E61FA97.3010301@mozilla.org> <001701c2e146$d3bc79a0$800310ac@tripathi.bhartitelesoft.com> <3E631930.6000100@mozilla.org> <002501c2e21e$60248ea0$800310ac@tripathi.bhartitelesoft.com> <3E64561E.3030207@mozilla.org> Message-ID: <000901c2e2e7$f7c4fc80$800310ac@tripathi.bhartitelesoft.com> Hello Gervase, I was playing around with the bugzilla version 2.17.3 on the landfill web-site with the time-tracker. I tried to change a state today but then the date remains for yesterday. Is this a bug. Or any parameter needs to be enabled now I don't have tweakconfig access. Looking forward to hear from you. Thanks and regards, Geeta ----- Original Message ----- From: "Gervase Markham" To: Sent: Tuesday, March 04, 2003 1:00 PM Subject: Re: bug life cycle in Bugzilla > geeta wrote: > > Hello Gervase, > > > > Is 2.17.3 stable. Can we go ahead with its implmentation in our > > organisation. > > Well, I can't answer that question for you! :-) > > Gerv > > ---- > To view or change your list settings, click here: > > From cmason at somanetworks.com Wed Mar 5 16:27:40 2003 From: cmason at somanetworks.com (Chris L. Mason) Date: Wed, 5 Mar 2003 11:27:40 -0500 Subject: Support for multiple "Locations" in bugzilla In-Reply-To: <20030228213030.GN29540@somanetworks.com> References: <20030228160833.GH16925@somanetworks.com> <20030228122106.G29082@sigkill.com> <20030228204414.GJ29540@somanetworks.com> <3E5FCC97.1000700@mozilla.org> <20030228210919.GK29540@somanetworks.com> <3E5FD267.8040306@mozilla.org> <20030228213030.GN29540@somanetworks.com> Message-ID: <20030305162740.GB26617@somanetworks.com> Okay, after reading the FAQ on custom fields, we're just going to use bug entry templates for the location. :) Thanks, Chris -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 232 bytes Desc: not available URL: From gerv at mozilla.org Wed Mar 5 17:59:05 2003 From: gerv at mozilla.org (Gervase Markham) Date: Wed, 05 Mar 2003 17:59:05 +0000 Subject: bug life cycle in Bugzilla In-Reply-To: <000901c2e2e7$f7c4fc80$800310ac@tripathi.bhartitelesoft.com> References: <002301c2d745$6587a960$800310ac@tripathi.bhartitelesoft.com> <20030227040954.B23574@sigkill.com> <000d01c2d7fa$7bb2b3a0$800310ac@tripathi.bhartitelesoft.com> <3E61FA97.3010301@mozilla.org> <001701c2e146$d3bc79a0$800310ac@tripathi.bhartitelesoft.com> <3E631930.6000100@mozilla.org> <002501c2e21e$60248ea0$800310ac@tripathi.bhartitelesoft.com> <3E64561E.3030207@mozilla.org> <000901c2e2e7$f7c4fc80$800310ac@tripathi.bhartitelesoft.com> Message-ID: <3E663AE9.8070602@mozilla.org> geeta wrote: > Hello Gervase, > I tried to change a state today but then the date remains for yesterday. I haven't used the time tracker, so I'm not very familiar with it, and I don't really understand what you are referring to. As I've said before, the newsgroup is the best place for questions - hopefully someone there can help you out. :-) Gerv From gerv at mozilla.org Wed Mar 5 17:59:52 2003 From: gerv at mozilla.org (Gervase Markham) Date: Wed, 05 Mar 2003 17:59:52 +0000 Subject: bug life cycle in Bugzilla In-Reply-To: <000901c2e2e7$f7c4fc80$800310ac@tripathi.bhartitelesoft.com> References: <002301c2d745$6587a960$800310ac@tripathi.bhartitelesoft.com> <20030227040954.B23574@sigkill.com> <000d01c2d7fa$7bb2b3a0$800310ac@tripathi.bhartitelesoft.com> <3E61FA97.3010301@mozilla.org> <001701c2e146$d3bc79a0$800310ac@tripathi.bhartitelesoft.com> <3E631930.6000100@mozilla.org> <002501c2e21e$60248ea0$800310ac@tripathi.bhartitelesoft.com> <3E64561E.3030207@mozilla.org> <000901c2e2e7$f7c4fc80$800310ac@tripathi.bhartitelesoft.com> Message-ID: <3E663B18.4060401@mozilla.org> geeta wrote: > Hello Gervase, > > I was playing around with the bugzilla version 2.17.3 on the landfill > web-site with the time-tracker. Oops. Sorry everyone. Thought this was personal mail, given that it had my name at the top. Still, Geeta, the newsgroup might be a good bet if no-one here can answer your question. Gerv From geeta.tripathi at bhartitelesoft.com Thu Mar 6 04:40:53 2003 From: geeta.tripathi at bhartitelesoft.com (geeta) Date: Thu, 6 Mar 2003 10:10:53 +0530 Subject: bug life cycle in Bugzilla References: <002301c2d745$6587a960$800310ac@tripathi.bhartitelesoft.com> <20030227040954.B23574@sigkill.com> <000d01c2d7fa$7bb2b3a0$800310ac@tripathi.bhartitelesoft.com> <3E61FA97.3010301@mozilla.org> <001701c2e146$d3bc79a0$800310ac@tripathi.bhartitelesoft.com> <3E631930.6000100@mozilla.org> <002501c2e21e$60248ea0$800310ac@tripathi.bhartitelesoft.com> <3E64561E.3030207@mozilla.org> <000901c2e2e7$f7c4fc80$800310ac@tripathi.bhartitelesoft.com> <3E663B18.4060401@mozilla.org> Message-ID: <001a01c2e39a$95baa0e0$800310ac@tripathi.bhartitelesoft.com> ok ----- Original Message ----- From: "Gervase Markham" To: Sent: Wednesday, March 05, 2003 11:29 PM Subject: Re: bug life cycle in Bugzilla > geeta wrote: > > Hello Gervase, > > > > I was playing around with the bugzilla version 2.17.3 on the landfill > > web-site with the time-tracker. > > Oops. Sorry everyone. Thought this was personal mail, given that it had > my name at the top. Still, Geeta, the newsgroup might be a good bet if > no-one here can answer your question. > > Gerv > > ---- > To view or change your list settings, click here: > > From jon at vmware.com Thu Mar 6 19:40:57 2003 From: jon at vmware.com (Jonathan Schatz) Date: 06 Mar 2003 11:40:57 -0800 Subject: problem with reports.cgi In-Reply-To: <20030305072630.GA1348@mango.home> References: <1046826060.32243.23.camel@jonschatz-lx.vmware.com> <20030305072630.GA1348@mango.home> Message-ID: <1046979657.17027.14.camel@jonschatz-lx.vmware.com> On Tue, 2003-03-04 at 23:26, Bradley Baetz wrote: > GD::Graph should be installed separately, not into the Bugzilla tree. > The template plugin should be a separate thing - what TT version do you > have? And are you sure that its installed correctly? [root at bugzilla jon]$ perl -MGD::Graph -e 'print $GD::Graph::VERSION, "\n"' 1.39 [root at bugzilla jon]$ perl -MTemplate::Plugin::GD::Graph::bars -e 'print "$Template::Plugin::GD::Graph::bars::VERSION\n"' 1.44 [root at bugzilla jon]$ perl -MTemplate::Plugin -e 'print "$Template::Plugin::VERSION\n"' 2.54 i reinstalled all of these modules just to make sure, but it didn't change anything. all of the tests passed though. i'm just curious as to why the call to Template::Plugin::GD::Graph::bars is made as Bugzilla::Template::Plugin::GD::Graph::bars (this is in Template::Plugins::_load). is this code generally considered to work? any other ideas? -jon -- Jonathan Schatz Engineering System Administrator VMware, Inc "Te occidere possunt sed te edere non possunt nefas est." From justdave at syndicomm.com Wed Mar 12 00:18:09 2003 From: justdave at syndicomm.com (David Miller) Date: Tue, 11 Mar 2003 19:18:09 -0500 Subject: Bugzilla #! perl location changing Message-ID: I wanted to issue a heads-up to everyone that we're going to be changing the default perl path on all Bugzilla files from /usr/bonsaitools/bin/perl to /usr/bin/perl sometime in the next few days. This has been a much sought-after change for a long time, and the main roadblock to doing so before have been objections from mozilla.org. Those objections have been alleviated by other means now, so we can change it without them giving us a hassle over it now. :) See http://bugzilla.mozilla.org/show_bug.cgi?id=196433 for details. If you're using a CVS copy of Bugzilla and actually have more than one perl on your machine, and your bonsaitools symlink happens to point at something other than /usr/bin/perl, then this will probably affect you. I suspect that's a very small percentage of folks though. :) -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From geeta.tripathi at bhartitelesoft.com Wed Mar 12 07:24:07 2003 From: geeta.tripathi at bhartitelesoft.com (geeta) Date: Wed, 12 Mar 2003 12:54:07 +0530 Subject: Bugzilla 2.16.1 References: Message-ID: <002001c2e868$62988240$800310ac@tripathi.bhartitelesoft.com> Hello David, I am again installing bugzilla seperately organisation wide, while accessing bugzilla I got the following errors- Directory index forbidden by rule: /home/bugzilla/apache/htdocs/ and on running index.cgi on command line I get the following error- Uncaught exception from user code: Uncaught exception from user code: Uncaught exception from user code: mkdir data/template/en: Permission denied at /usr/lib/perl5/site_perl/5.005/i386-linux/T emplate/Provider.pm line 368 Carp::croak('mkdir data/template/en: Permission denied') called at /usr/lib/perl5/5.0050 3/File/Path.pm line 140 File::Path::mkpath('data/template/en', undef, 511) called at /usr/lib/perl5/5.00503/File /Path.pm line 133 File::Path::mkpath('data/template/en/custom') called at /usr/lib/perl5/site_perl/5.005/i 386-linux/Template/Provider.pm line 368 Template::Provider::_init('Template::Provider=HASH(0x84b2244)', 'HASH(0x8356488)') calle d at /usr/lib/perl5/site_perl/5.005/i386-linux/Template/Base.pm line 71 Template::Base::new('Template::Provider', 'HASH(0x8356488)') called at /usr/lib/perl5/si te_perl/5.005/i386-linux/Template/Config.pm line 116 Template::Config::provider('Template::Config', 'HASH(0x8356488)') called at /usr/lib/per l5/site_perl/5.005/i386-linux/Template/Context.pm line 725 Template::Context::_init('Template::Context=HASH(0x84b2070)', 'HASH(0x8356488)') called at /usr/lib/perl5/site_perl/5.005/i386-linux/Template/Base.pm line 71 Template::Base::new('Template::Context', 'HASH(0x8356488)') called at /usr/lib/perl5/sit e_perl/5.005/i386-linux/Template/Config.pm line 208 Template::Config::context('Template::Config', 'HASH(0x8356488)') called at /usr/lib/perl 5/site_perl/5.005/i386-linux/Template/Service.pm line 159 Template::Service::_init('Template::Service=HASH(0x835aae8)', 'HASH(0x8356488)') called at /usr/lib/perl5/site_perl/5.005/i386-linux/Template/Base.pm line 71 Template::Base::new('Template::Service', 'HASH(0x8356488)') called at /usr/lib/perl5/sit e_perl/5.005/i386-linux/Template/Config.pm line 226 Template::Config::service('Template::Config', 'HASH(0x8356488)') called at /usr/lib/perl 5/site_perl/5.005/i386-linux/Template.pm line 126 Template::_init('Template=HASH(0x8357398)', 'HASH(0x8356488)') called at /usr/lib/perl5/ site_perl/5.005/i386-linux/Template/Base.pm line 71 Template::Base::new('Template', 'HASH(0x8356488)') called at globals.pl line 1593 require globals.pl called at CGI.pl line 53 require CGI.pl called at index.cgi line 39 require CGI.pl called at index.cgi line 39 I have installed Bugzilla earlier, but I have never got this error. Is there any permission issue, this is all I can understand from the error.............how do I resolve it. Please let me know of any further details needed from my end to resolve this issue. Looking forward to hear from you. regards, Geeta -------------- next part -------------- An HTML attachment was scrubbed... URL: From geeta.tripathi at bhartitelesoft.com Wed Mar 12 10:02:41 2003 From: geeta.tripathi at bhartitelesoft.com (geeta) Date: Wed, 12 Mar 2003 15:32:41 +0530 Subject: Bugzilla 2.16.1 References: <002001c2e868$62988240$800310ac@tripathi.bhartitelesoft.com> Message-ID: <007e01c2e87e$890191e0$800310ac@tripathi.bhartitelesoft.com> Please help, I am still waiting. regards, Geeta ----- Original Message ----- From: geeta To: developers at bugzilla.org Sent: Wednesday, March 12, 2003 12:54 PM Subject: Re: Bugzilla 2.16.1 Hello David, I am again installing bugzilla seperately organisation wide, while accessing bugzilla I got the following errors- Directory index forbidden by rule: /home/bugzilla/apache/htdocs/ and on running index.cgi on command line I get the following error- Uncaught exception from user code: Uncaught exception from user code: Uncaught exception from user code: mkdir data/template/en: Permission denied at /usr/lib/perl5/site_perl/5.005/i386-linux/T emplate/Provider.pm line 368- Check for File-Spec Carp::croak('mkdir data/template/en: Permission denied') called at /usr/lib/perl5/5.0050 3/File/Path.pm line 140 File::Path::mkpath('data/template/en', undef, 511) called at /usr/lib/perl5/5.00503/File /Path.pm line 133 File::Path::mkpath('data/template/en/custom') called at /usr/lib/perl5/site_perl/5.005/i 386-linux/Template/Provider.pm line 368-Check for File-Spec Template::Provider::_init('Template::Provider=HASH(0x84b2244)', 'HASH(0x8356488)') calle d at /usr/lib/perl5/site_perl/5.005/i386-linux/Template/Base.pm line 71-Base for Configurations Template::Base::new('Template::Provider', 'HASH(0x8356488)') called at /usr/lib/perl5/si te_perl/5.005/i386-linux/Template/Config.pm line 116 Template::Config::provider('Template::Config', 'HASH(0x8356488)') called at /usr/lib/per l5/site_perl/5.005/i386-linux/Template/Context.pm line 725-Template Template::Context::_init('Template::Context=HASH(0x84b2070)', 'HASH(0x8356488)') called at /usr/lib/perl5/site_perl/5.005/i386-linux/Template/Base.pm line 71 Template::Base::new('Template::Context', 'HASH(0x8356488)') called at /usr/lib/perl5/sit e_perl/5.005/i386-linux/Template/Config.pm line 208 Template::Config::context('Template::Config', 'HASH(0x8356488)') called at /usr/lib/perl 5/site_perl/5.005/i386-linux/Template/Service.pm line 159 Template::Service::_init('Template::Service=HASH(0x835aae8)', 'HASH(0x8356488)') called at /usr/lib/perl5/site_perl/5.005/i386-linux/Template/Base.pm line 71 Template::Base::new('Template::Service', 'HASH(0x8356488)') called at /usr/lib/perl5/sit e_perl/5.005/i386-linux/Template/Config.pm line 226 Template::Config::service('Template::Config', 'HASH(0x8356488)') called at /usr/lib/perl 5/site_perl/5.005/i386-linux/Template.pm line 126 Template::_init('Template=HASH(0x8357398)', 'HASH(0x8356488)') called at /usr/lib/perl5/ site_perl/5.005/i386-linux/Template/Base.pm line 71 Template::Base::new('Template', 'HASH(0x8356488)') called at globals.pl line 1593 require globals.pl called at CGI.pl line 53 require CGI.pl called at index.cgi line 39 require CGI.pl called at index.cgi line 39 I have installed Bugzilla earlier, but I have never got this error. Is there any permission issue, this is all I can understand from the error.............how do I resolve it. Please let me know of any further details needed from my end to resolve this issue. Looking forward to hear from you. regards, Geeta -------------- next part -------------- An HTML attachment was scrubbed... URL: From caseyg at chsamerica.com Wed Mar 12 14:49:09 2003 From: caseyg at chsamerica.com (Casey Gregoire) Date: Wed, 12 Mar 2003 09:49:09 -0500 Subject: Time Tracking Message-ID: Does anyone know of a patch that does time tacking in the sense of you can start a timer and stop it to track exactly how many hours you worked on a bug? Not just tracking from time of reporting to time of fixing, but tracking how much time an individual spends working on a bug? Our tracking system here has a feature for doing so. This is the ONLY good feature it has. But we would like to have it when we go to Bugzilla and need to know if we should write it in or not. Thank you, Casey Gregoire Programmer CHS of America 100 1st Ave S. Suite 601 St. Petersburg, FL 33701 Phone - (727) 824-0800 ext 1236 Every great achievement was once impossible. -- Anonymous I'd like to get started on time, if we can, inasmuch as we're late already. -- Larry Gelbart From gerv at mozilla.org Wed Mar 12 15:04:19 2003 From: gerv at mozilla.org (Gervase Markham) Date: Wed, 12 Mar 2003 15:04:19 +0000 Subject: Time Tracking In-Reply-To: References: Message-ID: <3E6F4C73.4020602@mozilla.org> Casey Gregoire wrote: > Does anyone know of a patch that does time tacking in the sense of you can > start a timer and stop it to track exactly how many hours you worked on a > bug? This question would have been better posed in the newsgroup. Have you investigated Bugzilla's time tracking features? What do they not have that you need? Gerv From PPrymmer at factset.com Wed Mar 12 15:17:05 2003 From: PPrymmer at factset.com (PPrymmer at factset.com) Date: Wed, 12 Mar 2003 10:17:05 -0500 Subject: Bugzilla #! perl location changing Message-ID: Could someone explain why $Config{startperl} extraction is not used for the tops of the files? That is why have any hard coded, possibly incorrect path? Note that h2ph and h2xs in the perl distrbution are extract from h2ph.PL and h2xs.PL respectively. It is my understanding that bugzilla will be moving toward a Makefile.PL based installation. Within a Makefile.PL it is even easier to do .PL extraction provided the hash passed to WriteMakefile() has an entry 'EXE_FILES' that points to an array ref. To make the example explicit you could have this set up (assuming that the index.PL file will become index.cgi and it is in the same directory as the Makefile.PL): WriteMakefile( # other stuff 'EXE_FILES' => [("index.cgi")], # even more hash elements ); Then the file 'index.PL' that ships with the tar ball can be written with a simple wrapper like so: use Config; use File::Basename qw(basename dirname); use Cwd; my $origdir = cwd; chdir dirname($0); my $file = basename($0, '.PL'); $file .= '.cgi'; open OUT,">$file" or die "Can't create $file: $!"; print OUT <<"!GROK!THIS!"; $Config{startperl} !GROK!THIS! print OUT <<'!NO!SUBS!'; # original content of index.cgi excluding top #! line !NO!SUBS! That avoids a hard coded #!/path altogether and takes the one specified in Config.pm for the perl that is used to run the initial: perl Makefile.PL installation step. Peter Prymmer David Miller cc: Sent by: developers-owner at b Subject: Bugzilla #! perl location changing ugzilla.org 03/11/2003 07:18 PM Please respond to developers I wanted to issue a heads-up to everyone that we're going to be changing the default perl path on all Bugzilla files from /usr/bonsaitools/bin/perl to /usr/bin/perl sometime in the next few days. This has been a much sought-after change for a long time, and the main roadblock to doing so before have been objections from mozilla.org. Those objections have been alleviated by other means now, so we can change it without them giving us a hassle over it now. :) See http://bugzilla.mozilla.org/show_bug.cgi?id=196433 for details. If you're using a CVS copy of Bugzilla and actually have more than one perl on your machine, and your bonsaitools symlink happens to point at something other than /usr/bin/perl, then this will probably affect you. I suspect that's a very small percentage of folks though. :) -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ ---- To view or change your list settings, click here: From caseyg at chsamerica.com Wed Mar 12 16:03:22 2003 From: caseyg at chsamerica.com (Casey Gregoire) Date: Wed, 12 Mar 2003 11:03:22 -0500 Subject: Time Tracking Message-ID: Well today I made a stupid mistake. And don't know how to reverse it. When I tried to setup Bugzilla 2.17.3 on my machine I ran the checksetup.pl script on my 2.16 Bugzilla database and now 2.16 does not work with my database. Apparently some fields are different in the database in the 2.17 version. How do I get it back to the way it was? Can I? Thanks, Casey Gregoire -----Original Message----- From: Gervase Markham [mailto:gerv at mozilla.org] Sent: Wednesday, March 12, 2003 10:04 AM To: developers at bugzilla.org Subject: Re: Time Tracking Casey Gregoire wrote: > Does anyone know of a patch that does time tacking in the sense of you can > start a timer and stop it to track exactly how many hours you worked on a > bug? This question would have been better posed in the newsgroup. Have you investigated Bugzilla's time tracking features? What do they not have that you need? Gerv ---- To view or change your list settings, click here: From justdave at syndicomm.com Wed Mar 12 16:38:36 2003 From: justdave at syndicomm.com (David Miller) Date: Wed, 12 Mar 2003 11:38:36 -0500 Subject: Bugzilla #! perl location changing In-Reply-To: References: Message-ID: On 3/12/03 10:17 AM -0500, PPrymmer at factset.com wrote: > Could someone explain why $Config{startperl} extraction > is not used for the tops of the files? That is why have any > hard coded, possibly incorrect path? Note that h2ph and > h2xs in the perl distrbution are extract from h2ph.PL and > h2xs.PL respectively. It is my understanding that bugzilla > will be moving toward a Makefile.PL based installation. > Within a Makefile.PL it is even easier to do .PL extraction > provided the hash passed to WriteMakefile() has an > entry 'EXE_FILES' that points to an array ref. To make the > example explicit you could have this set up (assuming > that the index.PL file will become index.cgi and it is in > the same directory as the Makefile.PL): You quoted a message I wrote: > See http://bugzilla.mozilla.org/show_bug.cgi?id=196433 for details. You'll notice on that bug that the long term goal is indeed intended to rewrite the #! lines on the files as they get installed with the path to the perl the installer was called with. This is a short-term thing to make it easier for people until that installer is ready. -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From caseyg at chsamerica.com Wed Mar 12 17:17:32 2003 From: caseyg at chsamerica.com (Casey Gregoire) Date: Wed, 12 Mar 2003 12:17:32 -0500 Subject: Time Tracking Message-ID: I have looked at the time tracking features now. What I was looking for is you start a timer, then when you are done working on that bug you go back to the bug and hit stop timer. And that time gets added in. (a comment would be added as well.) Thanks, Casey Gregoire -----Original Message----- From: Gervase Markham [mailto:gerv at mozilla.org] Sent: Wednesday, March 12, 2003 10:04 AM To: developers at bugzilla.org Subject: Re: Time Tracking Casey Gregoire wrote: > Does anyone know of a patch that does time tacking in the sense of you can > start a timer and stop it to track exactly how many hours you worked on a > bug? This question would have been better posed in the newsgroup. Have you investigated Bugzilla's time tracking features? What do they not have that you need? Gerv ---- To view or change your list settings, click here: From jake at bugzilla.org Wed Mar 12 17:22:37 2003 From: jake at bugzilla.org (Jake) Date: Wed, 12 Mar 2003 12:22:37 -0500 Subject: Time Tracking In-Reply-To: References: Message-ID: <3E6F6CDD.8040400@bugzilla.org> Ouch. We generally don't provide a script to revert the schema as it's not the direction we encourage :) About the only way to go back is either restore from backup or look at the changes in checksetup.pl from 2.16 to 2.17.3 and write a script that does the opposite. Casey Gregoire wrote: >Well today I made a stupid mistake. And don't know how to reverse it. When I >tried to setup Bugzilla 2.17.3 on my machine I ran the checksetup.pl script >on my 2.16 Bugzilla database and now 2.16 does not work with my database. >Apparently some fields are different in the database in the 2.17 version. >How do I get it back to the way it was? Can I? > > > From jake at bugzilla.org Wed Mar 12 17:26:08 2003 From: jake at bugzilla.org (Jake) Date: Wed, 12 Mar 2003 12:26:08 -0500 Subject: Bugzilla 2.16.1 In-Reply-To: <007e01c2e87e$890191e0$800310ac@tripathi.bhartitelesoft.com> References: <002001c2e868$62988240$800310ac@tripathi.bhartitelesoft.com> <007e01c2e87e$890191e0$800310ac@tripathi.bhartitelesoft.com> Message-ID: <3E6F6DB0.7040101@bugzilla.org> Did you run checksetup.pl and did it give you any errors (such as missing modules or permissions issues)? The newsgroup would probably be a more appropriate place for this as it has a larger audiance of potentially helpful people. geeta wrote: > Please help, I am still waiting. > > regards, > > Geeta > > ----- Original Message ----- > *From:* geeta > *To:* developers at bugzilla.org > *Sent:* Wednesday, March 12, 2003 12:54 PM > *Subject:* Re: Bugzilla 2.16.1 > > Hello David, > > I am again installing bugzilla seperately organisation wide, while > accessing bugzilla I got the following errors- > > Directory index forbidden by rule: /home/bugzilla/apache/htdocs/ > > and on running index.cgi on command line I get the following error- > > *Uncaught exception from user code: > Uncaught exception from user code: > Uncaught exception from user code: > mkdir data/template/en: Permission denied at > /usr/lib/perl5/site_perl/5.005/i386-linux/T > emplate/Provider.pm line 368- Check for File-Spec > Carp::croak('mkdir data/template/en: Permission denied') > called at /usr/lib/perl5/5.0050 > 3/File/Path.pm line 140 > File::Path::mkpath('data/template/en', undef, 511) called > at /usr/lib/perl5/5.00503/File > /Path.pm line 133 > File::Path::mkpath('data/template/en/custom') called at > /usr/lib/perl5/site_perl/5.005/i > 386-linux/Template/Provider.pm line 368-Check for File-Spec > > Template::Provider::_init('Template::Provider=HASH(0x84b2244)', > 'HASH(0x8356488)') calle > d at /usr/lib/perl5/site_perl/5.005/i386-linux/Template/Base.pm > line 71-Base for Configurations > Template::Base::new('Template::Provider', > 'HASH(0x8356488)') called at /usr/lib/perl5/si > te_perl/5.005/i386-linux/Template/Config.pm line 116 > Template::Config::provider('Template::Config', > 'HASH(0x8356488)') called at /usr/lib/per > l5/site_perl/5.005/i386-linux/Template/Context.pm line 725-Template > > Template::Context::_init('Template::Context=HASH(0x84b2070)', > 'HASH(0x8356488)') called > at /usr/lib/perl5/site_perl/5.005/i386-linux/Template/Base.pm line 71 > Template::Base::new('Template::Context', > 'HASH(0x8356488)') called at /usr/lib/perl5/sit > e_perl/5.005/i386-linux/Template/Config.pm line 208 > Template::Config::context('Template::Config', > 'HASH(0x8356488)') called at /usr/lib/perl > 5/site_perl/5.005/i386-linux/Template/Service.pm line 159 > > Template::Service::_init('Template::Service=HASH(0x835aae8)', > 'HASH(0x8356488)') called > at /usr/lib/perl5/site_perl/5.005/i386-linux/Template/Base.pm line 71 > Template::Base::new('Template::Service', > 'HASH(0x8356488)') called at /usr/lib/perl5/sit > e_perl/5.005/i386-linux/Template/Config.pm line 226 > Template::Config::service('Template::Config', > 'HASH(0x8356488)') called at /usr/lib/perl > 5/site_perl/5.005/i386-linux/Template.pm line 126 > Template::_init('Template=HASH(0x8357398)', > 'HASH(0x8356488)') called at /usr/lib/perl5/ > site_perl/5.005/i386-linux/Template/Base.pm line 71 > Template::Base::new('Template', 'HASH(0x8356488)') called > at globals.pl line 1593 > require globals.pl called at CGI.pl line 53 > require CGI.pl called at index.cgi line 39 > require CGI.pl called at index.cgi line 39* > > I have installed Bugzilla earlier, but I have never got this > error. Is there any permission issue, this is all I can understand > from the error.............how do I resolve it. > > Please let me know of any further details needed from my end to > resolve this issue. > > Looking forward to hear from you. > > regards, > > Geeta > From geeta.tripathi at bhartitelesoft.com Wed Mar 12 17:30:28 2003 From: geeta.tripathi at bhartitelesoft.com (geeta.tripathi at bhartitelesoft.com) Date: Wed, 12 Mar 2003 17:30:28 GMT Subject: Bugzilla 2.16.1 Message-ID: <200303121730.h2CHURT16500@mail.bhartitelesoft.com> Hello Jake, I have not missed installing any Bugzilla module. However, when I run checksetup.pl from the webuser id it talks of the DBD:Mysql missing. But when I run the checksetup.pl from root, I get no error messages. Is there any catch there....... I can't see the resolution to this issue in the newsgroup any other Mozilla site has the solution. Thanks and regards, Geeta Did you run checksetup.pl and did it give you any errors (such as > missing modules or permissions issues)? > > The newsgroup would probably be a more appropriate place for this as it > has a larger audiance of potentially helpful people. > > geeta wrote: > > > Please help, I am still waiting. > > > > regards, > > > > Geeta > > > > ----- Original Message ----- > > *From:* geeta > > *To:* developers at bugzilla.org > > *Sent:* Wednesday, March 12, 2003 12:54 PM > > *Subject:* Re: Bugzilla 2.16.1 > > > > Hello David, > > > > I am again installing bugzilla seperately organisation wide, while > > accessing bugzilla I got the following errors- > > > > Directory index forbidden by rule: /home/bugzilla/apache/htdocs/ > > > > and on running index.cgi on command line I get the following error- > > > > *Uncaught exception from user code: > > Uncaught exception from user code: > > Uncaught exception from user code: > > mkdir data/template/en: Permission denied at > > /usr/lib/perl5/site_perl/5.005/i386-linux/T > > emplate/Provider.pm line 368- Check for File-Spec > > Carp::croak('mkdir data/template/en: Permission denied') > > called at /usr/lib/perl5/5.0050 > > 3/File/Path.pm line 140 > > File::Path::mkpath('data/template/en', undef, 511) called > > at /usr/lib/perl5/5.00503/File > > /Path.pm line 133 > > File::Path::mkpath ('data/template/en/custom') called at > > /usr/lib/perl5/site_perl/5.005/i > > 386-linux/Template/Provider.pm line 368-Check for File-Spec > > > > Template::Provider::_init('Template::Provider=HASH (0x84b2244)', > > 'HASH(0x8356488)') calle > > d at /usr/lib/perl5/site_perl/5.005/i386- linux/Template/Base.pm > > line 71-Base for Configurations > > Template::Base::new('Template::Provider', > > 'HASH(0x8356488)') called at /usr/lib/perl5/si > > te_perl/5.005/i386-linux/Template/Config.pm line 116 > > Template::Config::provider ('Template::Config', > > 'HASH(0x8356488)') called at /usr/lib/per > > l5/site_perl/5.005/i386-linux/Template/Context.pm line 725-Template > > > > Template::Context::_init('Template::Context=HASH (0x84b2070)', > > 'HASH(0x8356488)') called > > at /usr/lib/perl5/site_perl/5.005/i386- linux/Template/Base.pm line 71 > > Template::Base::new('Template::Context', > > 'HASH(0x8356488)') called at /usr/lib/perl5/sit > > e_perl/5.005/i386-linux/Template/Config.pm line 208 > > Template::Config::context ('Template::Config', > > 'HASH(0x8356488)') called at /usr/lib/perl > > 5/site_perl/5.005/i386-linux/Template/Service.pm line 159 > > > > Template::Service::_init('Template::Service=HASH (0x835aae8)', > > 'HASH(0x8356488)') called > > at /usr/lib/perl5/site_perl/5.005/i386- linux/Template/Base.pm line 71 > > Template::Base::new('Template::Service', > > 'HASH(0x8356488)') called at /usr/lib/perl5/sit > > e_perl/5.005/i386-linux/Template/Config.pm line 226 > > Template::Config::service ('Template::Config', > > 'HASH(0x8356488)') called at /usr/lib/perl > > 5/site_perl/5.005/i386-linux/Template.pm line 126 > > Template::_init('Template=HASH(0x8357398)', > > 'HASH(0x8356488)') called at /usr/lib/perl5/ > > site_perl/5.005/i386-linux/Template/Base.pm line 71 > > Template::Base::new('Template', 'HASH (0x8356488)') called > > at globals.pl line 1593 > > require globals.pl called at CGI.pl line 53 > > require CGI.pl called at index.cgi line 39 > > require CGI.pl called at index.cgi line 39* > > > > I have installed Bugzilla earlier, but I have never got this > > error. Is there any permission issue, this is all I can understand > > from the error.............how do I resolve it. > > > > Please let me know of any further details needed from my end to > > resolve this issue. > > > > Looking forward to hear from you. > > > > regards, > > > > Geeta > > > > > ---- > To view or change your list settings, click here: > > ---------------------------------------------------------- http://www.bhartitelesoft.com Telecommunication and E-Commerce solutions for the future. From caseyg at chsamerica.com Wed Mar 12 19:02:23 2003 From: caseyg at chsamerica.com (Casey Gregoire) Date: Wed, 12 Mar 2003 14:02:23 -0500 Subject: Bugzilla 2.16.1 Message-ID: Well, since I don't have a back up - my fault.... And Bugzilla 2.17 seems to work fine right now, I will stick with it for now. Thanks, Casey Gregoire -----Original Message----- From: Jake [mailto:jake at bugzilla.org] Sent: Wednesday, March 12, 2003 12:26 PM To: developers at bugzilla.org; geeta.tripathi at bhartitelesoft.com Subject: Re: Bugzilla 2.16.1 Did you run checksetup.pl and did it give you any errors (such as missing modules or permissions issues)? The newsgroup would probably be a more appropriate place for this as it has a larger audiance of potentially helpful people. geeta wrote: > Please help, I am still waiting. > > regards, > > Geeta > > ----- Original Message ----- > *From:* geeta > *To:* developers at bugzilla.org > *Sent:* Wednesday, March 12, 2003 12:54 PM > *Subject:* Re: Bugzilla 2.16.1 > > Hello David, > > I am again installing bugzilla seperately organisation wide, while > accessing bugzilla I got the following errors- > > Directory index forbidden by rule: /home/bugzilla/apache/htdocs/ > > and on running index.cgi on command line I get the following error- > > *Uncaught exception from user code: > Uncaught exception from user code: > Uncaught exception from user code: > mkdir data/template/en: Permission denied at > /usr/lib/perl5/site_perl/5.005/i386-linux/T > emplate/Provider.pm line 368- Check for File-Spec > Carp::croak('mkdir data/template/en: Permission denied') > called at /usr/lib/perl5/5.0050 > 3/File/Path.pm line 140 > File::Path::mkpath('data/template/en', undef, 511) called > at /usr/lib/perl5/5.00503/File > /Path.pm line 133 > File::Path::mkpath('data/template/en/custom') called at > /usr/lib/perl5/site_perl/5.005/i > 386-linux/Template/Provider.pm line 368-Check for File-Spec > > Template::Provider::_init('Template::Provider=HASH(0x84b2244)', > 'HASH(0x8356488)') calle > d at /usr/lib/perl5/site_perl/5.005/i386-linux/Template/Base.pm > line 71-Base for Configurations > Template::Base::new('Template::Provider', > 'HASH(0x8356488)') called at /usr/lib/perl5/si > te_perl/5.005/i386-linux/Template/Config.pm line 116 > Template::Config::provider('Template::Config', > 'HASH(0x8356488)') called at /usr/lib/per > l5/site_perl/5.005/i386-linux/Template/Context.pm line 725-Template > > Template::Context::_init('Template::Context=HASH(0x84b2070)', > 'HASH(0x8356488)') called > at /usr/lib/perl5/site_perl/5.005/i386-linux/Template/Base.pm line 71 > Template::Base::new('Template::Context', > 'HASH(0x8356488)') called at /usr/lib/perl5/sit > e_perl/5.005/i386-linux/Template/Config.pm line 208 > Template::Config::context('Template::Config', > 'HASH(0x8356488)') called at /usr/lib/perl > 5/site_perl/5.005/i386-linux/Template/Service.pm line 159 > > Template::Service::_init('Template::Service=HASH(0x835aae8)', > 'HASH(0x8356488)') called > at /usr/lib/perl5/site_perl/5.005/i386-linux/Template/Base.pm line 71 > Template::Base::new('Template::Service', > 'HASH(0x8356488)') called at /usr/lib/perl5/sit > e_perl/5.005/i386-linux/Template/Config.pm line 226 > Template::Config::service('Template::Config', > 'HASH(0x8356488)') called at /usr/lib/perl > 5/site_perl/5.005/i386-linux/Template.pm line 126 > Template::_init('Template=HASH(0x8357398)', > 'HASH(0x8356488)') called at /usr/lib/perl5/ > site_perl/5.005/i386-linux/Template/Base.pm line 71 > Template::Base::new('Template', 'HASH(0x8356488)') called > at globals.pl line 1593 > require globals.pl called at CGI.pl line 53 > require CGI.pl called at index.cgi line 39 > require CGI.pl called at index.cgi line 39* > > I have installed Bugzilla earlier, but I have never got this > error. Is there any permission issue, this is all I can understand > from the error.............how do I resolve it. > > Please let me know of any further details needed from my end to > resolve this issue. > > Looking forward to hear from you. > > regards, > > Geeta > ---- To view or change your list settings, click here: From pschranz at caracal-tech.com Wed Mar 12 23:20:30 2003 From: pschranz at caracal-tech.com (Paul Schranz) Date: Wed, 12 Mar 2003 15:20:30 -0800 Subject: Win32 Platform Message-ID: <001201c2e8ed$f918a3f0$9800000a@glenndesktop> Bugzilla developers, Is today a good day to pull from CVS a stable version of Bugzilla? >>You are almost certainly better off getting the 2.17 version from CVS (after consultation >>with the Bugzilla Team to make sure you are pulling on a stable day) because we'll be >>doing a load of work to make the Win32 experience more pleasant than it is now. Thanks, Paul -------------- next part -------------- An HTML attachment was scrubbed... URL: From gerv at mozilla.org Wed Mar 12 23:36:28 2003 From: gerv at mozilla.org (Gervase Markham) Date: Wed, 12 Mar 2003 23:36:28 +0000 Subject: Regarding the error in CGI.pl script in Bugzilla-2.16.1 In-Reply-To: <200303121726.h2CHQTT16428@mail.bhartitelesoft.com> References: <200303121726.h2CHQTT16428@mail.bhartitelesoft.com> Message-ID: <3E6FC47C.3080706@mozilla.org> geeta.tripathi at bhartitelesoft.com wrote: > Hello Gerv, > > Reference to the earlier mail, please let me know the > solution for the following error while compiling CGI.pl > script. I don't know offhand, I'm afraid. Geeta - it seems that you are having an unusual amount of trouble installing and configuring Bugzilla. Perhaps your company should consider hiring one of the consultants listed on this page: http://www.bugzilla.org/consulting.html They would be able to devote as much time as was needed to helping you with your problems. Gerv From gerv at mozilla.org Wed Mar 12 23:38:01 2003 From: gerv at mozilla.org (Gervase Markham) Date: Wed, 12 Mar 2003 23:38:01 +0000 Subject: Time Tracking In-Reply-To: References: Message-ID: <3E6FC4D9.40704@mozilla.org> Casey Gregoire wrote: > I have looked at the time tracking features now. What I was looking for is > you start a timer, then when you are done working on that bug you go back to > the bug and hit stop timer. And that time gets added in. (a comment would be > added as well.) There is no feature like that, sorry. What happens if you have two timers running at once? How does it know about holidays or illness? The calculation of time spent is far better done by a human, IMO. Gerv From dswegen at software.plasmon.com Thu Mar 13 10:25:10 2003 From: dswegen at software.plasmon.com (Dave Swegen) Date: Thu, 13 Mar 2003 10:25:10 +0000 Subject: CVS integration patch Message-ID: <20030313102509.GF28150@software.plasmon> I was wondering if someone could give me a pointer about what to do with a set of changes/additions to Bugzilla that we've made that allow greater integration between CVS and BZ. Should I submit these changes, and if so as a patch or a tarball (which would be a lot easier ;) The changes we've made build on the existing script contrib/bugzilla_email_append.pl, and adds some more stuff. More specifically the changes do the following: * Allow a CVS commit log entry to be appended to one or more bugs: This requires a patch to CVS to allow other scripts to cope with filenames with spaces in them, a script that is called by CVSROOT/loginfo, and additions to bugzilla_email_append.pl. * Autolinkify CVS commit log entries in BZ comments to point to a ViewCVS install. This allows the user to look at general info about the file, the old and new revision, as well as a diff between them: It currently only requires a smallish patch against globals.pl (but do it properly will probably require yet another parameter added to BZ). It does not does not do anything to control the bug (i.e. changing status, etc). At the moment this stuff Works For Us (TM). If it is of interest I'd be happy to put in some work to clean it up to some degree. Just let me know what you'd like to see. Cheers Dave From geeta.tripathi at bhartitelesoft.com Thu Mar 13 11:46:02 2003 From: geeta.tripathi at bhartitelesoft.com (geeta) Date: Thu, 13 Mar 2003 17:16:02 +0530 Subject: Fw: Bugzilla 2.16.1-Urgent! Message-ID: <000a01c2e956$26dd6ce0$800310ac@tripathi.bhartitelesoft.com> Hello Dave/Gervase/Jake, I know its bugging to bother you everywhile with issues, but I am again installing Bugzilla I am finding some other kind of problems details not there in the news group & neither I have encountered in my earlier installations. This is one of my very last mails on the installation issues please help- > What could be the reason for the following error while compiling CGI.pl - > > > Subroutine confess redefined at /usr/lib/perl5/5.00503/Exporter.pm line 157 > Exporter::export('Carp', 'main') called at > /usr/lib/perl5/5.00503/Exporter.pm line > 182 > Exporter::import('Carp') called at globals.pl line 80 > main::BEGIN() called at globals.pl line 80 > eval {...} called at globals.pl line 80 > require globals.pl called at CGI.pl line 53 > Subroutine croak redefined at /usr/lib/perl5/5.00503/Exporter.pm line 157 > Exporter::export('Carp', 'main') called at > /usr/lib/perl5/5.00503/Exporter.pm line > 182 > Exporter::import('Carp') called at globals.pl line 80 > main::BEGIN() called at globals.pl line 80 > eval {...} called at globals.pl line 80 > require globals.pl called at CGI.pl line 53 > Subroutine carp redefined at /usr/lib/perl5/5.00503/Exporter.pm line 157 > Exporter::export('Carp', 'main') called at > /usr/lib/perl5/5.00503/Exporter.pm line > 182 > Exporter::import('Carp') called at globals.pl line 80 > main::BEGIN() called at globals.pl line 80 > eval {...} called at globals.pl line 80 > require globals.pl called at CGI.pl line 53 > > There is no clue in the news group for this error. > > Please help. meanwhile while checking from the server error log I get the > error Directory Index forbidden by rule, > other warnings that I get are > Address already in use: make_sock: could not bind to > port 9090, I have installed Apache and defined the port 9090 already another version of apache is installed. > > Now, if I have to check the permissions , please let me know the steps. > regards, > > Geeta From jth at mikrobitti.fi Thu Mar 13 11:51:08 2003 From: jth at mikrobitti.fi (Jouni Heikniemi) Date: Thu, 13 Mar 2003 13:51:08 +0200 Subject: Win32 Platform In-Reply-To: <001201c2e8ed$f918a3f0$9800000a@glenndesktop> Message-ID: <5.1.0.14.2.20030313134937.041bde58@mikrobitti.fi> At 15:20 12.3.2003 -0800, you wrote: >Is today a good day to pull from CVS a stable version of Bugzilla? As good as they've been recently - but if you really want your Win32 installation smoother, you'll wait for bug 84876 to land. Jouni From geeta.tripathi at bhartitelesoft.com Thu Mar 13 11:51:20 2003 From: geeta.tripathi at bhartitelesoft.com (geeta) Date: Thu, 13 Mar 2003 17:21:20 +0530 Subject: Fw: Bugzilla 2.16.1-Urgent! Message-ID: <000c01c2e956$e3467520$800310ac@tripathi.bhartitelesoft.com> ----- Original Message ----- From: "geeta" To: Sent: Thursday, March 13, 2003 5:16 PM Subject: Fw: Bugzilla 2.16.1-Urgent! > > Hello Dave/Gervase/Jake, > > I know its bugging to bother you everywhile with issues, but I am again > installing Bugzilla I am finding some other kind of problems details not > there in the news group & neither I have encountered in my earlier > installations. This is one of my very last mails on the installation issues > please help- > > > > What could be the reason for the following error while compiling CGI.pl - > > > > > > Subroutine confess redefined at /usr/lib/perl5/5.00503/Exporter.pm line > 157 > > Exporter::export('Carp', 'main') called at > > /usr/lib/perl5/5.00503/Exporter.pm line > > 182 > > Exporter::import('Carp') called at globals.pl line 80 > > main::BEGIN() called at globals.pl line 80 > > eval {...} called at globals.pl line 80 > > require globals.pl called at CGI.pl line 53 > > Subroutine croak redefined at /usr/lib/perl5/5.00503/Exporter.pm line 157 > > Exporter::export('Carp', 'main') called at > > /usr/lib/perl5/5.00503/Exporter.pm line > > 182 > > Exporter::import('Carp') called at globals.pl line 80 > > main::BEGIN() called at globals.pl line 80 > > eval {...} called at globals.pl line 80 > > require globals.pl called at CGI.pl line 53 > > Subroutine carp redefined at /usr/lib/perl5/5.00503/Exporter.pm line 157 > > Exporter::export('Carp', 'main') called at > > /usr/lib/perl5/5.00503/Exporter.pm line > > 182 > > Exporter::import('Carp') called at globals.pl line 80 > > main::BEGIN() called at globals.pl line 80 > > eval {...} called at globals.pl line 80 > > require globals.pl called at CGI.pl line 53 > > > > There is no clue in the news group for this error. > > > > Please help. meanwhile while checking from the server error log I get the > > error Directory Index forbidden by rule, > > other warnings that I get are > > Address already in use: make_sock: could not bind to > > port 9090, I have installed Apache and defined the port 9090 already > another version of apache is installed. > > > > > Now, if I have to check the permissions , please let me know the steps. > > regards, > > > > Geeta > > > ---- > To view or change your list settings, click here: > > From jamessmith449 at lycos.co.uk Thu Mar 13 15:13:00 2003 From: jamessmith449 at lycos.co.uk (James Smith) Date: Thu, 13 Mar 2003 10:13:00 -0500 Subject: Bugzilla Extensions References: <20030313102509.GF28150@software.plasmon> Message-ID: <002801c2e973$0b8e9640$1fec55c8@supercabletv.net.co> Hi I'm new to bugzilla as a developer. I wanna know if bugzilla has, for example, Corba IDL Interfaces to allow use bugzilla "api" from other languages like java. Or at least if 'bugzilla' has plans to do something similar. I've serched all around the bugzilla site and couldn't found any reference about this matter. I'm asking you this, 'cause I'm planning develop some extensions to bugzilla. Thanks Hope you can help. Bye James From caseyg at chsamerica.com Thu Mar 13 15:15:53 2003 From: caseyg at chsamerica.com (Casey Gregoire) Date: Thu, 13 Mar 2003 10:15:53 -0500 Subject: Time Tracking Message-ID: We want more of a punch in, punch out on a bug. I can just add my own script for it, I just wanted to see what the channel thought of it. May be it could be an option? (If I wrote properly.) or just a patch. Thanks, Casey Gregoire -----Original Message----- From: Gervase Markham [mailto:gerv at mozilla.org] Sent: Wednesday, March 12, 2003 6:38 PM To: developers at bugzilla.org Subject: Re: Time Tracking Casey Gregoire wrote: > I have looked at the time tracking features now. What I was looking for is > you start a timer, then when you are done working on that bug you go back to > the bug and hit stop timer. And that time gets added in. (a comment would be > added as well.) There is no feature like that, sorry. What happens if you have two timers running at once? How does it know about holidays or illness? The calculation of time spent is far better done by a human, IMO. Gerv ---- To view or change your list settings, click here: From jason at pyeron.com Thu Mar 13 15:30:50 2003 From: jason at pyeron.com (Jason Pyeron) Date: Thu, 13 Mar 2003 10:30:50 -0500 (EST) Subject: Time Tracking In-Reply-To: Message-ID: This sounds like a flex time system integrated with the qa software. Should I assume you are doing this for billing? If so you might just want to "link" bugzilla into your flex time system. This is going a little off topic, I would be more than happy to discuss how we handle this issue, off list of course. Sincerely, Jason Pyeron On Thu, 13 Mar 2003, Casey Gregoire wrote: We want more of a punch in, punch out on a bug. I can just add my own script for it, I just wanted to see what the channel thought of it. May be it could be an option? (If I wrote properly.) or just a patch. Thanks, Casey Gregoire -----Original Message----- From: Gervase Markham [mailto:gerv at mozilla.org] Sent: Wednesday, March 12, 2003 6:38 PM To: developers at bugzilla.org Subject: Re: Time Tracking Casey Gregoire wrote: > I have looked at the time tracking features now. What I was looking for is > you start a timer, then when you are done working on that bug you go back to > the bug and hit stop timer. And that time gets added in. (a comment would be > added as well.) There is no feature like that, sorry. What happens if you have two timers running at once? How does it know about holidays or illness? The calculation of time spent is far better done by a human, IMO. Gerv ---- To view or change your list settings, click here: ---- To view or change your list settings, click here: -- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- - - - Jason Pyeron http://www.pyerotechnics.com - - Owner & Lead Pyerotechnics Development, Inc. - - +1 410 808 6646 (c) 500 West University Parkway #1S - - +1 410 467 2266 (f) Baltimore, Maryland 21210-3253 - - - -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- This message is for the designated recipient only and may contain privileged, proprietary, or otherwise private information. If you have received it in error, purge the message from your system and notify the sender immediately. Any other use of the email by you is prohibited. From caseyg at chsamerica.com Thu Mar 13 15:30:00 2003 From: caseyg at chsamerica.com (Casey Gregoire) Date: Thu, 13 Mar 2003 10:30:00 -0500 Subject: Time Tracking Message-ID: Well no, its not for billing though. Its for the same function as what is in Bugzilla now, but it is meant to have less room for 'fudging' time worked. Thanks, Casey Gregoire -----Original Message----- From: Jason Pyeron [mailto:jason at pyeron.com] Sent: Thursday, March 13, 2003 10:31 AM To: developers at bugzilla.org Subject: Re: Time Tracking This sounds like a flex time system integrated with the qa software. Should I assume you are doing this for billing? If so you might just want to "link" bugzilla into your flex time system. This is going a little off topic, I would be more than happy to discuss how we handle this issue, off list of course. Sincerely, Jason Pyeron On Thu, 13 Mar 2003, Casey Gregoire wrote: We want more of a punch in, punch out on a bug. I can just add my own script for it, I just wanted to see what the channel thought of it. May be it could be an option? (If I wrote properly.) or just a patch. Thanks, Casey Gregoire -----Original Message----- From: Gervase Markham [mailto:gerv at mozilla.org] Sent: Wednesday, March 12, 2003 6:38 PM To: developers at bugzilla.org Subject: Re: Time Tracking Casey Gregoire wrote: > I have looked at the time tracking features now. What I was looking for is > you start a timer, then when you are done working on that bug you go back to > the bug and hit stop timer. And that time gets added in. (a comment would be > added as well.) There is no feature like that, sorry. What happens if you have two timers running at once? How does it know about holidays or illness? The calculation of time spent is far better done by a human, IMO. Gerv ---- To view or change your list settings, click here: ---- To view or change your list settings, click here: -- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- - - - Jason Pyeron http://www.pyerotechnics.com - - Owner & Lead Pyerotechnics Development, Inc. - - +1 410 808 6646 (c) 500 West University Parkway #1S - - +1 410 467 2266 (f) Baltimore, Maryland 21210-3253 - - - -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- This message is for the designated recipient only and may contain privileged, proprietary, or otherwise private information. If you have received it in error, purge the message from your system and notify the sender immediately. Any other use of the email by you is prohibited. ---- To view or change your list settings, click here: From usmlekaplan at yahoo.com Thu Mar 13 16:48:22 2003 From: usmlekaplan at yahoo.com (usmle Prep) Date: Thu, 13 Mar 2003 08:48:22 -0800 (PST) Subject: buglist.cgi times out Message-ID: <20030313164822.31713.qmail@web20210.mail.yahoo.com> Hi Friends, Buglist.cgi times out on my bugzilla installation.I did verify that the permissions are good and the same buglist.cgi script runs on another instance of bugzilla on the same server(different database). I feel it's something to do with the database. Do any one know what might have gone wrong with the database? One more thing is if I run a older version of buglist.cgi it works fine but it doesn't have customfields support. Thanks a lot for your help, Srikant Server error! The server encountered an internal error and was unable to complete your request. Error message: Premature end of script headers: buglist.cgi If you think this is a server error, please contact the webmaster Error 500 __________________________________________________ Do you Yahoo!? Yahoo! Web Hosting - establish your business online http://webhosting.yahoo.com From myk at mozilla.org Thu Mar 13 18:13:39 2003 From: myk at mozilla.org (Myk Melez) Date: Thu, 13 Mar 2003 10:13:39 -0800 Subject: buglist.cgi times out In-Reply-To: <20030313164822.31713.qmail@web20210.mail.yahoo.com> References: <20030313164822.31713.qmail@web20210.mail.yahoo.com> Message-ID: <3E70CA53.6050006@mozilla.org> Is buglist.cgi timing out or throwing a server error? If the problem is a server error, look in your web server logs for clues to what the error is. If the problem is buglist.cgi timing out, on the other hand, you may have database corruption. To find out, add "&debug=1" to the working buglist.cgi URL to see the query it produces, then connect to both databases with the mysql client and run "explain" (http://www.mysql.com/doc/en/EXPLAIN.html) on each one to see how MySQL expects to optimize the query. If there are differences, the problem is probably database corruption, and the best solution is to dump and rebuild the database using mysqldump. This question is better answered in the netscape.public.mozilla.webtools newsgroup. cc:ing and setting follow-ups appropriately. -myk usmle Prep wrote: >Hi Friends, >Buglist.cgi times out on my bugzilla installation.I >did verify that the permissions are good and the same >buglist.cgi script runs on another instance of >bugzilla on the same server(different database). I >feel it's something to do with the database. Do any >one know what might have gone wrong with the >database? One more thing is if I run a older version >of buglist.cgi it works fine but it doesn't have >customfields support. >Thanks a lot for your help, >Srikant >Server error! > >The server encountered an internal error and was >unable to complete your request. > > Error message: >Premature end of script headers: buglist.cgi > >If you think this is a server error, please contact >the webmaster >Error 500 > >__________________________________________________ >Do you Yahoo!? >Yahoo! Web Hosting - establish your business online >http://webhosting.yahoo.com >---- >To view or change your list settings, click here: > > > From justdave at syndicomm.com Thu Mar 13 18:41:09 2003 From: justdave at syndicomm.com (David Miller) Date: Thu, 13 Mar 2003 13:41:09 -0500 Subject: Fw: Bugzilla 2.16.1-Urgent! In-Reply-To: <000a01c2e956$26dd6ce0$800310ac@tripathi.bhartitelesoft.com> References: <000a01c2e956$26dd6ce0$800310ac@tripathi.bhartitelesoft.com> Message-ID: On 3/13/03 5:16 PM +0530, geeta wrote: > Hello Dave/Gervase/Jake, > > I know its bugging to bother you everywhile with issues, but I am again > installing Bugzilla I am finding some other kind of problems details not > there in the news group & neither I have encountered in my earlier > installations. This is one of my very last mails on the installation issues > please help- It's not so much the quantity of what you ask, it's the way you ask. This is a community-supported project. We answer questions when we can on the list, but lots of other people here that are working on Bugzilla can answer questions, too. Addressing your messages to a specific person is a good way to not get them answered. That person may be busy at the time and not have time to post, or they might not know the answer. And because you addressed it to a specific person, others might avoid answering because they're assuming the person you addressed it to is going to answer. You'll get much better results if you don't address your questions to a specific person. Also, the mailing list you're posting on is intended for people who are developing Bugzilla, planning to contribute, or are working on making large customizations to their installations. Discussion here is generally centered more around delevopment direction or specific alterations to Bugzilla code. Most of your questions have been general end-user installation questions, which would be much better dealt with on the mozilla-webtools mailing list, which has a much larger audience and many more people who have installed Bugzilla in the past that may have run into the problems you're having and have a solution for you. The other reason people have been avoiding answering your posts is because the above has already been explained to you multiple times on this list and in private emails and you haven't been paying attention. It's hard to help someone who doesn't follow the helpful instructions they've already been given. As Gerv mentioned in an earlier post, if you're having this much trouble dealing with it, you might want to look into hiring one of the consultants to set it up for you, because you're obviously having a pretty hard time of it yourself. Now on to your questions (Despite the above rant, I'm not a hateful spiteful person and I'll still try to help - just keep the above in mind in the future) >> What could be the reason for the following error while compiling CGI.pl - >> >> >> Subroutine confess redefined at /usr/lib/perl5/5.00503/Exporter.pm line > 157 You are using Perl 5.00503 and you have the "use Carp" line and the "die_with_dignity" sub uncommented in globals.pl. That sub doesn't work well on 5.00503. It's generally useful on 5.6.0 and up. Comment out the "use Carp" line, the die_with_dignity sub, and the SIG{__DIE__} trap right under the sub. That function is only intended for debugging purposes anyway, you don't want to run Bugzilla with that enabled on a day to day basis. >> Please help. meanwhile while checking from the server error log I get the >> error Directory Index forbidden by rule, The Options line for your Bugzilla direction in your httpd.conf file needs to include the "Indexes" option. Likewise, you need to find you DirectoryIndex directive and make sure "index.cgi" is listed in it. >> other warnings that I get are >> Address already in use: make_sock: could not bind to >> port 9090, I have installed Apache and defined the port 9090 already > another version of apache is installed. Is the other copy of Apache also using port 9090? If so, use a different port on this one. -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From gerv at mozilla.org Thu Mar 13 18:45:36 2003 From: gerv at mozilla.org (Gervase Markham) Date: Thu, 13 Mar 2003 18:45:36 +0000 Subject: Time Tracking In-Reply-To: References: Message-ID: <3E70D1D0.1010004@mozilla.org> Jason Pyeron wrote: > This sounds like a flex time system integrated with the qa software. > Should I assume you are doing this for billing? If so you might just want > to "link" bugzilla into your flex time system. This seems to me to be a smarter route than trying to get Bugzilla to support this stuff, which is rather away from our main mission. Gerv From justdave at syndicomm.com Thu Mar 13 18:49:03 2003 From: justdave at syndicomm.com (David Miller) Date: Thu, 13 Mar 2003 13:49:03 -0500 Subject: Stalled post to developers In-Reply-To: <1047580660.10891.23.camel@hawk> References: <200303131836.h2DIa2XC020480@sinclair.syndicomm.com> <1047580660.10891.23.camel@hawk> Message-ID: On 3/13/03 10:37 AM -0800, Joseph Phillips wrote: > 1. Is it possible, in the latest stable bugzilla version 2.16.2, to > delete users? In previous versions, I had to go into the database to > manually delete users. There's an option in editparams.cgi to enable user deletion. Do note that the user delete code is not very good at cleaning up the database integrity at the moment. If you delete a user, you'll need to run sanitycheck.cgi and manually clean up anything it breaks. My personal recommendation is to edit the user and change their email address to "deleteduserXXX at nowhere.tld" where XXX is their userID from the database, then just make their real name be "Deleted User". > 2. Would it be ok to dump the database from an old bugzilla 2.14.1 > installation, and then import this database into a freshly installed > 2.16.2 installation? This should work fine. Just make sure you run checksetup.pl to upgrade the database to the 2.16.2 format after you restore it. -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From justdave at syndicomm.com Thu Mar 13 19:12:10 2003 From: justdave at syndicomm.com (David Miller) Date: Thu, 13 Mar 2003 14:12:10 -0500 Subject: Fw: Bugzilla 2.16.1-Urgent! In-Reply-To: References: <000a01c2e956$26dd6ce0$800310ac@tripathi.bhartitelesoft.com> Message-ID: On 3/13/03 1:41 PM -0500, David Miller wrote: > The Options line for your Bugzilla direction in your httpd.conf file needs > to include the "Indexes" option. Likewise, you need to find you > DirectoryIndex directive and make sure "index.cgi" is listed in it. Oops, I meant "The Options line for your Bugzilla *directory*" in the above sentence. -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From caseyg at chsamerica.com Thu Mar 13 19:40:20 2003 From: caseyg at chsamerica.com (Casey Gregoire) Date: Thu, 13 Mar 2003 14:40:20 -0500 Subject: Time Tracking Message-ID: I figured I would throw it out there and see how people liked it. Thanks, Casey Gregoire -----Original Message----- From: Gervase Markham [mailto:gerv at mozilla.org] Sent: Thursday, March 13, 2003 1:46 PM To: developers at bugzilla.org Subject: Re: Time Tracking Jason Pyeron wrote: > This sounds like a flex time system integrated with the qa software. > Should I assume you are doing this for billing? If so you might just want > to "link" bugzilla into your flex time system. This seems to me to be a smarter route than trying to get Bugzilla to support this stuff, which is rather away from our main mission. Gerv ---- To view or change your list settings, click here: From jphillips at amphus.com Thu Mar 13 19:58:31 2003 From: jphillips at amphus.com (Joseph Phillips) Date: 13 Mar 2003 11:58:31 -0800 Subject: checkboxes on editusers.cgi Message-ID: <1047585510.10891.69.camel@hawk> In the editusers.cgi page, in the "Group Access" and "Privileges" section, I notice that there are two columns of checkboxes standing vertically parallel to each other. What is the difference between these two checkbox columns? Thanks. From jon at vmware.com Thu Mar 13 20:05:59 2003 From: jon at vmware.com (Jonathan Schatz) Date: 13 Mar 2003 12:05:59 -0800 Subject: checkboxes on editusers.cgi In-Reply-To: <1047585510.10891.69.camel@hawk> References: <1047585510.10891.69.camel@hawk> Message-ID: <1047585959.32510.6.camel@jonschatz-lx.vmware.com> On Thu, 2003-03-13 at 11:58, Joseph Phillips wrote: > In the editusers.cgi page, in the "Group Access" and "Privileges" > section, I notice that there are two columns of checkboxes standing > vertically parallel to each other. > > What is the difference between these two checkbox columns? Thanks. one grants membership to the group, the other gives the user the ability to grant others membership in the same group. -jon -- Jonathan Schatz Engineering System Administrator VMware, Inc "Te occidere possunt sed te edere non possunt nefas est." From jphillips at amphus.com Thu Mar 13 20:18:02 2003 From: jphillips at amphus.com (Joseph Phillips) Date: 13 Mar 2003 12:18:02 -0800 Subject: checkboxes on editusers.cgi In-Reply-To: <1047585959.32510.6.camel@jonschatz-lx.vmware.com> References: <1047585510.10891.69.camel@hawk> <1047585959.32510.6.camel@jonschatz-lx.vmware.com> Message-ID: <1047586682.10891.74.camel@hawk> I just figured that out right before your reply. The problem is the description text isn't formatted well on the web page, so it doesn't line up directly over each column. I realized that the "|" is meant as a dividing line between the two pieces of text description. The part before the "|" applies to the column on the left, and the part after the "|" applies to the column on the right. On Thu, 2003-03-13 at 12:05, Jonathan Schatz wrote: > On Thu, 2003-03-13 at 11:58, Joseph Phillips wrote: > > In the editusers.cgi page, in the "Group Access" and "Privileges" > > section, I notice that there are two columns of checkboxes standing > > vertically parallel to each other. > > > > What is the difference between these two checkbox columns? Thanks. > > one grants membership to the group, the other gives the user the ability > to grant others membership in the same group. > > -jon From justdave at syndicomm.com Thu Mar 13 20:22:33 2003 From: justdave at syndicomm.com (David Miller) Date: Thu, 13 Mar 2003 15:22:33 -0500 Subject: checkboxes on editusers.cgi In-Reply-To: <1047586682.10891.74.camel@hawk> References: <1047585510.10891.69.camel@hawk> <1047585959.32510.6.camel@jonschatz-lx.vmware.com> <1047586682.10891.74.camel@hawk> Message-ID: On 3/13/03 12:18 PM -0800, Joseph Phillips wrote: > I just figured that out right before your reply. > > The problem is the description text isn't formatted well on the web > page, so it doesn't line up directly over each column. I realized that > the "|" is meant as a dividing line between the two pieces of text > description. The part before the "|" applies to the column on the left, > and the part after the "|" applies to the column on the right. Actually, the | is supposed to be a line connecting the first column to the top row of text. They're spaced above each other to avoid making the page too wide. Can give others access to this group | Is a member of this group [ ] [ ] admin [ ] [ ] tweakparams etc That does seem to confuse people from time to time, but I have yet to see a better idea for how to display it. :) -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From usmlekaplan at yahoo.com Thu Mar 13 22:11:15 2003 From: usmlekaplan at yahoo.com (usmle Prep) Date: Thu, 13 Mar 2003 14:11:15 -0800 (PST) Subject: buglist.cgi times out In-Reply-To: <3E70CA53.6050006@mozilla.org> Message-ID: <20030313221115.32302.qmail@web20203.mail.yahoo.com> Hi Myk, Thanks a lot for taking the time to help me.I really appreciate that.I did run a explain on the query that is getting executed but seems like it's fine from the explain stand point of view. Only 40 rows are present in bugs database.I used mysqldump to recreate the database but still it is behaving the same way .Timing out. Do u have any more suggestions. Thanks again, Srikant --- Myk Melez wrote: > Is buglist.cgi timing out or throwing a server > error? If the problem is > a server error, look in your web server logs for > clues to what the error > is. If the problem is buglist.cgi timing out, on > the other hand, you > may have database corruption. > > To find out, add "&debug=1" to the working > buglist.cgi URL to see the > query it produces, then connect to both databases > with the mysql client > and run "explain" > (http://www.mysql.com/doc/en/EXPLAIN.html) on each > one > to see how MySQL expects to optimize the query. If > there are > differences, the problem is probably database > corruption, and the best > solution is to dump and rebuild the database using > mysqldump. > > This question is better answered in the > netscape.public.mozilla.webtools > newsgroup. cc:ing and setting follow-ups > appropriately. > > -myk > > > usmle Prep wrote: > > >Hi Friends, > >Buglist.cgi times out on my bugzilla installation.I > >did verify that the permissions are good and the > same > >buglist.cgi script runs on another instance of > >bugzilla on the same server(different database). I > >feel it's something to do with the database. Do any > >one know what might have gone wrong with the > >database? One more thing is if I run a older > version > >of buglist.cgi it works fine but it doesn't have > >customfields support. > >Thanks a lot for your help, > >Srikant > >Server error! > > > >The server encountered an internal error and was > >unable to complete your request. > > > > Error message: > >Premature end of script headers: buglist.cgi > > > >If you think this is a server error, please contact > >the webmaster > >Error 500 > > > >__________________________________________________ > >Do you Yahoo!? > >Yahoo! Web Hosting - establish your business online > >http://webhosting.yahoo.com > >---- > >To view or change your list settings, click here: > > > > > > > > ---- > To view or change your list settings, click here: > __________________________________________________ Do you Yahoo!? Yahoo! Web Hosting - establish your business online http://webhosting.yahoo.com From jphillips at amphus.com Thu Mar 13 23:01:55 2003 From: jphillips at amphus.com (Joseph Phillips) Date: 13 Mar 2003 15:01:55 -0800 Subject: groups and entering bugs Message-ID: <1047596515.10891.181.camel@hawk> Is there a way to make it so that when a user enter a new bug, and when that user belongs to more than one group, that if he selects a particular component for the project he is submitting the bug one, that the act of selecting the particular component will force the bug to be submitted as belonging to one particular group? Currently, the user who belongs to more than one group must choose which group with which he would like to submit the bug. This is a problem, because I can't rely on the user to make the correct choice. Also, is there a way to prevent the user from entering the bug as a public bug? This can happen if the user does not choose any of the groups when entering the bug. From gerv at mozilla.org Fri Mar 14 07:39:15 2003 From: gerv at mozilla.org (Gervase Markham) Date: Fri, 14 Mar 2003 07:39:15 +0000 Subject: Bugzilla Extensions In-Reply-To: <002801c2e973$0b8e9640$1fec55c8@supercabletv.net.co> References: <20030313102509.GF28150@software.plasmon> <002801c2e973$0b8e9640$1fec55c8@supercabletv.net.co> Message-ID: <3E718723.1060909@mozilla.org> James Smith wrote: > I'm new to bugzilla as a developer. I wanna know if bugzilla has, for > example, Corba IDL Interfaces to allow use bugzilla "api" from other > languages like java. Or at least if 'bugzilla' has plans to do something > similar. We have no Corba IDL interfaces, no. Bugzilla is not a library, and so does not have an API in that sense. > I've serched all around the bugzilla site and couldn't found any reference > about this matter. > > I'm asking you this, 'cause I'm planning develop some extensions to > bugzilla. What sort of extensions? We'd be able to answer your question much better if you were more specific. Gerv From gerv at mozilla.org Fri Mar 14 07:40:47 2003 From: gerv at mozilla.org (Gervase Markham) Date: Fri, 14 Mar 2003 07:40:47 +0000 Subject: CVS integration patch In-Reply-To: <20030313102509.GF28150@software.plasmon> References: <20030313102509.GF28150@software.plasmon> Message-ID: <3E71877F.5040208@mozilla.org> Dave Swegen wrote: > I was wondering if someone could give me a pointer about what to do > with a set of changes/additions to Bugzilla that we've made that allow > greater integration between CVS and BZ. Should I submit these changes, and > if so as a patch or a tarball (which would be a lot easier ;) You should attach them as diff -u to a Bugzilla bug (as it says to do on http://www.bugzilla.org ;-) > At the moment this stuff Works For Us (TM). If it is of interest I'd be > happy to put in some work to clean it up to some degree. Just let me > know what you'd like to see. Well, it would need to be seen first. :-) It may turn out that no-one has enough time to do anything with it, but attaching it to a bug is a good first step. Gerv From justdave at syndicomm.com Fri Mar 14 08:37:09 2003 From: justdave at syndicomm.com (David Miller) Date: Fri, 14 Mar 2003 03:37:09 -0500 Subject: groups and entering bugs In-Reply-To: <1047596515.10891.181.camel@hawk> References: <1047596515.10891.181.camel@hawk> Message-ID: On 3/13/03 3:01 PM -0800, Joseph Phillips wrote: > Is there a way to make it so that when a user enter a new bug, and when > that user belongs to more than one group, that if he selects a > particular component for the project he is submitting the bug one, that > the act of selecting the particular component will force the bug to be > submitted as belonging to one particular group? > > Currently, the user who belongs to more than one group must choose which > group with which he would like to submit the bug. This is a problem, > because I can't rely on the user to make the correct choice. > > Also, is there a way to prevent the user from entering the bug as a > public bug? This can happen if the user does not choose any of the > groups when entering the bug. You can do this per product (not per component) in 2.17.3 and up. It has the capability of making specific groups be mandatory on specific products (they don't even get a checkbox to uncheck, it just happens) -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From dswegen at software.plasmon.com Fri Mar 14 09:05:59 2003 From: dswegen at software.plasmon.com (Dave Swegen) Date: Fri, 14 Mar 2003 09:05:59 +0000 Subject: CVS integration patch In-Reply-To: <3E71877F.5040208@mozilla.org> References: <20030313102509.GF28150@software.plasmon> <3E71877F.5040208@mozilla.org> Message-ID: <20030314090559.GG28150@software.plasmon> On Fri, Mar 14, 2003 at 07:40:47AM +0000, Gervase Markham wrote: > Dave Swegen wrote: > >I was wondering if someone could give me a pointer about what to do > >with a set of changes/additions to Bugzilla that we've made that allow > >greater integration between CVS and BZ. Should I submit these changes, and > >if so as a patch or a tarball (which would be a lot easier ;) > > You should attach them as diff -u to a Bugzilla bug (as it says to do on > http://www.bugzilla.org ;-) But doing that precludes me from being bone-idle ;) > > >At the moment this stuff Works For Us (TM). If it is of interest I'd be > >happy to put in some work to clean it up to some degree. Just let me > >know what you'd like to see. > > Well, it would need to be seen first. :-) It may turn out that no-one > has enough time to do anything with it, but attaching it to a bug is a > good first step. Right. I'll see if I can get this stuff done in The Very Near Future. Cheers Dave From jphillips at amphus.com Fri Mar 14 20:50:18 2003 From: jphillips at amphus.com (Joseph Phillips) Date: 14 Mar 2003 12:50:18 -0800 Subject: "Group X Bugs Access" drop-down menu item Message-ID: <1047675017.12408.23.camel@hawk> I have a bugzilla 2.10 installation. This has 4 products. When the user goes to enter a new bug on any of these products, towards the bottom of the page there are 2 drop-down menu items. The first one has two choices: 1. "People not in the 'Group 1 Bugs Access' group can see this bug" 2. "Only people in the 'Group 1 Bugs Access' group can see this bug" The second drop-down menu item has the same two choices, but for 'Group 2' instead. I do not have "usebuggroups" or "usebuggroupsentry" turned on (and I never have in the past). So why are these menu items appearing in bugzilla? I don't want them to. Thanks. From justdave at syndicomm.com Fri Mar 14 21:46:35 2003 From: justdave at syndicomm.com (David Miller) Date: Fri, 14 Mar 2003 16:46:35 -0500 Subject: "Group X Bugs Access" drop-down menu item In-Reply-To: <1047675017.12408.23.camel@hawk> References: <1047675017.12408.23.camel@hawk> Message-ID: I'm copying this to the mozilla-webtools list because this is more of a tech-support question than a development issue. Please direct any replies there. On 3/14/03 12:50 PM -0800, Joseph Phillips wrote: > I have a bugzilla 2.10 installation. 2.10 is full of security holes. You need to upgrade. > This has 4 products. When the > user goes to enter a new bug on any of these products, towards the > bottom of the page there are 2 drop-down menu items. The first one has > two choices: > > 1. "People not in the 'Group 1 Bugs Access' group can see this bug" > 2. "Only people in the 'Group 1 Bugs Access' group can see this bug" > > The second drop-down menu item has the same two choices, but for 'Group > 2' instead. > > I do not have "usebuggroups" or "usebuggroupsentry" turned on (and I > never have in the past). So why are these menu items appearing in > bugzilla? I don't want them to. Thanks. Because the groups exist. Delete the groups and they won't show up. The "usebuggroups" only deals with whether products get tied to groups or not, not whether you can actually use groups (which is kinda confusing - it works better in 2.17.3 :) -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From etzwane at schwag.org Sat Mar 15 00:10:42 2003 From: etzwane at schwag.org (Sean McAfee) Date: Fri, 14 Mar 2003 19:10:42 -0500 Subject: An alternate approach to custom fields Message-ID: <20030315001042.BB0B1BCFA@diggity.schwag.org> My company is looking to replace its current incident-tracking system (TeamTrack). We want to keep as many of its good features as possible while doing away with the flaws that are a continual source of headaches--the horrible performance, mostly, although its proprietary, black-box nature is also annoying. I've examined a good number of free systems, and Bugzilla appears to offer most of the features that we'd like. One feature is lacking, however, that we'd really hate to do without: custom fields. I've read the "Living Without Custom Fields" essay at . I had little success trying to apply the existing custom-fields patch. None of the other suggestions really address all of the situations that might arise here. For example, one of our workflows describes the testing path of individual returned parts. Each part is run through a dozen or more tests. The result of each test is recorded via a drop-down list on our current system's web interface. Different tests are performed at different times. Trying to squeeze all of this information into a single free-form text field and depending on users to keep the field formatted in a way that's suitable for querying seems like a recipe for disaster. I've thought up a possible way of implementing custom fields by combining options 2, 4, and 6 from the "Living Without" page ("Bug entry templates", "Co-opt an existing field", and "Hacking" respectively). It seems feasible to me, but I'm not sufficiently familiar with the Bugzilla internals to feel really confident about my assessment. So I thought I'd toss my plan out on the developers' list and see what y'all thought. In a nutshell, the plan is this: 1. Co-opt an unused text field and use it to store a serialized representation of my desired custom fields--using XML, say. 2. Add a call to the bug entry and bug editing templates that expands the serialized custom fields into additional HTML form elements. 3. Modify the targets of the previous two pages such that they collapse the extra form elements back into serialized form. Even a basic approach like this would supply much of the functionality we'd hate to have to do without, but I can easily see building more sophisticated layers on top of it: state-dependent views, per-project fields, etc. As far as I can tell, the extra features I'm proposing would be largely orthogonal to the existing Bugzilla codebase, and thus easy to maintain as new versions of Bugzilla are released. The only exception is querying; I wouldn't want to have to figure out how to incorporate searching on custom fields into the existing framework. I'm willing to live without this feature, however, at least via the web browser. I can envision a new Perl query library that transparently presents the custom fields alongside the standard ones. So, how does all of this sound? Has it been tried before, with disastrous results? Any obvious gotchas that I'm missing? Thanks in advance for any advice. -- Sean McAfee etzwane at schwag.org From geeta.tripathi at bhartitelesoft.com Sat Mar 15 08:30:14 2003 From: geeta.tripathi at bhartitelesoft.com (geeta) Date: Sat, 15 Mar 2003 14:00:14 +0530 Subject: Need help with this one a bugging error Message-ID: <002001c2eacd$27525e60$800310ac@tripathi.bhartitelesoft.com> Hello , When I run index.cgi from my command line my script runs ok. But when I run from the apache gives the following errors- I understand that this is a mysql shared library issue , I have done the work arounds. But, still getting the same errors. Another issue I have installled Msql-Mysql-modules-1.2209, but checksetup.pl is always giving found v 2.0409 instead of 1.2209. Now, is there any way to resolve this issue. Enclosing the error that I got- Uncaught exception from user code: install_driver(mysql) failed: Uncaught exception from user code: Uncaught exception from user code: Can't load '/usr/lib/perl5/site_perl/5.005/i386-linux/auto/DBD/mysql/mysql.so' for module DBD::mysql: libmysqlclient.so.10: cannot open shared object file: No such file or di rectory at /usr/lib/perl5/5.00503/i386-linux/DynaLoader.pm line 169. at (eval 4) line 3 Carp::croak('Can\'t load \'/usr/lib/perl5/site_perl/5.005/i386-linux/auto/DBD...') called at /usr/lib/perl5/5.00503/i386-linux/DynaLoader.pm line 93 DynaLoader::croak('Can\'t load \'/usr/lib/perl5/site_perl/5.005/i386-linux/auto/DBD ...') called at /usr/lib/perl5/5.00503/i386-linux/DynaLoader.pm line 169 DynaLoader::bootstrap('DBD::mysql', 2.0409) called at /usr/lib/perl5/site_perl/5.00 5/i386-linux/DBD/mysql.pm line 21 require DBD/mysql.pm called at (eval 4) line 3 eval 'package DBI::_firesafe; require DBD::mysql ;' called at /usr/lib/perl5/site_perl/5.005/i386-linux/DBI.pm line 478 DBI::install_driver('DBI', 'mysql') called at /usr/lib/perl5/site_perl/5.005/i386-l inux/DBI.pm line 402 DBI::connect('DBI', 'DBI:mysql:host=localhost;database=test;port=3306', 'test', 'te sttest') called at globals.pl line 140 main::ConnectToDatabase() called at index.cgi line 46 eval 'package DBI::_firesafe; require DBD::mysql ;' called at /usr/lib/perl5/site_perl/5.005/i386-linux/DBI.pm line 478 DBI::install_driver('DBI', 'mysql') called at /usr/lib/perl5/site_perl/5.005/i386-l inux/DBI.pm line 402 DBI::connect('DBI', 'DBI:mysql:host=localhost;database=test;port=3306', 'test', 'te sttest') called at globals.pl line 140 main::ConnectToDatabase() called at index.cgi line 46 Perhaps a required shared library or dll isn't installed where expected at globals.pl line 140 Carp::croak('install_driver(mysql) failed: Uncaught exception from user code:...') called at /usr/lib/perl5/site_perl/5.005/i386-linux/DBI.pm line 499 DBI::install_driver('DBI', 'mysql') called at /usr/lib/perl5/site_perl/5.005/i386-l inux/DBI.pm line 402 DBI::connect('DBI', 'DBI:mysql:host=localhost;database=test;port=3306', 'test', 'te sttest') called at globals.pl line 140 main::ConnectToDatabase() called at index.cgi line 46 Looking forward to hear from you. Thanks and regards, Geeta -------------- next part -------------- An HTML attachment was scrubbed... URL: From bbaetz at acm.org Sat Mar 15 08:59:44 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Sat, 15 Mar 2003 19:59:44 +1100 Subject: An alternate approach to custom fields In-Reply-To: <20030315001042.BB0B1BCFA@diggity.schwag.org> References: <20030315001042.BB0B1BCFA@diggity.schwag.org> Message-ID: <20030315085943.GA31727@mango.home> On Fri, Mar 14, 2003 at 07:10:42PM -0500, Sean McAfee wrote: > For example, one of our workflows describes the testing path of individual > returned parts. Each part is run through a dozen or more tests. The result > of each test is recorded via a drop-down list on our current system's web > interface. Different tests are performed at different times. Trying to > squeeze all of this information into a single free-form text field and > depending on users to keep the field formatted in a way that's suitable > for querying seems like a recipe for disaster. If the tests are just a pass/fail basis, then you could use keywords or (in 2.17) bug flags to represent the various tests. If you sometimes run tests more than once, then you'll have to use bug flags. If they're not just pass/fail, you could probably hack the +/-/? bits to have more options. If they're freeform comments, then it could be more complicated, if you don't want to just have them in there as normal comments > In a nutshell, the plan is this: > I suspect that the effort involved in doing that (especially if you want to search) is probably more than the effort required to get custom fields done properly. There are bugs with searches on some of the flag fields, but they're still easier to fix than having the search routines somehow grovel through xml (or any other format) > As far as I can tell, the extra features I'm proposing would be largely > orthogonal to the existing Bugzilla codebase, and thus easy to maintain as > new versions of Bugzilla are released. The only exception is querying; I > wouldn't want to have to figure out how to incorporate searching on custom > fields into the existing framework. I'm willing to live without this > feature, however, at least via the web browser. I can envision a new Perl > query library that transparently presents the custom fields alongside the > standard ones. > Thats the plan (using Bugzilla::Search, basically) > So, how does all of this sound? Has it been tried before, with disastrous > results? Any obvious gotchas that I'm missing? > I'm not aware of anyone having tried to handle custom fields via xml serialisation. The issue with custom fields is not that we don't know how to do it, but that none of the developers have time to do it. Bradley From geeta.tripathi at bhartitelesoft.com Sat Mar 15 11:32:58 2003 From: geeta.tripathi at bhartitelesoft.com (geeta) Date: Sat, 15 Mar 2003 17:02:58 +0530 Subject: Need help with this one a bugging error Message-ID: <001601c2eae6$a4d20480$800310ac@tripathi.bhartitelesoft.com> Hello David, This problem is solved this issue occured because the mysql permissions were set at chmod 700 which we changed to 750. I detected this problem when I was able to compile index.cgi at root and not as the apache user. Sorry for the trouble and thanks for the support. regards, Geeta ----- Original Message ----- From: geeta To: developers at bugzilla.org Sent: Saturday, March 15, 2003 2:00 PM Subject: Need help with this one a bugging error Hello , When I run index.cgi from my command line my script runs ok. But when I run from the apache gives the following errors- I understand that this is a mysql shared library issue , I have done the work arounds. But, still getting the same errors. Another issue I have installled Msql-Mysql-modules-1.2209, but checksetup.pl is always giving found v 2.0409 instead of 1.2209. Now, is there any way to resolve this issue. Enclosing the error that I got- Uncaught exception from user code: install_driver(mysql) failed: Uncaught exception from user code: Uncaught exception from user code: Can't load '/usr/lib/perl5/site_perl/5.005/i386-linux/auto/DBD/mysql/mysql.so' for module DBD::mysql: libmysqlclient.so.10: cannot open shared object file: No such file or di rectory at /usr/lib/perl5/5.00503/i386-linux/DynaLoader.pm line 169. at (eval 4) line 3 Carp::croak('Can\'t load \'/usr/lib/perl5/site_perl/5.005/i386-linux/auto/DBD...') called at /usr/lib/perl5/5.00503/i386-linux/DynaLoader.pm line 93 DynaLoader::croak('Can\'t load \'/usr/lib/perl5/site_perl/5.005/i386-linux/auto/DBD ...') called at /usr/lib/perl5/5.00503/i386-linux/DynaLoader.pm line 169 DynaLoader::bootstrap('DBD::mysql', 2.0409) called at /usr/lib/perl5/site_perl/5.00 5/i386-linux/DBD/mysql.pm line 21 require DBD/mysql.pm called at (eval 4) line 3 eval 'package DBI::_firesafe; require DBD::mysql ;' called at /usr/lib/perl5/site_perl/5.005/i386-linux/DBI.pm line 478 DBI::install_driver('DBI', 'mysql') called at /usr/lib/perl5/site_perl/5.005/i386-l inux/DBI.pm line 402 DBI::connect('DBI', 'DBI:mysql:host=localhost;database=test;port=3306', 'test', 'te sttest') called at globals.pl line 140 main::ConnectToDatabase() called at index.cgi line 46 eval 'package DBI::_firesafe; require DBD::mysql ;' called at /usr/lib/perl5/site_perl/5.005/i386-linux/DBI.pm line 478 DBI::install_driver('DBI', 'mysql') called at /usr/lib/perl5/site_perl/5.005/i386-l inux/DBI.pm line 402 DBI::connect('DBI', 'DBI:mysql:host=localhost;database=test;port=3306', 'test', 'te sttest') called at globals.pl line 140 main::ConnectToDatabase() called at index.cgi line 46 Perhaps a required shared library or dll isn't installed where expected at globals.pl line 140 Carp::croak('install_driver(mysql) failed: Uncaught exception from user code:...') called at /usr/lib/perl5/site_perl/5.005/i386-linux/DBI.pm line 499 DBI::install_driver('DBI', 'mysql') called at /usr/lib/perl5/site_perl/5.005/i386-l inux/DBI.pm line 402 DBI::connect('DBI', 'DBI:mysql:host=localhost;database=test;port=3306', 'test', 'te sttest') called at globals.pl line 140 main::ConnectToDatabase() called at index.cgi line 46 Looking forward to hear from you. Thanks and regards, Geeta -------------- next part -------------- An HTML attachment was scrubbed... URL: From gerv at mozilla.org Sat Mar 15 21:02:25 2003 From: gerv at mozilla.org (Gervase Markham) Date: Sat, 15 Mar 2003 21:02:25 +0000 Subject: Lost bugmail Message-ID: <3E7394E1.7020209@mozilla.org> Anyone with an @mozilla.org address will have lost some bugmail over the past few days. I lost all mine since midnight GMT last night, but it may vary for other people. If you've said something to me in bugmail and I don't respond in the next few days, please issue a "ping?" and normal service will be resumed. This has been a public service announcement. Gerv From jamessmith449 at lycos.co.uk Sun Mar 16 21:48:58 2003 From: jamessmith449 at lycos.co.uk (James Smith) Date: Sun, 16 Mar 2003 16:48:58 -0500 Subject: Bugzilla Extensions References: <20030313102509.GF28150@software.plasmon> <002801c2e973$0b8e9640$1fec55c8@supercabletv.net.co> <3E718723.1060909@mozilla.org> Message-ID: <003401c2ec05$e788f7a0$1fec55c8@supercabletv.net.co> Hi It's not really bugzilla extensions, the kind of things I'm planning to develop are mobile "applications" using J2ME or other mobile technology. The idea I'm thinking about, consist in develop a mobile application that allows use some of the functionalities of bugzilla (i.e. notifications, consult bug status...). So, I would like to know if there is any experience or any projects, that, from an external application (probably running in other language i.e. java, C++...), communicates and/or uses some of the bugzilla functionalities. Thanks again Bye ----- Original Message ----- From: "Gervase Markham" To: Sent: Friday, March 14, 2003 2:39 AM Subject: Re: Bugzilla Extensions > James Smith wrote: > > I'm new to bugzilla as a developer. I wanna know if bugzilla has, for > > example, Corba IDL Interfaces to allow use bugzilla "api" from other > > languages like java. Or at least if 'bugzilla' has plans to do something > > similar > > We have no Corba IDL interfaces, no. Bugzilla is not a library, and so > does not have an API in that sense > > > I've serched all around the bugzilla site and couldn't found any reference > > about this matter > > > > I'm asking you this, 'cause I'm planning develop some extensions to > > bugzilla > > What sort of extensions? We'd be able to answer your question much > better if you were more specific > > Gerv > > ---- > To view or change your list settings, click here: > > > From jason at pyeron.com Sun Mar 16 22:49:48 2003 From: jason at pyeron.com (Jason Pyeron) Date: Sun, 16 Mar 2003 17:49:48 -0500 (EST) Subject: Bugzilla Extensions In-Reply-To: <003401c2ec05$e788f7a0$1fec55c8@supercabletv.net.co> Message-ID: James, Please review bug 188570, http://bugzilla.mozilla.org/show_bug.cgi?id=188570 Does this cover your issues? pardon the cc, if you would like to discuss by phone, feel free, or at least of the dev list. Sincerely, Jason Pyeron -- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- - - - Jason Pyeron http://www.pyerotechnics.com - - Owner & Lead Pyerotechnics Development, Inc. - - +1 410 808 6646 (c) 500 West University Parkway #1S - - +1 410 467 2266 (f) Baltimore, Maryland 21210-3253 - - - -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- This message is for the designated recipient only and may contain privileged, proprietary, or otherwise private information. If you have received it in error, purge the message from your system and notify the sender immediately. Any other use of the email by you is prohibited. On Sun, 16 Mar 2003, James Smith wrote: Hi It's not really bugzilla extensions, the kind of things I'm planning to develop are mobile "applications" using J2ME or other mobile technology. The idea I'm thinking about, consist in develop a mobile application that allows use some of the functionalities of bugzilla (i.e. notifications, consult bug status...). So, I would like to know if there is any experience or any projects, that, from an external application (probably running in other language i.e. java, C++...), communicates and/or uses some of the bugzilla functionalities. Thanks again Bye ----- Original Message ----- From: "Gervase Markham" To: Sent: Friday, March 14, 2003 2:39 AM Subject: Re: Bugzilla Extensions > James Smith wrote: > > I'm new to bugzilla as a developer. I wanna know if bugzilla has, for > > example, Corba IDL Interfaces to allow use bugzilla "api" from other > > languages like java. Or at least if 'bugzilla' has plans to do something > > similar > > We have no Corba IDL interfaces, no. Bugzilla is not a library, and so > does not have an API in that sense > > > I've serched all around the bugzilla site and couldn't found any reference > > about this matter > > > > I'm asking you this, 'cause I'm planning develop some extensions to > > bugzilla > > What sort of extensions? We'd be able to answer your question much > better if you were more specific > > Gerv > > ---- > To view or change your list settings, click here: > > > ---- To view or change your list settings, click here: From caseyg at chsamerica.com Mon Mar 17 15:11:01 2003 From: caseyg at chsamerica.com (Casey Gregoire) Date: Mon, 17 Mar 2003 10:11:01 -0500 Subject: custom fields patch Message-ID: Will the current custom fields patch work with Bugzilla 2.17? it says its for 2.16 but I don't know enough about the code to know whether or not it will work. Thank you Casey Gregoire From jphillips at amphus.com Mon Mar 17 20:10:35 2003 From: jphillips at amphus.com (Joseph Phillips) Date: 17 Mar 2003 12:10:35 -0800 Subject: CVS update, reset or preserve bugzilla Message-ID: <1047931835.24003.12.camel@hawk> I tried to send this to the mozilla-webtools list, but it is rejected since I'm not on the whitelist. The rejection email instructs me to send mail to the mozilla.org postmaster in order to be added to the list. However, this email address is not included. Would someone kindly send me this email? Thanks. Below is the message I am seeking an answer on, in case anyone on this list would be kind enough to respond. Thank you. I have an existing installation of bugzilla, version 2.14.1. I am interested to run the "Update Existing Install" procedure as described on the http://www.bugzilla.org/download.html webpage. The procedure basically involves logging into the bugzilla CVS server and running the command "$ cvs update -rBugzilla_Stable -dP". I want to know if this will reset all customization that I have made on the cgi/html files on this bugzilla installation to the default version from the public CVS server? From etzwane at schwag.org Mon Mar 17 23:24:53 2003 From: etzwane at schwag.org (Sean McAfee) Date: Mon, 17 Mar 2003 18:24:53 -0500 Subject: An alternate approach to custom fields In-Reply-To: <20030315085943.GA31727@mango.home> Message-ID: <20030317232453.CA162BCAD@diggity.schwag.org> Bradley Baetz wrote: >On Fri, Mar 14, 2003 at 07:10:42PM -0500, Sean McAfee wrote: >> For example, one of our workflows describes the testing path of individual >> returned parts. Each part is run through a dozen or more tests. The result >> of each test is recorded via a drop-down list on our current system's web >> interface. Different tests are performed at different times. Trying to >> squeeze all of this information into a single free-form text field and >> depending on users to keep the field formatted in a way that's suitable >> for querying seems like a recipe for disaster. >If the tests are just a pass/fail basis, then you could use keywords or >(in 2.17) bug flags to represent the various tests. If you sometimes run >tests more than once, then you'll have to use bug flags. The keyword approach doesn't seem appropriate. As I understand it, that would involve creating a trio of keywords (pass/fail/unknown) for each possible test, right? But then there would be no way to prevent these keywords from being applied to inappropriate incidents, or to enforce exclusivity. I didn't know about bug flags, since I've been evaluating 2.16.2. But I don't see any mention of them in the 2.17.3 documentation. How do they work? >> In a nutshell, the plan is this: > >I suspect that the effort involved in doing that (especially if you want >to search) is probably more than the effort required to get custom >fields done properly. [later] >The issue with custom fields is not that we don't know how to do it, but >that none of the developers have time to do it. Of course, ideally I'd like to have an official, proper custom fields implementation rather than hacking my own. But I have a time limit I'm up against; the license for our existing system expires later this year--sometime this summer, IIRC. The custom fields patch has been around and unfinalized for over a year a half already, with no end in sight, so I clearly can't depend on it being ready in time. The advantage to my approach, as I see it, is that it would be largely independent of the evolution of the Bugzilla codebase. Alternately, I could jump in and try to help get custom fields done the "right" way. My boss has even suggested that it might be possible to use some of my at-work time for this. Is it reasonably possible that a new developer could bring himself up to speed and drive the completion of an official custom fields patch in, say, a few months? -- Sean McAfee etzwane at schwag.org From gerv at mozilla.org Tue Mar 18 08:31:06 2003 From: gerv at mozilla.org (Gervase Markham) Date: Tue, 18 Mar 2003 08:31:06 +0000 Subject: CVS update, reset or preserve bugzilla In-Reply-To: <1047931835.24003.12.camel@hawk> References: <1047931835.24003.12.camel@hawk> Message-ID: <3E76D94A.3040604@mozilla.org> Joseph Phillips wrote: > The procedure basically involves logging into the bugzilla CVS server > and running the command "$ cvs update -rBugzilla_Stable -dP". > > I want to know if this will reset all customization that I have made on > the cgi/html files on this bugzilla installation to the default version > from the public CVS server? CVS will do its best to merge the updates with your changes, but if it doesn't succeed, you will have to integrate the updates with your changes manually. I suggest you take a copy of your Bugzilla tree, run the update on that, and fix the conflicts at your leisure. You can then copy the code back over when upgrade time comes (and run checksetup.pl.) Gerv From gerv at mozilla.org Tue Mar 18 08:31:42 2003 From: gerv at mozilla.org (Gervase Markham) Date: Tue, 18 Mar 2003 08:31:42 +0000 Subject: custom fields patch In-Reply-To: References: Message-ID: <3E76D96E.3000301@mozilla.org> Casey Gregoire wrote: > Will the current custom fields patch work with Bugzilla 2.17? it says its > for 2.16 but I don't know enough about the code to know whether or not it > will work. The bug would have been a better place to ask. I'd say it's very unlikely... but why not try it? Gerv From gerv at mozilla.org Tue Mar 18 08:34:13 2003 From: gerv at mozilla.org (Gervase Markham) Date: Tue, 18 Mar 2003 08:34:13 +0000 Subject: Bugzilla Extensions In-Reply-To: <003401c2ec05$e788f7a0$1fec55c8@supercabletv.net.co> References: <20030313102509.GF28150@software.plasmon> <002801c2e973$0b8e9640$1fec55c8@supercabletv.net.co> <3E718723.1060909@mozilla.org> <003401c2ec05$e788f7a0$1fec55c8@supercabletv.net.co> Message-ID: <3E76DA05.4090603@mozilla.org> James Smith wrote: > Hi > > It's not really bugzilla extensions, the kind of things I'm planning to > develop are mobile "applications" using J2ME or other mobile technology. You mean "mobile" as in "running on mobile hardware (PDAs)", as opposed to "mobile code"? > The idea I'm thinking about, consist in develop a mobile application that > allows use some of the functionalities of bugzilla (i.e. notifications, > consult bug status...). > > So, I would like to know if there is any experience or any projects, that, > from an external application (probably running in other language i.e. java, > C++...), communicates and/or uses some of the bugzilla functionalities. The best way to do this is to use a WWW library of some sort to make requests of the Bugzilla server, and receive the results back as HTML (for display) or RDF/XML (for further processing.) The obvious way to plug into notifications is to have the software check for and display the bugmail that gets sent. Gerv From jason at pyeron.com Tue Mar 18 12:12:44 2003 From: jason at pyeron.com (Jason Pyeron) Date: Tue, 18 Mar 2003 07:12:44 -0500 (EST) Subject: Bugzilla Extensions In-Reply-To: <3E76DA05.4090603@mozilla.org> Message-ID: Is this a reoccurring discussion every few months? On Tue, 18 Mar 2003, Gervase Markham wrote: James Smith wrote: > Hi > > It's not really bugzilla extensions, the kind of things I'm planning to > develop are mobile "applications" using J2ME or other mobile technology. You mean "mobile" as in "running on mobile hardware (PDAs)", as opposed to "mobile code"? > The idea I'm thinking about, consist in develop a mobile application that > allows use some of the functionalities of bugzilla (i.e. notifications, > consult bug status...). > > So, I would like to know if there is any experience or any projects, that, > from an external application (probably running in other language i.e. java, > C++...), communicates and/or uses some of the bugzilla functionalities. The best way to do this is to use a WWW library of some sort to make requests of the Bugzilla server, and receive the results back as HTML (for display) or RDF/XML (for further processing.) The obvious way to plug into notifications is to have the software check for and display the bugmail that gets sent. Gerv ---- To view or change your list settings, click here: -- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- - - - Jason Pyeron http://www.pyerotechnics.com - - Owner & Lead Pyerotechnics Development, Inc. - - +1 410 808 6646 (c) 500 West University Parkway #1S - - +1 410 467 2266 (f) Baltimore, Maryland 21210-3253 - - - -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- This message is for the designated recipient only and may contain privileged, proprietary, or otherwise private information. If you have received it in error, purge the message from your system and notify the sender immediately. Any other use of the email by you is prohibited. From justdave at syndicomm.com Tue Mar 18 15:13:15 2003 From: justdave at syndicomm.com (David Miller) Date: Tue, 18 Mar 2003 10:13:15 -0500 Subject: An alternate approach to custom fields In-Reply-To: <20030317232453.CA162BCAD@diggity.schwag.org> References: <20030317232453.CA162BCAD@diggity.schwag.org> Message-ID: On 3/17/03 6:24 PM -0500, Sean McAfee wrote: > Alternately, I could jump in and try to help get custom fields done the > "right" way. My boss has even suggested that it might be possible to use > some of my at-work time for this. Is it reasonably possible that a new > developer could bring himself up to speed and drive the completion of an > official custom fields patch in, say, a few months? I would say that's certainly possible, and we'd certainly welcome the help. :) -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From etzwane at schwag.org Tue Mar 18 21:07:52 2003 From: etzwane at schwag.org (Sean McAfee) Date: Tue, 18 Mar 2003 16:07:52 -0500 Subject: An alternate approach to custom fields In-Reply-To: Message-ID: <20030318210752.EC9ADBCFA@diggity.schwag.org> David Miller wrote: >On 3/17/03 6:24 PM -0500, Sean McAfee wrote: >> Alternately, I could jump in and try to help get custom fields done the >> "right" way. My boss has even suggested that it might be possible to use >> some of my at-work time for this. Is it reasonably possible that a new >> developer could bring himself up to speed and drive the completion of an >> official custom fields patch in, say, a few months? >I would say that's certainly possible, and we'd certainly welcome the help. :) Ah, excellent! I'll probably be taking that route, then. Is the proper place for discussing the details here, on this list, or via the comments section of 91307, or elsewhere? -- Sean McAfee etzwane at schwag.org From justdave at syndicomm.com Tue Mar 18 20:49:06 2003 From: justdave at syndicomm.com (David Miller) Date: Tue, 18 Mar 2003 15:49:06 -0500 Subject: An alternate approach to custom fields In-Reply-To: <20030318210752.EC9ADBCFA@diggity.schwag.org> References: <20030318210752.EC9ADBCFA@diggity.schwag.org> Message-ID: On 3/18/03 4:07 PM -0500, Sean McAfee wrote: > David Miller wrote: >>On 3/17/03 6:24 PM -0500, Sean McAfee wrote: >>> Alternately, I could jump in and try to help get custom fields done the >>> "right" way. My boss has even suggested that it might be possible to use >>> some of my at-work time for this. Is it reasonably possible that a new >>> developer could bring himself up to speed and drive the completion of an >>> official custom fields patch in, say, a few months? > >>I would say that's certainly possible, and we'd certainly welcome the >>help. :) > > Ah, excellent! I'll probably be taking that route, then. > > Is the proper place for discussing the details here, on this list, or via > the comments section of 91307, or elsewhere? Either place probably works. comments of that bug helps keep it together for latecomers so they don't have to go digging through the archives... -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From myk at mozilla.org Wed Mar 19 14:01:13 2003 From: myk at mozilla.org (Myk Melez) Date: Wed, 19 Mar 2003 06:01:13 -0800 Subject: An alternate approach to custom fields In-Reply-To: <20030317232453.CA162BCAD@diggity.schwag.org> References: <20030317232453.CA162BCAD@diggity.schwag.org> Message-ID: <3E787829.4020703@mozilla.org> Sean McAfee wrote: >Bradley Baetz wrote: > > >>On Fri, Mar 14, 2003 at 07:10:42PM -0500, Sean McAfee wrote: >> >> >>>For example, one of our workflows describes the testing path of individual >>>returned parts. Each part is run through a dozen or more tests. The result >>>of each test is recorded via a drop-down list on our current system's web >>>interface. Different tests are performed at different times. Trying to >>>squeeze all of this information into a single free-form text field and >>>depending on users to keep the field formatted in a way that's suitable >>>for querying seems like a recipe for disaster. >>> >>> > > > >>If the tests are just a pass/fail basis, then you could use keywords or >>(in 2.17) bug flags to represent the various tests. If you sometimes run >>tests more than once, then you'll have to use bug flags. >> >> > >The keyword approach doesn't seem appropriate. As I understand it, that >would involve creating a trio of keywords (pass/fail/unknown) for each >possible test, right? But then there would be no way to prevent these >keywords from being applied to inappropriate incidents, or to enforce >exclusivity. > >I didn't know about bug flags, since I've been evaluating 2.16.2. But I >don't see any mention of them in the 2.17.3 documentation. How do they >work? > > Bug flags are a way to record the status of some characteristic of a bug whose status can be defined in a boolean fashion. For example, bug fixes in the Bugzilla community must be approved by the module owner or his designee before they can be checked in. The Bugzilla community uses an "approved" flag in their own Bugzilla installation (bugzilla.mozilla.org) to indicate whether or not a bug has been approved for check in. A bug marked "approved+" has been approved for check in, while one marked "approved-" has been rejected. Bug flags are configurable and product/component-specific, so you can add as many flags as you want and restrict them to certain products or product/component combinations. The UI for setting flags on the bug report page consists of a table listing active flags and a pull-down menu showing their status. To change a flag's status, users select a different status from the pull-down menu. The options are "" (no value, a.k.a. flag not set), "+" (flag set positively), "-" (flag set negatively). Flags are similar to trios of keywords in the functionality they provide but have the following advantages over keywords; 1. They can be made product/component-specific. 2. They provide a structured UI for making changes instead of making the user type magical values into an arbitrary text field. 3. They can be deactivated/reactivated without deleting records in the database, so you can remove a flag from the UI while still keeping the data around in case it becomes useful again. Based on your description of the situation, I think bug flags are a good solution to your problem, with a few potential issues. First, you mention that you need a way to prevent the values from being applied to inappropriate incidents. Flags can be restricted by product/component combinations, so if your restrictions are all product/component-based, then this is not a problem. Otherwise you would have to add functionality to flags to restrict them by your other criteria. Second, flags do not enforce exclusivity, so this feature would have to be added. If you are interested, I can give you administrator access to a test installation on the Bugzilla testing server (http://landfill.bugzilla.org/) so you can test flags and see if they would work for what you want to do. -myk -------------- next part -------------- An HTML attachment was scrubbed... URL: From JWilmoth at starbucks.com Wed Mar 19 16:41:03 2003 From: JWilmoth at starbucks.com (Jon Wilmoth) Date: Wed, 19 Mar 2003 08:41:03 -0800 Subject: An alternate approach to custom fields Message-ID: I'm interested in custom fields and was under the impression the patch was "almost there"! While I think the flags are a good thing, my situation requires non-boolean values and disparate groups of information (i.e. what project the issue/bug is for AS WELL as what internal business unit(s) the issue impacts). My timeframe is relatively small for implementing this functionality, so I'd like to offer my assistance on finalizing the custom fields approach immediately. -----Original Message----- From: Myk Melez [mailto:myk at mozilla.org] Sent: Wednesday, March 19, 2003 6:01 AM To: Sean McAfee Cc: developers at bugzilla.org Subject: Re: An alternate approach to custom fields Sean McAfee wrote: Bradley Baetz wrote: On Fri, Mar 14, 2003 at 07:10:42PM -0500, Sean McAfee wrote: For example, one of our workflows describes the testing path of individual returned parts. Each part is run through a dozen or more tests. The result of each test is recorded via a drop-down list on our current system's web interface. Different tests are performed at different times. Trying to squeeze all of this information into a single free-form text field and depending on users to keep the field formatted in a way that's suitable for querying seems like a recipe for disaster. If the tests are just a pass/fail basis, then you could use keywords or (in 2.17) bug flags to represent the various tests. If you sometimes run tests more than once, then you'll have to use bug flags. The keyword approach doesn't seem appropriate. As I understand it, that would involve creating a trio of keywords (pass/fail/unknown) for each possible test, right? But then there would be no way to prevent these keywords from being applied to inappropriate incidents, or to enforce exclusivity. I didn't know about bug flags, since I've been evaluating 2.16.2. But I don't see any mention of them in the 2.17.3 documentation. How do they work? Bug flags are a way to record the status of some characteristic of a bug whose status can be defined in a boolean fashion.? For example, bug fixes in the Bugzilla community must be approved by the module owner or his designee before they can be checked in.? The Bugzilla community uses an "approved" flag in their own Bugzilla installation (bugzilla.mozilla.org) to indicate whether or not a bug has been approved for check in.? A bug marked "approved+" has been approved for check in, while one marked "approved-" has been rejected. Bug flags are configurable and product/component-specific, so you can add as many flags as you want and restrict them to certain products or product/component combinations.? The UI for setting flags on the bug report page consists of a table listing active flags and a pull-down menu showing their status.? To change a flag's status, users select a different status from the pull-down menu.? The options are "" (no value, a.k.a. flag not set), "+" (flag set positively), "-" (flag set negatively). Flags are similar to trios of keywords in the functionality they provide but have the following advantages over keywords; 1. They can be made product/component-specific. 2. They provide a structured UI for making changes instead of making the user type magical values into an arbitrary text field. 3. They can be deactivated/reactivated without deleting records in the database, so you can remove a flag from the UI while still keeping the data around in case it becomes useful again. Based on your description of the situation, I think bug flags are a good solution to your problem, with a few potential issues.? First, you mention that you need a way to prevent the values from being applied to inappropriate incidents.? Flags can be restricted by product/component combinations, so if your restrictions are all product/component-based, then this is not a problem.? Otherwise you would have to add functionality to flags to restrict them by your other criteria.? Second, flags do not enforce exclusivity, so this feature would have to be added. If you are interested, I can give you administrator access to a test installation on the Bugzilla testing server (http://landfill.bugzilla.org/) so you can test flags and see if they would work for what you want to do. -myk From usmlekaplan at yahoo.com Wed Mar 19 16:41:58 2003 From: usmlekaplan at yahoo.com (usmle Prep) Date: Wed, 19 Mar 2003 08:41:58 -0800 (PST) Subject: Can't use an undefined value as an ARRAY reference at /var/www/html/bugzilla/process_bug.cgi line 562 Message-ID: <20030319164158.99195.qmail@web20202.mail.yahoo.com> When trying to uncheck the CC list check box on the bugs form I am getting the following error.Did any one of you get it before. The line causing it is $cc_add=join(',',@{$::MFORM{newcc}}); When I looked on the web I found that perl is known to be behaving like this and it can be fixed if @{$::MFORM{newcc}} be assigned in 2 steps instead of one which I don't know how to do.Please let me know if you have fixed this before. Thanks a lot, Sri Bug processed Content-type: text/html Software error: Can't use an undefined value as an ARRAY reference at /var/www/html/bugzilla/process_bug.cgi line 562. __________________________________________________ Do you Yahoo!? Yahoo! Platinum - Watch CBS' NCAA March Madness, live on your desktop! http://platinum.yahoo.com From caseyg at chsamerica.com Thu Mar 20 15:01:42 2003 From: caseyg at chsamerica.com (Casey Gregoire) Date: Thu, 20 Mar 2003 10:01:42 -0500 Subject: Bug server database Message-ID: The other day I asked about the converting a database back to 2.16 from 2.17. I am wondering, why is that the database changed so much that you can not use the older version with the data? Also, would this mean the mechanism for transferring bugs from one database to another will not work across these two versions? I have not messed with that option much so I don't know much about it. Thank you, Casey Gregoire Programmer CHS of America 100 1st Ave S. Suite 601 St. Petersburg, FL 33701 Phone - (727) 824-0800 ext 1236 Every great achievement was once impossible. -- Anonymous I'd like to get started on time, if we can, inasmuch as we're late already. -- Larry Gelbart Nothing is accomplished without passion. -- My Fortune Cookie From usmlekaplan at yahoo.com Thu Mar 20 15:57:08 2003 From: usmlekaplan at yahoo.com (usmle Prep) Date: Thu, 20 Mar 2003 07:57:08 -0800 (PST) Subject: Resending: Can't use an undefined value as an ARRAY reference Message-ID: <20030320155708.40683.qmail@web20202.mail.yahoo.com> Hi, I am resending it.Hopefully some body might have seen it before. Thanks for u'r help, Sri When trying to uncheck the CC list check box on the bugs form I am getting the following error.Did any one of you get it before. The line causing it is $cc_add=join(',',@{$::MFORM{newcc}}); When I looked on the web I found that perl is known to be behaving like this and it can be fixed if @{$::MFORM{newcc}} be assigned in 2 steps instead of one which I don't know how to do.Please let me know if you have fixed this before. Thanks a lot, Sri Bug processed Content-type: text/html Software error: Can't use an undefined value as an ARRAY reference at /var/www/html/bugzilla/process_bug.cgi line 562. __________________________________________________ Do you Yahoo!? Yahoo! Platinum - Watch CBS' NCAA March Madness, live on your desktop! http://platinum.yahoo.com From justdave at syndicomm.com Thu Mar 20 16:12:47 2003 From: justdave at syndicomm.com (David Miller) Date: Thu, 20 Mar 2003 11:12:47 -0500 Subject: Bug server database In-Reply-To: References: Message-ID: On 3/20/03 10:01 AM -0500, Casey Gregoire wrote: > The other day I asked about the converting a database back to 2.16 from > 2.17. I am wondering, why is that the database changed so much that you can > not use the older version with the data? Also, would this mean the mechanism > for transferring bugs from one database to another will not work across > these two versions? I have not messed with that option much so I don't know > much about it. The externally visible field names remained the same in most places, just the way they're represented in the database is different. So the XML transfer should still work. In the case of new fields, whichever one doesn't recognize them will just list the unrecognized ones in the comment area. -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From caseyg at chsamerica.com Thu Mar 20 16:12:35 2003 From: caseyg at chsamerica.com (Casey Gregoire) Date: Thu, 20 Mar 2003 11:12:35 -0500 Subject: Bug server database Message-ID: Ok. SO I COULD use that to transfer my database from 2.17 back to 2.16 if I had to? I don't plan to at this time. But I needed to be able to answer the question when asked. Thanks, Casey Gregoire -----Original Message----- From: David Miller [mailto:justdave at syndicomm.com] Sent: Thursday, March 20, 2003 11:13 AM To: developers at bugzilla.org Subject: Re: Bug server database On 3/20/03 10:01 AM -0500, Casey Gregoire wrote: > The other day I asked about the converting a database back to 2.16 from > 2.17. I am wondering, why is that the database changed so much that you can > not use the older version with the data? Also, would this mean the mechanism > for transferring bugs from one database to another will not work across > these two versions? I have not messed with that option much so I don't know > much about it. The externally visible field names remained the same in most places, just the way they're represented in the database is different. So the XML transfer should still work. In the case of new fields, whichever one doesn't recognize them will just list the unrecognized ones in the comment area. -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ ---- To view or change your list settings, click here: From gerv at mozilla.org Thu Mar 20 17:45:04 2003 From: gerv at mozilla.org (Gervase Markham) Date: Thu, 20 Mar 2003 17:45:04 +0000 Subject: Another win for the good guys... Message-ID: <3E79FE20.9010600@mozilla.org> http://bugs.xfree86.org/ Gerv From jon at vmware.com Thu Mar 20 18:56:33 2003 From: jon at vmware.com (Jonathan Schatz) Date: 20 Mar 2003 10:56:33 -0800 Subject: Another win for the good guys... In-Reply-To: <3E79FE20.9010600@mozilla.org> References: <3E79FE20.9010600@mozilla.org> Message-ID: <1048186592.20523.5.camel@jonschatz-lx.vmware.com> On Thu, 2003-03-20 at 09:45, Gervase Markham wrote: > http://bugs.xfree86.org/ add VMware on the list as well. I went beta with our bugzilla server today, and (hopefully) within a week gnats will be gone. granted, this isn't a publicly accessible db (we have a siebel sr tracking system for external customer issues), but still. -jon -- Jonathan Schatz Engineering System Administrator VMware, Inc "Te occidere possunt sed te edere non possunt nefas est." From gerv at mozilla.org Fri Mar 21 00:00:23 2003 From: gerv at mozilla.org (Gervase Markham) Date: Fri, 21 Mar 2003 00:00:23 +0000 Subject: Another win for the good guys... In-Reply-To: <1048186592.20523.5.camel@jonschatz-lx.vmware.com> References: <3E79FE20.9010600@mozilla.org> <1048186592.20523.5.camel@jonschatz-lx.vmware.com> Message-ID: <3E7A5617.9070602@mozilla.org> Jonathan Schatz wrote: > On Thu, 2003-03-20 at 09:45, Gervase Markham wrote: > >>http://bugs.xfree86.org/ > > add VMware on the list as well. I went beta with our bugzilla server > today, and (hopefully) within a week gnats will be gone. We need a Hall Of Fame on the website. Gerv From kiko at async.com.br Fri Mar 21 00:48:05 2003 From: kiko at async.com.br (Christian Reis) Date: Thu, 20 Mar 2003 21:48:05 -0300 Subject: Another win for the good guys... In-Reply-To: <3E7A5617.9070602@mozilla.org>; from gerv@mozilla.org on Fri, Mar 21, 2003 at 12:00:23AM +0000 References: <3E79FE20.9010600@mozilla.org> <1048186592.20523.5.camel@jonschatz-lx.vmware.com> <3E7A5617.9070602@mozilla.org> Message-ID: <20030320214805.I15922@blackjesus.async.com.br> On Fri, Mar 21, 2003 at 12:00:23AM +0000, Gervase Markham wrote: > Jonathan Schatz wrote: > > On Thu, 2003-03-20 at 09:45, Gervase Markham wrote: > > > >>http://bugs.xfree86.org/ > > > > add VMware on the list as well. I went beta with our bugzilla server > > today, and (hopefully) within a week gnats will be gone. > > We need a Hall Of Fame on the website. Somebody explain to me why they are using an ancient version of Bugzilla (2.14) with no version information and the old and sluggish query Javascript? Take care, -- Christian Reis, Senior Engineer, Async Open Source, Brazil. http://async.com.br/~kiko/ | [+55 16] 261 2331 | NMFL From justdave at syndicomm.com Fri Mar 21 02:00:42 2003 From: justdave at syndicomm.com (David Miller) Date: Thu, 20 Mar 2003 21:00:42 -0500 Subject: Another win for the good guys... In-Reply-To: <20030320214805.I15922@blackjesus.async.com.br> References: <3E79FE20.9010600@mozilla.org> <1048186592.20523.5.camel@jonschatz-lx.vmware.com> <3E7A5617.9070602@mozilla.org> <20030320214805.I15922@blackjesus.async.com.br> Message-ID: On 3/20/03 9:48 PM -0300, Christian Reis wrote: > On Fri, Mar 21, 2003 at 12:00:23AM +0000, Gervase Markham wrote: >> Jonathan Schatz wrote: >> > On Thu, 2003-03-20 at 09:45, Gervase Markham wrote: >> > >> >>http://bugs.xfree86.org/ >> > >> > add VMware on the list as well. I went beta with our bugzilla server >> > today, and (hopefully) within a week gnats will be gone. >> >> We need a Hall Of Fame on the website. > > Somebody explain to me why they are using an ancient version of Bugzilla > (2.14) with no version information and the old and sluggish query > Javascript? Notice that it redirects into cgi-bin... That's the Debian version of Bugzilla, and 2.14.something is what's in Debian-stable right now. -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From etzwane at schwag.org Fri Mar 21 08:20:04 2003 From: etzwane at schwag.org (Sean McAfee) Date: Fri, 21 Mar 2003 03:20:04 -0500 Subject: An alternate approach to custom fields In-Reply-To: Message-ID: <20030321082004.02294C00B@diggity.schwag.org> "Jon Wilmoth" wrote: >I'm interested in custom fields and was under the impression the patch was >"almost there"! So was I, but other comments I've read suggest that the final implementation may be quite different from what currently exists, from which I gather that the existing implementation isn't considered satisfactory. >While I think the flags are a good thing, my situation >requires non-boolean values and disparate groups of information (i.e. what >project the issue/bug is for AS WELL as what internal business unit(s) >the issue impacts). My timeframe is relatively small for implementing >this functionality, so I'd like to offer my assistance on finalizing the >custom fields approach immediately. As for me, I need at least a few days to become more familiar with the code and schema. Then, I was planning to write up a requirements document and solicit comments...unless someone else takes the helm in the meantime, I suppose. -- Sean McAfee -- etzwane at schwag.org From gerv at mozilla.org Fri Mar 21 08:30:14 2003 From: gerv at mozilla.org (Gervase Markham) Date: Fri, 21 Mar 2003 08:30:14 +0000 Subject: An alternate approach to custom fields In-Reply-To: <20030321082004.02294C00B@diggity.schwag.org> References: <20030321082004.02294C00B@diggity.schwag.org> Message-ID: <3E7ACD96.1050305@mozilla.org> Sean McAfee wrote: > "Jon Wilmoth" wrote: > As for me, I need at least a few days to become more familiar with the code > and schema. Then, I was planning to write up a requirements document and > solicit comments...unless someone else takes the helm in the meantime, I > suppose. bbaetz outlined how we would like this to work in a message on the 1st of this month: >>A Google for 'database "mapping tables"' doesn't define the term. Do you >>> mean lots of tables like: > > No, I mean tables like: > > bug_id, field_id, value > > We will need several of those, however - one for string types, and one > for number types, so that indexes work optimally. For stuff like os's, > (field_id, value) is an FK [Foreign Key] to another table which has the > name<-> number mapping. > > (If we ever support email types that way, then wed need a third one, to > fk to the profiles table) > >>> with slight variations depending on whether it's a 1->1 or 1->many or >>> enum-style mapping? And doing joins everywhere to connect it all up? > > 1->1 vs 1->many is only how many entries there are. We may want > (another) separate table for that to add the UNIQUE constraint - I > haven't thought too much about that. Basically, you need to run your design past him. If he likes it, we like it :-) Here are some thoughts. Important criteria: - If you aren't using custom fields, performance should not be affected Possible field types: - Single string - Single value from enum (but can't use MySQL-specific ENUM type) - Single email address (optional, unless you need it yourself) - Multiple values from enum (optional) - Multiple email addresses (very optional) We need to discuss and find good UI for displaying the custom fields on the show_bug page, and querying for them. Gerv From chicks at chicks.net Fri Mar 21 17:15:32 2003 From: chicks at chicks.net (Christopher Hicks) Date: Fri, 21 Mar 2003 12:15:32 -0500 (EST) Subject: An alternate approach to custom fields In-Reply-To: <3E7ACD96.1050305@mozilla.org> Message-ID: On Fri, 21 Mar 2003, Gervase Markham wrote: > bbaetz outlined how we would like this to work in a message on the 1st > of this month: bbaetz' answer sounds good as far as storing the data, but he left out a table for storing this mapping information. Having the mapping information in the database has always been essential in my experience. If this information doesn't ride along with the data you have to maintain it as part of the code which would undermine the value of making the custom fields generalizations in the first place For instance: ID Name Type Multi FKID -- ---- ---- ----- ---- 1 dept fk 1 1 2 manager email 0 NULL 3 fixedon fk 1 2 4 serverroom text 0 NULL "ID" would be a uniqie identifier for each custom field. "Name" would be a human-readable unique identifier. Additionally, a presentable UI name would probably be beneficial, but I omitted that for brevity. "Type" would indicate what kind of thing it is. 'text', 'email', 'float' and 'int' have been mentioned and would be basic types each with their own table. 'FK' would be for foreign keys. "FKID" would identify the foreign key. Above I've used an integer to indicate this which would point to a portion of a codes table. bbaetz answer implied he'd prefer to have a seperate table for each foreign key, so having a table name here might be necessary for his approval. I'd encourage having a single table for such simple code lookups however. If you're not using relational integrity and you're not stuck in the Orthodox Relational perspective, it tends to lead to a lot less clutter and it makes creating an interface to let people maintain codes easier. "Multi" would indicate whether multiple instances of that field are allowed. -- The death of democracy is not likely to be an assassination from ambush. It will be a slow extinction from apathy, indifference, and undernourishment. -Robert Maynard Hutchins, educator (1899-1977) From caseyg at chsamerica.com Fri Mar 21 20:42:56 2003 From: caseyg at chsamerica.com (Casey Gregoire) Date: Fri, 21 Mar 2003 15:42:56 -0500 Subject: Lines really long in comment fields Message-ID: When you enter a comment in 2.17 with out manually hitting enter to go down to the next line the line will not word wrap at all when you view the bug. Is this a problem I should put into the bug db? Or is it meant to be this way? Thank you, Casey Gregoire Programmer CHS of America 100 1st Ave S. Suite 601 St. Petersburg, FL 33701 Phone - (727) 824-0800 ext 1236 Every great achievement was once impossible. -- Anonymous I'd like to get started on time, if we can, inasmuch as we're late already. -- Larry Gelbart Nothing is accomplished without passion. -- My Fortune Cookie From justdave at syndicomm.com Fri Mar 21 20:53:10 2003 From: justdave at syndicomm.com (David Miller) Date: Fri, 21 Mar 2003 15:53:10 -0500 Subject: Lines really long in comment fields In-Reply-To: References: Message-ID: On 3/21/03 3:42 PM -0500, Casey Gregoire wrote: > When you enter a comment in 2.17 with out manually hitting enter to go down > to the next line the line will not word wrap at all when you view the bug. > Is this a problem I should put into the bug db? Or is it meant to be this > way? It means your browser doesn't honor a "netscapism" on the TEXTAREA attribute which Netscape, Mozilla, and MSIE all honor. Fixing Bugzilla to work without depending on that non-standard browser feature is http://bugzilla.mozilla.org/show_bug.cgi?id=11901 -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From bbaetz at acm.org Fri Mar 21 22:18:25 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Sat, 22 Mar 2003 09:18:25 +1100 Subject: An alternate approach to custom fields In-Reply-To: References: <3E7ACD96.1050305@mozilla.org> Message-ID: <20030321221825.GA1234@mango.home> On Fri, Mar 21, 2003 at 12:15:32PM -0500, Christopher Hicks wrote: > On Fri, 21 Mar 2003, Gervase Markham wrote: > > bbaetz outlined how we would like this to work in a message on the 1st > > of this month: > > bbaetz' answer sounds good as far as storing the data, but he left out a > table for storing this mapping information. We already have a fielddefs table - we'd just have to extend it a bit > Having the mapping > information in the database has always been essential in my experience. > If this information doesn't ride along with the data you have to maintain > it as part of the code which would undermine the value of making the > custom fields generalizations in the first place > > For instance: > > "FKID" would identify the foreign key. Above I've used an integer to > indicate this which would point to a portion of a codes table. bbaetz > answer implied he'd prefer to have a seperate table for each foreign key, > so having a table name here might be necessary for his approval. I'd > encourage having a single table for such simple code lookups however. If > you're not using relational integrity and you're not stuck in the Orthodox > Relational perspective, it tends to lead to a lot less clutter and it > makes creating an interface to let people maintain codes easier. We will be using referential integrity. That said, we could probably enforce it via a trigger, except for mysql. > > "Multi" would indicate whether multiple instances of that field are > allowed. > That precludes a unique constraint when its not. Now, this would all be simpler if we coudl store it in multiple tables, and then have a VIEW with a UNION ALL query to bring them all together. But.... Bradley From caseyg at chsamerica.com Fri Mar 21 23:01:04 2003 From: caseyg at chsamerica.com (Casey Gregoire) Date: Fri, 21 Mar 2003 18:01:04 -0500 Subject: Lines really long in comment fields Message-ID: Which patch do I apply? All I get is errors when I try to apply it. Thanks, Casey Gregoire -----Original Message----- From: David Miller [mailto:justdave at syndicomm.com] Sent: Friday, March 21, 2003 3:53 PM To: developers at bugzilla.org Subject: Re: Lines really long in comment fields On 3/21/03 3:42 PM -0500, Casey Gregoire wrote: > When you enter a comment in 2.17 with out manually hitting enter to go down > to the next line the line will not word wrap at all when you view the bug. > Is this a problem I should put into the bug db? Or is it meant to be this > way? It means your browser doesn't honor a "netscapism" on the TEXTAREA attribute which Netscape, Mozilla, and MSIE all honor. Fixing Bugzilla to work without depending on that non-standard browser feature is http://bugzilla.mozilla.org/show_bug.cgi?id=11901 -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ ---- To view or change your list settings, click here: From JWilmoth at starbucks.com Sat Mar 22 00:59:05 2003 From: JWilmoth at starbucks.com (Jon Wilmoth) Date: Fri, 21 Mar 2003 16:59:05 -0800 Subject: An alternate approach to custom fields Message-ID: Can some one who has a solid understanding of the bugzilla inners workings describe the relationship between the fielddefs table, field-descs.html.tmpl and their effect on the ui? I've changed description values in the table and the field-descs with no affect on the bug entry, query, or query results page. I'd also like to see a merger of the 'customfields' (from patch) and the 'fielddefs' tables. I think it would be valuable for an admin perspective to make fields required/validated and change the label/description on a per install basis. This would also allow single "field" definition/maintenance paradigm. Furthermore, it will allow for quick resolution of a number of bugs that are waiting for this (i.e. "must have by" field, etc.) Thus something like (fielddefs): Name - human-readable unique identifier (pk). An numeric ID column wouldn't be necessary? Label - The human readable text presented to the user (Template would need to be refactored to apply this to existing fields). Type - ? is this necessary? What value does this provide if validation regexps are enforced against the content? constraint restricted to valid data types. Multi - flag indicating user may select multiple values SortKey - Display order Required - flag indicating a value must be submitted. Should be able to be overridden at lower (i.e. product specific levels). Valexp - regular expression describing the valid pattern of input. If null, no validation is performed. (also overrideable at lower level). System - flag indicating this field is part of the base distribution and must be present for the correct functioning of Bugzilla (i.e. a not deletable flag). -----Original Message----- From: Bradley Baetz [mailto:bbaetz at acm.org] Sent: Friday, March 21, 2003 2:18 PM To: developers at bugzilla.org Subject: Re: An alternate approach to custom fields On Fri, Mar 21, 2003 at 12:15:32PM -0500, Christopher Hicks wrote: > On Fri, 21 Mar 2003, Gervase Markham wrote: > > bbaetz outlined how we would like this to work in a message on the 1st > > of this month: > > bbaetz' answer sounds good as far as storing the data, but he left out a > table for storing this mapping information. We already have a fielddefs table - we'd just have to extend it a bit > Having the mapping > information in the database has always been essential in my experience. > If this information doesn't ride along with the data you have to maintain > it as part of the code which would undermine the value of making the > custom fields generalizations in the first place > > For instance: > > "FKID" would identify the foreign key. Above I've used an integer to > indicate this which would point to a portion of a codes table. bbaetz > answer implied he'd prefer to have a seperate table for each foreign key, > so having a table name here might be necessary for his approval. I'd > encourage having a single table for such simple code lookups however. If > you're not using relational integrity and you're not stuck in the Orthodox > Relational perspective, it tends to lead to a lot less clutter and it > makes creating an interface to let people maintain codes easier. We will be using referential integrity. That said, we could probably enforce it via a trigger, except for mysql. > > "Multi" would indicate whether multiple instances of that field are > allowed. > That precludes a unique constraint when its not. Now, this would all be simpler if we coudl store it in multiple tables, and then have a VIEW with a UNION ALL query to bring them all together. But.... Bradley ---- To view or change your list settings, click here: From gerv at mozilla.org Sat Mar 22 10:03:04 2003 From: gerv at mozilla.org (Gervase Markham) Date: Sat, 22 Mar 2003 10:03:04 +0000 Subject: An alternate approach to custom fields In-Reply-To: References: Message-ID: <3E7C34D8.80903@mozilla.org> Jon Wilmoth wrote: > Can some one who has a solid understanding of the bugzilla inners > workings describe the relationship between the fielddefs table, > field-descs.html.tmpl and their effect on the ui? I've changed > description values in the table and the field-descs with no affect on > the bug entry, query, or query results page. The description field is deprecated; it can't live in there for localisation reasons. Fields should be uniquely identified by a short string consisting of letters and underscores only, which can be used as a form field name, database column name etc. > I'd also like to see a merger of the 'customfields' (from patch) and the > 'fielddefs' tables. Please don't think you need to base your work on the existing patch, or necessarily use any of the ideas or schemas contained therein. > Thus something like (fielddefs): > Name - human-readable unique identifier (pk). An numeric ID column > wouldn't be necessary? Always have a numerical ID. :-) > Label - The human readable text presented to the user (Template would > need to be refactored to apply this to existing fields). This should not be in the table - see above, and the field-descs.html.tmpl template in Bugzilla. > Type - ? is this necessary? What value does this provide if validation > regexps are enforced against the content? constraint restricted to valid > data types. We need a type, because we want to store strings as strings and integers as integers, for speed reasons. > Multi - flag indicating user may select multiple values > > SortKey - Display order Where these have appeared elsewhere in Bugzilla, they have been a pain. Why is it needed? > Required - flag indicating a value must be submitted. Should be able to > be overridden at lower (i.e. product specific levels). You mean on bug entry? We don't need this in the table - policy such as this is enforced at a much higher level. > Valexp - regular expression describing the valid pattern of input. If > null, no validation is performed. (also overrideable at lower level). Again, I'm not convinced we need this (other than the definition of enum types.) > System - flag indicating this field is part of the base distribution and > must be present for the correct functioning of Bugzilla (i.e. a not > deletable flag). Surely the point of custom fields is that they are custom? Or is this a way to distinguish those fields currently in the fielddefs table from the new ones? Gerv From gerv at mozilla.org Sat Mar 22 10:20:06 2003 From: gerv at mozilla.org (Gervase Markham) Date: Sat, 22 Mar 2003 10:20:06 +0000 Subject: Hall of Fame Message-ID: <3E7C38D6.90508@mozilla.org> Here's a quick draft before I go for the weekend. Any important projects or companies I've missed? Gerv Free Software Projects: Linux Kernel: http://bugzilla.kernel.org/ Gnome: http://bugzilla.gnome.org/ KDE: http://bugs.kde.org/ XFree86: http://bugs.xfree86.org/ Mozilla: http://bugzilla.mozilla.org Open Office: http://www.openoffice.org/issues/query.cgi Apache Project: http://nagoya.apache.org/bugzilla/ Wine: http://bugs.winehq.com/ AbiWord: http://bugzilla.abisource.com/ Squid: http://www.squid-cache.org/bugs/ Lyx: http://bugzilla.lyx.org/ ProFTPD: http://bugs.proftpd.org/ xiph.org: http://bugs.xiph.org/ Spam Assassin: http://bugzilla.spamassassin.org/ Darwin: http://www.opendarwin.org/bugzilla/ OpenSSH: http://bugzilla.mindrot.org/ (and many more smaller ones) Linux Distributions: Red Hat: http://bugzilla.redhat.com/bugzilla/ Mandrake: https://qa.mandrakesoft.com/cgi-bin/index.cgi Gentoo: http://bugs.gentoo.org/ TurboLinux: http://the.turbolinux.co.jp/bugzilla/ Conectiva: http://distro.conectiva.com/bugzilla/ Linux From Scratch: http://bugs.linuxfromscratch.org/ Blue Linux: http://bugzilla.bluelinux.org/ SOT Linux: https://www.sotlinux.org/bugzilla/ Trustix: http://www.trustix.net/bugs/ SourceMage: http://bugs.sourcemage.org/ Companies: Ximian: http://bugzilla.ximian.com/ NASA: http://itos.gsfc.nasa.gov/~bugzilla/ SGI: http://oss.sgi.com/bugzilla/ Id Software: http://zerowing.idsoftware.com/bugzilla/ IBM: http://www-124.ibm.com/developerworks/oss/linux/bugzilla/ From bbaetz at acm.org Sat Mar 22 10:30:03 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Sat, 22 Mar 2003 21:30:03 +1100 Subject: Hall of Fame In-Reply-To: <3E7C38D6.90508@mozilla.org> References: <3E7C38D6.90508@mozilla.org> Message-ID: <20030322103003.GA27106@mango.home> GCC is moving to bugzilla 2.17. RSN, I believe Bradley From bugreport at peshkin.net Sat Mar 22 15:49:11 2003 From: bugreport at peshkin.net (Joel Peshkin) Date: Sat, 22 Mar 2003 07:49:11 -0800 Subject: Hall of Fame In-Reply-To: <3E7C38D6.90508@mozilla.org> References: <3E7C38D6.90508@mozilla.org> Message-ID: <3E7C85F7.7070403@peshkin.net> I will seek permission from my employer to put their name on the list. -Joel From dberlin at dberlin.org Sat Mar 22 15:53:45 2003 From: dberlin at dberlin.org (Daniel Berlin) Date: Sat, 22 Mar 2003 10:53:45 -0500 Subject: Hall of Fame In-Reply-To: <20030322103003.GA27106@mango.home> Message-ID: <7660DB2B-5C7E-11D7-B64A-000393575BCC@dberlin.org> On Saturday, March 22, 2003, at 05:30 AM, Bradley Baetz wrote: > GCC is moving to bugzilla 2.17. 3, ATM, plus a few local hacks (platform and os are useless for us. We have a target, host, and build field. This is because GCC can be built, for example, on an i686-pc-linux-gnu to run on a powerpc-apple-darwin6.5 machine targeting arm-elf. We end up with bugs specific to targets, hosts, hosts x target, etc.) > RSN, I believe > Yes. I just need to finsih some doc patches, and then we'll change over. > Bradley > ---- > To view or change your list settings, click here: > From justdave at syndicomm.com Sat Mar 22 16:05:40 2003 From: justdave at syndicomm.com (David Miller) Date: Sat, 22 Mar 2003 11:05:40 -0500 Subject: Lines really long in comment fields In-Reply-To: References: Message-ID: On 3/21/03 6:01 PM -0500, Casey Gregoire wrote: > Which patch do I apply? All I get is errors when I try to apply it. I don't think there's a working one on there yet. There were a few prototypes done, but it's been a long time and the code has changed a lot. (I think the existing patches are 2.14 era) -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From justdave at syndicomm.com Sat Mar 22 16:32:13 2003 From: justdave at syndicomm.com (David Miller) Date: Sat, 22 Mar 2003 11:32:13 -0500 Subject: An alternate approach to custom fields In-Reply-To: <20030321221825.GA1234@mango.home> References: <3E7ACD96.1050305@mozilla.org> <20030321221825.GA1234@mango.home> Message-ID: On 3/22/03 9:18 AM +1100, Bradley Baetz wrote: >> "Multi" would indicate whether multiple instances of that field are >> allowed. >> > > That precludes a unique constraint when its not. > > Now, this would all be simpler if we coudl store it in multiple tables, > and then have a VIEW with a UNION ALL query to bring them all together. MySQL 4 supports that. :) -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From justdave at syndicomm.com Sat Mar 22 16:32:01 2003 From: justdave at syndicomm.com (David Miller) Date: Sat, 22 Mar 2003 11:32:01 -0500 Subject: An alternate approach to custom fields In-Reply-To: <3E7C34D8.80903@mozilla.org> References: <3E7C34D8.80903@mozilla.org> Message-ID: On 3/22/03 10:03 AM +0000, Gervase Markham wrote: > Jon Wilmoth wrote: > >> SortKey - Display order > > Where these have appeared elsewhere in Bugzilla, they have been a pain. > Why is it needed? fielddefs already has sort keys. It's used to define the order the fields are listed in when sending bugmail for new bugs. That's not something new he's adding. >> Valexp - regular expression describing the valid pattern of input. If >> null, no validation is performed. (also overrideable at lower level). > > Again, I'm not convinced we need this (other than the definition of enum > types.) I can see the use for this in general for strings.... say you put in a free-text version number field because you have too many versions to list in a popup. Your version numbers are in a specific format that you can always recognize though. You could use this to ensure anything entered in that field is actually a legal version number. >> System - flag indicating this field is part of the base distribution and >> must be present for the correct functioning of Bugzilla (i.e. a not >> deletable flag). > > Surely the point of custom fields is that they are custom? Or is this a > way to distinguish those fields currently in the fielddefs table from > the new ones? I think that's what he meant, since it was suggested to extend the existing fielddefs table for that purpose. -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From aponnampalam at sapient.com Sun Mar 23 13:01:18 2003 From: aponnampalam at sapient.com (Anandha Ponnampalam) Date: Sun, 23 Mar 2003 13:01:18 -0000 Subject: Upgrading from v2.14.x to v2.17.x Message-ID: <4207EEFD29960040BAFE87E3D5C1DAB617D71D@lonbmsx01.sapient.com> I've found some documentation on how to upgrade the code base (using cvs, tarball or patch) from one version to next. How do i upgrade the database from v2.14.x to v2.17.x. Have there been any schema changes? Is there some mechanism for rollback? Is there a recommended version of mySQL? thanks anandha -------------- next part -------------- An HTML attachment was scrubbed... URL: From bbaetz at acm.org Sun Mar 23 20:10:18 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Mon, 24 Mar 2003 07:10:18 +1100 Subject: An alternate approach to custom fields In-Reply-To: References: <3E7ACD96.1050305@mozilla.org> <20030321221825.GA1234@mango.home> Message-ID: <20030323201016.GA1241@mango.home> On Sat, Mar 22, 2003 at 11:32:13AM -0500, David Miller wrote: > > MySQL 4 supports that. :) Not the view part. And when 5.0 does, I'd be surpised if predicates were pushed down in the first release. Bradley From gerv at mozilla.org Mon Mar 24 07:35:24 2003 From: gerv at mozilla.org (Gervase Markham) Date: Mon, 24 Mar 2003 07:35:24 +0000 Subject: Upgrading from v2.14.x to v2.17.x In-Reply-To: <4207EEFD29960040BAFE87E3D5C1DAB617D71D@lonbmsx01.sapient.com> References: <4207EEFD29960040BAFE87E3D5C1DAB617D71D@lonbmsx01.sapient.com> Message-ID: <3E7EB53C.8080806@mozilla.org> Anandha Ponnampalam wrote: > Upgrading from v2.14.x to v2.17.x > > I've found some documentation on how to upgrade the code base (using > cvs, tarball or patch) from one version to next. > > How do i upgrade the database from v2.14.x to v2.17.x. Have there been > any schema changes? Is there some mechanism for rollback? To upgrade, you run "checksetup.pl". There are lots of schema changes. There is no rollback mechanism - back up your database. :-) Gerv From dswegen at software.plasmon.com Mon Mar 24 12:45:31 2003 From: dswegen at software.plasmon.com (Dave Swegen) Date: Mon, 24 Mar 2003 12:45:31 +0000 Subject: Licensing question Message-ID: <20030324124531.GH9460@software.plasmon> I'm in the process of cleaning (relatively speaking ;) up our CVS integration code, and it adds a script to bz. So the required MPL boilerplate has been tacked on. My understanding is that the 'Initial Developer' section should be set to my company, rather than Netscape. Is this a correct assumption? Cheers Dave From gerv at mozilla.org Mon Mar 24 13:13:07 2003 From: gerv at mozilla.org (Gervase Markham) Date: Mon, 24 Mar 2003 13:13:07 +0000 Subject: Licensing question In-Reply-To: <20030324124531.GH9460@software.plasmon> References: <20030324124531.GH9460@software.plasmon> Message-ID: <3E7F0463.7000709@mozilla.org> Dave Swegen wrote: > I'm in the process of cleaning (relatively speaking ;) up our CVS > integration code, and it adds a script to bz. So the required MPL > boilerplate has been tacked on. > > My understanding is that the 'Initial Developer' section should be set > to my company, rather than Netscape. Is this a correct assumption? If the script is entirely your work, rather than based on an existing Bugzilla file, then yes. Gerv From dswegen at software.plasmon.com Mon Mar 24 13:19:21 2003 From: dswegen at software.plasmon.com (Dave Swegen) Date: Mon, 24 Mar 2003 13:19:21 +0000 Subject: Licensing question In-Reply-To: <3E7F0463.7000709@mozilla.org> References: <20030324124531.GH9460@software.plasmon> <3E7F0463.7000709@mozilla.org> Message-ID: <20030324131921.GI9460@software.plasmon> On Mon, Mar 24, 2003 at 01:13:07PM +0000, Gervase Markham wrote: > Dave Swegen wrote: > >I'm in the process of cleaning (relatively speaking ;) up our CVS > >integration code, and it adds a script to bz. So the required MPL > >boilerplate has been tacked on. > > > >My understanding is that the 'Initial Developer' section should be set > >to my company, rather than Netscape. Is this a correct assumption? > > If the script is entirely your work, rather than based on an existing > Bugzilla file, then yes. > > Gerv That's what I thought. Cheers. Dave From caseyg at chsamerica.com Mon Mar 24 14:05:27 2003 From: caseyg at chsamerica.com (Casey Gregoire) Date: Mon, 24 Mar 2003 09:05:27 -0500 Subject: Lines really long in comment fields Message-ID: Is this a problem that comes from when you enter the bug or just when it is displayed? Thanks, Casey Gregoire -----Original Message----- From: David Miller [mailto:justdave at syndicomm.com] Sent: Saturday, March 22, 2003 11:06 AM To: developers at bugzilla.org Subject: Re: Lines really long in comment fields On 3/21/03 6:01 PM -0500, Casey Gregoire wrote: > Which patch do I apply? All I get is errors when I try to apply it. I don't think there's a working one on there yet. There were a few prototypes done, but it's been a long time and the code has changed a lot. (I think the existing patches are 2.14 era) -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ ---- To view or change your list settings, click here: From caseyg at chsamerica.com Mon Mar 24 14:15:25 2003 From: caseyg at chsamerica.com (Casey Gregoire) Date: Mon, 24 Mar 2003 09:15:25 -0500 Subject: Hall of Fame Message-ID: Do you only want links directly to the company Bugzilla page? We may not be publicly giving access to Bugzilla at this time. I am still seeking permission from my employer as well for this. Thanks, Casey Gregoire -----Original Message----- From: Daniel Berlin [mailto:dberlin at dberlin.org] Sent: Saturday, March 22, 2003 10:54 AM To: developers at bugzilla.org Subject: Re: Hall of Fame On Saturday, March 22, 2003, at 05:30 AM, Bradley Baetz wrote: > GCC is moving to bugzilla 2.17. 3, ATM, plus a few local hacks (platform and os are useless for us. We have a target, host, and build field. This is because GCC can be built, for example, on an i686-pc-linux-gnu to run on a powerpc-apple-darwin6.5 machine targeting arm-elf. We end up with bugs specific to targets, hosts, hosts x target, etc.) > RSN, I believe > Yes. I just need to finsih some doc patches, and then we'll change over. > Bradley > ---- > To view or change your list settings, click here: > ---- To view or change your list settings, click here: From justdave at syndicomm.com Mon Mar 24 14:22:38 2003 From: justdave at syndicomm.com (David Miller) Date: Mon, 24 Mar 2003 09:22:38 -0500 Subject: Lines really long in comment fields In-Reply-To: References: Message-ID: Right now it's a problem when you enter the bug. It *should* be a problem when the bug is displayed. The intended fix is to get rid of the hard-wrap attributes on the text boxes and either do display-time wrapping when the comments are displayed to the user, or hard-wrap the text server-side after it's uploaded instead of depending on the browser to do it like we do currently. My personal vote is to do it display time, as it makes things more flexible and also automatically fixes all the old, currently-broken comments. On 3/24/03 9:05 AM -0500, Casey Gregoire wrote: > Is this a problem that comes from when you enter the bug or just when it is > displayed? > > Thanks, > Casey Gregoire > > -----Original Message----- > From: David Miller [mailto:justdave at syndicomm.com] > Sent: Saturday, March 22, 2003 11:06 AM > To: developers at bugzilla.org > Subject: Re: Lines really long in comment fields > > On 3/21/03 6:01 PM -0500, Casey Gregoire wrote: > >> Which patch do I apply? All I get is errors when I try to apply it. > > I don't think there's a working one on there yet. There were a few > prototypes done, but it's been a long time and the code has changed a lot. > (I think the existing patches are 2.14 era) -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From gerv at mozilla.org Mon Mar 24 14:35:59 2003 From: gerv at mozilla.org (Gervase Markham) Date: Mon, 24 Mar 2003 14:35:59 +0000 Subject: Hall of Fame In-Reply-To: References: Message-ID: <3E7F17CF.7020605@mozilla.org> Casey Gregoire wrote: > Do you only want links directly to the company Bugzilla page? We may not be > publicly giving access to Bugzilla at this time. I am still seeking > permission from my employer as well for this. Given that we cannot list every Bugzilla or Bugzilla-using company out there, and that this is a marketing tool, I propose the following criteria for HOF entry: Either: - Public Bugzilla we can link to (public as in "you can search it") - High-profile open source project or well-known company Or: - Private Bugzilla - Extremely well-known company or company with kudos attached (e.g. NASA) The latter would be a pretty short list, I think. I suppose we could have a top section with logos and links, and a bottom section which is just a list of company/project names. Gerv From caseyg at chsamerica.com Mon Mar 24 14:33:04 2003 From: caseyg at chsamerica.com (Casey Gregoire) Date: Mon, 24 Mar 2003 09:33:04 -0500 Subject: Lines really long in comment fields Message-ID: Is there a script to manually fix the fields that are wrong on the database? Thanks, Casey Gregoire -----Original Message----- From: David Miller [mailto:justdave at syndicomm.com] Sent: Monday, March 24, 2003 9:23 AM To: developers at bugzilla.org Subject: Re: Lines really long in comment fields Right now it's a problem when you enter the bug. It *should* be a problem when the bug is displayed. The intended fix is to get rid of the hard-wrap attributes on the text boxes and either do display-time wrapping when the comments are displayed to the user, or hard-wrap the text server-side after it's uploaded instead of depending on the browser to do it like we do currently. My personal vote is to do it display time, as it makes things more flexible and also automatically fixes all the old, currently-broken comments. On 3/24/03 9:05 AM -0500, Casey Gregoire wrote: > Is this a problem that comes from when you enter the bug or just when it is > displayed? > > Thanks, > Casey Gregoire > > -----Original Message----- > From: David Miller [mailto:justdave at syndicomm.com] > Sent: Saturday, March 22, 2003 11:06 AM > To: developers at bugzilla.org > Subject: Re: Lines really long in comment fields > > On 3/21/03 6:01 PM -0500, Casey Gregoire wrote: > >> Which patch do I apply? All I get is errors when I try to apply it. > > I don't think there's a working one on there yet. There were a few > prototypes done, but it's been a long time and the code has changed a lot. > (I think the existing patches are 2.14 era) -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ ---- To view or change your list settings, click here: From justdave at syndicomm.com Mon Mar 24 14:45:25 2003 From: justdave at syndicomm.com (David Miller) Date: Mon, 24 Mar 2003 09:45:25 -0500 Subject: Lines really long in comment fields In-Reply-To: References: Message-ID: Not that I'm aware of. On 3/24/03 9:33 AM -0500, Casey Gregoire wrote: > Is there a script to manually fix the fields that are wrong on the database? -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From caseyg at chsamerica.com Mon Mar 24 18:27:36 2003 From: caseyg at chsamerica.com (Casey Gregoire) Date: Mon, 24 Mar 2003 13:27:36 -0500 Subject: Lines really long in comment fields Message-ID: If the mozilla team uses Bugzilla for bugs, what do you use for internal issue ticket tracking? (As in problems with hard ware and computers?) Thanks, Casey Gregoire -----Original Message----- From: David Miller [mailto:justdave at syndicomm.com] Sent: Friday, March 21, 2003 3:53 PM To: developers at bugzilla.org Subject: Re: Lines really long in comment fields On 3/21/03 3:42 PM -0500, Casey Gregoire wrote: > When you enter a comment in 2.17 with out manually hitting enter to go down > to the next line the line will not word wrap at all when you view the bug. > Is this a problem I should put into the bug db? Or is it meant to be this > way? It means your browser doesn't honor a "netscapism" on the TEXTAREA attribute which Netscape, Mozilla, and MSIE all honor. Fixing Bugzilla to work without depending on that non-standard browser feature is http://bugzilla.mozilla.org/show_bug.cgi?id=11901 -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ ---- To view or change your list settings, click here: From gerv at mozilla.org Mon Mar 24 20:08:06 2003 From: gerv at mozilla.org (Gervase Markham) Date: Mon, 24 Mar 2003 20:08:06 +0000 Subject: Internal issue tracking In-Reply-To: References: Message-ID: <3E7F65A6.4050601@mozilla.org> Casey Gregoire wrote: > If the mozilla team uses Bugzilla for bugs, what do you use for internal > issue ticket tracking? (As in problems with hard ware and computers?) The Bugzilla developers list is not really the right place to ask this question, and mozilla.org doesn't have that much of a concept of "internal"; but the answer is "Bugzilla". http://bugzilla.mozilla.org/buglist.cgi?query_format=&short_desc_type=allwordssubstr&short_desc=&product=mozilla.org&component=Server+Operations&long_desc_type=substring&long_desc=&bug_file_loc_type=allwordssubstr&bug_file_loc=&status_whiteboard_type=allwordssubstr&status_whiteboard=&keywords_type=allwords&keywords=&bug_status=UNCONFIRMED&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&emailassigned_to1=1&emailtype1=exact&email1=&emailassigned_to2=1&emailreporter2=1&emailqa_contact2=1&emailtype2=exact&email2=&bugidtype=include&bug_id=&votes=&changedin=&chfieldfrom=&chfieldto=Now&chfieldvalue=&cmdtype=doit&remaction=run&namedcmd=2.18&order=Reuse+same+sort+as+last+time&field0-0-0=noop&type0-0-0=noop&value0-0-0= Gerv From myk at mozilla.org Mon Mar 24 20:22:33 2003 From: myk at mozilla.org (Myk Melez) Date: Mon, 24 Mar 2003 12:22:33 -0800 Subject: Hall of Fame In-Reply-To: <3E7F17CF.7020605@mozilla.org> References: <3E7F17CF.7020605@mozilla.org> Message-ID: <3E7F6909.7020809@mozilla.org> Gervase Markham wrote: > Casey Gregoire wrote: > >> Do you only want links directly to the company Bugzilla page? We may >> not be >> publicly giving access to Bugzilla at this time. I am still seeking >> permission from my employer as well for this. > > > Given that we cannot list every Bugzilla or Bugzilla-using company out > there, Why not? > and that this is a marketing tool, Part of the marketing message is that everybody uses it, so you should too. If the list gets too long we can break private companies down by category (internet, health care, etc.), with the corollary benefit that companies looking into Bugzilla can see similar companies already using it. Note, however, that we will never list every Bugzilla installation out there, since many will choose to remain private. I think our list isn't nearly long enough at this point for us to worry about having too many names on it. -myk From jimw at bugopolis.com Mon Mar 24 20:42:57 2003 From: jimw at bugopolis.com (Jim Walters) Date: 24 Mar 2003 12:42:57 -0800 Subject: Hall of Fame In-Reply-To: <3E7F6909.7020809@mozilla.org> References: <3E7F17CF.7020605@mozilla.org> <3E7F6909.7020809@mozilla.org> Message-ID: <1048538576.9294.66.camel@control> For easy tracking of those sites with Internet facing servers you might be able to put in a hyperlink (or keyword?) into the index.cgi template and use the Google "Page-Specific Search" Link search to find instances of pages that link back to your "for-tracking-down-in-the-wild" embedded link. I can't imagine anyone really minding. Regards, Jim On Mon, 2003-03-24 at 12:22, Myk Melez wrote: > Gervase Markham wrote: > > > Casey Gregoire wrote: > > > >> Do you only want links directly to the company Bugzilla page? We may > >> not be > >> publicly giving access to Bugzilla at this time. I am still seeking > >> permission from my employer as well for this. > > > > > > Given that we cannot list every Bugzilla or Bugzilla-using company out > > there, > > Why not? > > > and that this is a marketing tool, > > Part of the marketing message is that everybody uses it, so you should > too. If the list gets too long we can break private companies down by > category (internet, health care, etc.), with the corollary benefit that > companies looking into Bugzilla can see similar companies already using > it. Note, however, that we will never list every Bugzilla installation > out there, since many will choose to remain private. I think our list > isn't nearly long enough at this point for us to worry about having too > many names on it. > > -myk > > > ---- > To view or change your list settings, click here: > __________________________ Jim Walters Director of Technology Bugopolis, Inc. phone: +1 206 447 8315 email: jimw at bugopolis.com web: http://www.bugopolis.com _________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: From jason at pyeron.com Mon Mar 24 20:48:11 2003 From: jason at pyeron.com (Jason Pyeron) Date: Mon, 24 Mar 2003 15:48:11 -0500 (EST) Subject: Hall of Fame In-Reply-To: <1048538576.9294.66.camel@control> Message-ID: I am whipping up a RFE: http://bugzilla.mozilla.org/show_bug.cgi?id=199018 Jason Pyeron On 24 Mar 2003, Jim Walters wrote: For easy tracking of those sites with Internet facing servers you might be able to put in a hyperlink (or keyword?) into the index.cgi template and use the Google "Page-Specific Search" Link search to find instances of pages that link back to your "for-tracking-down-in-the-wild" embedded link. I can't imagine anyone really minding. Regards, Jim On Mon, 2003-03-24 at 12:22, Myk Melez wrote: > Gervase Markham wrote: > > > Casey Gregoire wrote: > > > >> Do you only want links directly to the company Bugzilla page? We may > >> not be > >> publicly giving access to Bugzilla at this time. I am still seeking > >> permission from my employer as well for this. > > > > > > Given that we cannot list every Bugzilla or Bugzilla-using company out > > there, > > Why not? > > > and that this is a marketing tool, > > Part of the marketing message is that everybody uses it, so you should > too. If the list gets too long we can break private companies down by > category (internet, health care, etc.), with the corollary benefit that > companies looking into Bugzilla can see similar companies already using > it. Note, however, that we will never list every Bugzilla installation > out there, since many will choose to remain private. I think our list > isn't nearly long enough at this point for us to worry about having too > many names on it. > > -myk > > > ---- > To view or change your list settings, click here: > __________________________ Jim Walters Director of Technology Bugopolis, Inc. phone: +1 206 447 8315 email: jimw at bugopolis.com web: http://www.bugopolis.com _________________________ -- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- - - - Jason Pyeron http://www.pyerotechnics.com - - Owner & Lead Pyerotechnics Development, Inc. - - +1 410 808 6646 (c) 500 West University Parkway #1S - - +1 410 467 2266 (f) Baltimore, Maryland 21210-3253 - - - -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- This message is for the designated recipient only and may contain privileged, proprietary, or otherwise private information. If you have received it in error, purge the message from your system and notify the sender immediately. Any other use of the email by you is prohibited. From gerv at mozilla.org Mon Mar 24 21:30:30 2003 From: gerv at mozilla.org (Gervase Markham) Date: Mon, 24 Mar 2003 21:30:30 +0000 Subject: An alternate approach to custom fields In-Reply-To: References: <3E7C34D8.80903@mozilla.org> Message-ID: <3E7F78F6.4040101@mozilla.org> David Miller wrote: > On 3/22/03 10:03 AM +0000, Gervase Markham wrote: >>>SortKey - Display order >> >>Where these have appeared elsewhere in Bugzilla, they have been a pain. >>Why is it needed? > > fielddefs already has sort keys. It's used to define the order the fields > are listed in when sending bugmail for new bugs. That's not something new > he's adding. Fair enough. It's the UI for managing them that sucks; presumably, this would be (is) factory-set to something sane. >>>Valexp - regular expression describing the valid pattern of input. If >>>null, no validation is performed. (also overrideable at lower level). >> >>Again, I'm not convinced we need this (other than the definition of enum >>types.) > > I can see the use for this in general for strings.... say you put in a > free-text version number field because you have too many versions to list > in a popup. Your version numbers are in a specific format that you can > always recognize though. You could use this to ensure anything entered in > that field is actually a legal version number. But what error would you throw if it failed? For it to be user-friendly, you'd also have to define an error tag; unless this was derived automatically from the field name... "-format-error" or something. Gerv From jason at pyeron.com Mon Mar 24 23:20:57 2003 From: jason at pyeron.com (Jason Pyeron) Date: Mon, 24 Mar 2003 18:20:57 -0500 (EST) Subject: Hall of Fame In-Reply-To: <3E7F6909.7020809@mozilla.org> Message-ID: Does any one want to put there $.02 in? see http://bugzilla.mozilla.org/show_bug.cgi?id=199018 I kinda dont want to work with this right now, yall seem to have more spirit, and I got a big case on our lap. Of course if a match gets lit I will be active. Jason Pyeron On Mon, 24 Mar 2003, Myk Melez wrote: Gervase Markham wrote: > Casey Gregoire wrote: > >> Do you only want links directly to the company Bugzilla page? We may >> not be >> publicly giving access to Bugzilla at this time. I am still seeking >> permission from my employer as well for this. > > > Given that we cannot list every Bugzilla or Bugzilla-using company out > there, Why not? > and that this is a marketing tool, Part of the marketing message is that everybody uses it, so you should too. If the list gets too long we can break private companies down by category (internet, health care, etc.), with the corollary benefit that companies looking into Bugzilla can see similar companies already using it. Note, however, that we will never list every Bugzilla installation out there, since many will choose to remain private. I think our list isn't nearly long enough at this point for us to worry about having too many names on it. -myk ---- To view or change your list settings, click here: -- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- - - - Jason Pyeron http://www.pyerotechnics.com - - Owner & Lead Pyerotechnics Development, Inc. - - +1 410 808 6646 (c) 500 West University Parkway #1S - - +1 410 467 2266 (f) Baltimore, Maryland 21210-3253 - - - -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- This message is for the designated recipient only and may contain privileged, proprietary, or otherwise private information. If you have received it in error, purge the message from your system and notify the sender immediately. Any other use of the email by you is prohibited. From gerv at mozilla.org Tue Mar 25 08:32:52 2003 From: gerv at mozilla.org (Gervase Markham) Date: Tue, 25 Mar 2003 08:32:52 +0000 Subject: SQL call formatting style Message-ID: <3E801434.1000406@mozilla.org> Guys, The new DBI stuff has raised a few issues about how we write SQL statements in Perl code (see bug 180635 for some discussion.) IMO, it's important to have a consistent style, because it makes code easier to read. What that style is is far less important than the consistency. One of the factors is multi-line strings. Myk: am I correct in saying that you have asked for "separate quotes for " . "each line of the string" because it makes logs look better than "one string which runs over multiple lines"? My suggestion is as follows: my ($comma, $separated, $list, $of, $vars, $on, $one, $line) = $dbh->some_longwinded_methodname("SELECT some SQL " . "ON multiple lines " . "IF necessary " . "AND enclosed in quotes " . "BECAUSE we all understand them " . "WITH placeholders ?, ?, ?", undef, $placeholder, $variables, $here); but obviously it's only a suggestion. The last few lines can be block-shifted to the left if necessary to make more room. In contrast, bbaetz recently went for: my ($id, $login, $name, $mybugslink) = $dbh->selectrow_array(qq{SELECT userid, login_name, realname, mybugslink FROM profiles WHERE $cond}, undef, $val); Gerv From bbaetz at acm.org Tue Mar 25 08:55:42 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Tue, 25 Mar 2003 19:55:42 +1100 Subject: SQL call formatting style In-Reply-To: <3E801434.1000406@mozilla.org> References: <3E801434.1000406@mozilla.org> Message-ID: <20030325085542.GA3234@mango.home> See my comments in the bug for why I think my way is better. Lots of " . " suck, basically, especially since we are no longer doing interpoliating, and " . sqlQuote($foo) . " everywhere, but are using laceholders instead. Bradley From geeta.tripathi at bhartitelesoft.com Tue Mar 25 11:25:42 2003 From: geeta.tripathi at bhartitelesoft.com (geeta) Date: Tue, 25 Mar 2003 16:55:42 +0530 Subject: SQL call formatting style References: <3E801434.1000406@mozilla.org> Message-ID: <000701c2f2c1$492b9c80$800310ac@tripathi.bhartitelesoft.com> Hello Gervase, I have a small issue, when I add QA and CC users while netering bugs , the Bugzilla shows that email sent to CC and QA but they don't receive this mail. regards, Geeta From chicks at chicks.net Tue Mar 25 13:53:22 2003 From: chicks at chicks.net (Christopher Hicks) Date: Tue, 25 Mar 2003 08:53:22 -0500 (EST) Subject: SQL call formatting style In-Reply-To: <3E801434.1000406@mozilla.org> Message-ID: On Tue, 25 Mar 2003, Gervase Markham wrote: > My suggestion is as follows: > > my ($comma, $separated, $list, $of, $vars, $on, $one, $line) = > $dbh->some_longwinded_methodname("SELECT some SQL " . > "ON multiple lines " . > "IF necessary " . > "AND enclosed in quotes " . > "BECAUSE we all understand them " . > "WITH placeholders ?, ?, ?", > undef, > $placeholder, $variables, $here); > > but obviously it's only a suggestion. The last few lines can be > block-shifted to the left if necessary to make more room. > > In contrast, bbaetz recently went for: > > my ($id, > $login, > $name, > $mybugslink) = $dbh->selectrow_array(qq{SELECT userid, > login_name, > realname, > mybugslink > FROM profiles > WHERE $cond}, > undef, > $val); Having the SQL statement in a scalar makes the $dbh calls much cleaner. It also allows you to easily replace extra whitespace with a single space if you'd like to do so for logging. $sql = qq{ SELECT userid, login_name, realname, mybugslink FROM profiles WHERE $cond }; Having more width for the SQL statement lets you do thinks like $sql = qq{ SELECT userid*256 as userid, left(login_name,2) as login_name, right(realname,10) as realname, substr(mybugslink,2,20) as mybugslink FROM profiles WHERE $cond }; An acceptable alternative should allow for the SQL statement to be built up across a chunk of code: my $sql = "select "; $sql .= join(', ', at whatweneed); $sql .= " from sometable"; > One of the factors is multi-line strings. Myk: am I correct in saying > that you have asked for > "separate quotes for " . > "each line of the string" > because it makes logs look better than > "one string which runs > over multiple lines"? By putting the SQL in a scalar it's easy to gclean it up before logging it: $logsql = $sql; $logsql =~ s/\s+/ /g; -- The death of democracy is not likely to be an assassination from ambush. It will be a slow extinction from apathy, indifference, and undernourishment. -Robert Maynard Hutchins, educator (1899-1977) From dswegen at software.plasmon.com Tue Mar 25 13:53:49 2003 From: dswegen at software.plasmon.com (Dave Swegen) Date: Tue, 25 Mar 2003 13:53:49 +0000 Subject: CVS integration patch Message-ID: <20030325135349.GQ9460@software.plasmon> I thought I'd let those showed interest in this know that the patches required to get CVS commit logs into bz comments have been attached to bug 199116 (http://bugzilla.mozilla.org/show_bug.cgi?id=199116). Cheers Dave From bbaetz at acm.org Tue Mar 25 14:03:57 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Wed, 26 Mar 2003 01:03:57 +1100 Subject: SQL call formatting style In-Reply-To: References: <3E801434.1000406@mozilla.org> Message-ID: <20030325140357.GA6017@mango.home> On Tue, Mar 25, 2003 at 08:53:22AM -0500, Christopher Hicks wrote: > > Having the SQL statement in a scalar makes the $dbh calls much cleaner. > It also allows you to easily replace extra whitespace with a single space > if you'd like to do so for logging. logging would be done via the DBI trace stuff, so thats not an issue. As I mentioed in the bug, though. I especially prefer whitespace when debugging. Splitting up the various clauses is a lot easer on teh eyes than one jumbled-together string. Well, thats a separate issue. Our more coplicated sql queris (ie BugList.pm) have to be autogenarated to a scalar. The majority are just simple selects, though. > > An acceptable alternative should allow for the SQL statement to be built > up across a chunk of code: > > my $sql = "select "; > $sql .= join(', ', at whatweneed); > $sql .= " from sometable"; *cough* subselect *cough* > By putting the SQL in a scalar it's easy to gclean it up before logging > it: See above Bradley From sergeyli at pisem.net Tue Mar 25 14:53:01 2003 From: sergeyli at pisem.net (Sergey A. Lipnevich) Date: Tue, 25 Mar 2003 09:53:01 -0500 Subject: SQL call formatting style In-Reply-To: <20030325085542.GA3234@mango.home> References: <3E801434.1000406@mozilla.org> <20030325085542.GA3234@mango.home> Message-ID: <3E806D4D.5030104@pisem.net> I started coding SQL like this (as scalar, akin to shell's cat << EOF ... EOF) some time ago and now I hate when the language that doesn't allow that (e.g. Java). In situations like Java+SQL I'm willing to go extra length and store SQL in external config file, but not concatenate clauses together in the code. It's absolutely worth it, and parameterized (prepared) statements with question marks are what they call "enabler" in this case. My $.0029/line. Bradley Baetz wrote: > See my comments in the bug for why I think my way is better. Lots of " . > " suck, basically, especially since we are no longer doing > interpoliating, and " . sqlQuote($foo) . " everywhere, but are using > laceholders instead. > > Bradley > ---- > To view or change your list settings, click here: > > From jason at pyeron.com Tue Mar 25 16:04:57 2003 From: jason at pyeron.com (Jason Pyeron) Date: Tue, 25 Mar 2003 11:04:57 -0500 (EST) Subject: SQL call formatting style In-Reply-To: <3E806D4D.5030104@pisem.net> Message-ID: Guaranteed not to like this opinion or your money back... Here this is from our coding policy: Scan ability of parameters is highly important. If a string or formula is so long that it has to wrap multiple lines, then it should be the last item which needs to be read for the snippet of code. When wrapping lines always use a concatenation symbol. When there are no other parameters or the long string is the last parameter: foo( "This is a really long dialog message, it may be necessary to wrap"+ "lines, and if it is passed by value to a function, special"+ "formatting rules may apply." ); bar(99,0,varX, "This is a really long dialog message, it may be necessary to wrap"+ "lines, and if it is passed by value to a function, special"+ "formatting rules may apply." ); But when there are other parameters after the long string, the long string must be stored in a local var: string msg= "This is a really long dialog message, it may be necessary to wrap"+ "lines, and if it is passed by value to a function, special"+ "formatting rules may apply."; bar(msg,99,0,varX); Jason Pyeron On Tue, 25 Mar 2003, Sergey A. Lipnevich wrote: I started coding SQL like this (as scalar, akin to shell's cat << EOF ... EOF) some time ago and now I hate when the language that doesn't allow that (e.g. Java). In situations like Java+SQL I'm willing to go extra length and store SQL in external config file, but not concatenate clauses together in the code. It's absolutely worth it, and parameterized (prepared) statements with question marks are what they call "enabler" in this case. My $.0029/line. Bradley Baetz wrote: > See my comments in the bug for why I think my way is better. Lots of " . > " suck, basically, especially since we are no longer doing > interpoliating, and " . sqlQuote($foo) . " everywhere, but are using > laceholders instead. > > Bradley > ---- > To view or change your list settings, click here: > > ---- To view or change your list settings, click here: -- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- - - - Jason Pyeron http://www.pyerotechnics.com - - Owner & Lead Pyerotechnics Development, Inc. - - +1 410 808 6646 (c) 500 West University Parkway #1S - - +1 410 467 2266 (f) Baltimore, Maryland 21210-3253 - - - -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- This message is for the designated recipient only and may contain privileged, proprietary, or otherwise private information. If you have received it in error, purge the message from your system and notify the sender immediately. Any other use of the email by you is prohibited. From myk at mozilla.org Tue Mar 25 17:00:04 2003 From: myk at mozilla.org (Myk Melez) Date: Tue, 25 Mar 2003 09:00:04 -0800 Subject: SQL call formatting style In-Reply-To: <3E801434.1000406@mozilla.org> References: <3E801434.1000406@mozilla.org> Message-ID: <3E808B14.3000002@mozilla.org> Gervase Markham wrote: > Guys, > > The new DBI stuff has raised a few issues about how we write SQL > statements in Perl code (see bug 180635 for some discussion.) IMO, > it's important to have a consistent style, because it makes code > easier to read. What that style is is far less important than the > consistency. > > One of the factors is multi-line strings. Myk: am I correct in saying > that you have asked for > "separate quotes for " . > "each line of the string" > because it makes logs look better than > "one string which runs > over multiple lines"? Yes, because newlines mess up the structured format in which "mysqladmin processlist" displays queries, making it hard to identify errant queries, especially programmatically. Otherwise I'd be on bbaetz' side, since I find multi-line strings easier to work with, but being able to work with the MySQL process list is really important for Bugzilla system administrators, so I think we should use concatenation or newline substitution where possible (queries with strings that may have newlines in them can't be easily stripped of newlines with s/\s+/ /g). -myk From sergeyli at pisem.net Tue Mar 25 18:19:38 2003 From: sergeyli at pisem.net (Sergey A. Lipnevich) Date: Tue, 25 Mar 2003 13:19:38 -0500 Subject: SQL call formatting style In-Reply-To: References: <3E806D4D.5030104@pisem.net> Message-ID: I didn't think this discussion is about /parameters/, and so far I'm justified by other opinions. The policy itself looks good but has nothing to do with how to code these long parameters, as one multiline value or several concatenated chunks. Sorry if I misunderstood... Jason Pyeron wrote: > Guaranteed not to like this opinion or your money back... > > Here this is from our coding policy: > > Scan ability of parameters is highly important. If a string or formula is > so long that it has to wrap multiple lines, then it should be the last > item which needs to be read for the snippet of code. > > When wrapping lines always use a concatenation symbol. > > When there are no other parameters or the long string is the last > parameter: > > foo( > "This is a really long dialog message, it may be necessary to wrap"+ > "lines, and if it is passed by value to a function, special"+ > "formatting rules may apply." > ); > > bar(99,0,varX, > "This is a really long dialog message, it may be necessary to wrap"+ > "lines, and if it is passed by value to a function, special"+ > "formatting rules may apply." > ); > > But when there are other parameters after the long string, the long string > must be stored in a local var: > > string msg= > "This is a really long dialog message, it may be necessary to wrap"+ > "lines, and if it is passed by value to a function, special"+ > "formatting rules may apply."; > > bar(msg,99,0,varX); > > Jason Pyeron > > On Tue, 25 Mar 2003, Sergey A. Lipnevich wrote: > > I started coding SQL like this (as scalar, akin to shell's cat << EOF > ... EOF) some time ago and now I hate when the language that doesn't > allow that (e.g. Java). In situations like Java+SQL I'm willing to go > extra length and store SQL in external config file, but not concatenate > clauses together in the code. It's absolutely worth it, and > parameterized (prepared) statements with question marks are what they > call "enabler" in this case. My $.0029/line. > > Bradley Baetz wrote: > >>See my comments in the bug for why I think my way is better. Lots of " . >>" suck, basically, especially since we are no longer doing >>interpoliating, and " . sqlQuote($foo) . " everywhere, but are using >>laceholders instead. >> >>Bradley >>---- >>To view or change your list settings, click here: >> >> > > > ---- > To view or change your list settings, click here: > > > From justdave at syndicomm.com Tue Mar 25 18:38:27 2003 From: justdave at syndicomm.com (David Miller) Date: Tue, 25 Mar 2003 13:38:27 -0500 Subject: SQL call formatting style In-Reply-To: <000701c2f2c1$492b9c80$800310ac@tripathi.bhartitelesoft.com> References: <3E801434.1000406@mozilla.org> <000701c2f2c1$492b9c80$800310ac@tripathi.bhartitelesoft.com> Message-ID: On 3/25/03 4:55 PM +0530, geeta wrote: > Hello Gervase, > > I have a small issue, when I add QA and CC users while netering bugs , the > Bugzilla shows that email sent to CC and QA but they don't receive this > mail. Geeta: This is WAY off-topic for the thread you are replying to. Please start a new thread/subject when you have a new question. Also, it would be REALLY helpful if you could read the page at the following URL before posting here again: http://www.catb.org/~esr/faqs/smart-questions.html -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From jason at pyeron.com Tue Mar 25 19:06:47 2003 From: jason at pyeron.com (Jason Pyeron) Date: Tue, 25 Mar 2003 14:06:47 -0500 (EST) Subject: SQL call formatting style In-Reply-To: Message-ID: Sergey, I understand our policy does not explain why to use a concatenation, but here is the gist: when new lines are embedded by formatting it is not obvious or explicit as to their existence. It is always better to use a "\n". This point was supported by Myk's statement: ... I find multi-line strings easier to work with, but being able to work with the MySQL process list is really important for Bugzilla system administrators, so I think we should use concatenation ... On Tue, 25 Mar 2003, Sergey A. Lipnevich wrote: I didn't think this discussion is about /parameters/, and so far I'm justified by other opinions. The policy itself looks good but has nothing to do with how to code these long parameters, as one multiline value or several concatenated chunks. Sorry if I misunderstood... Jason Pyeron wrote: ... > When wrapping lines always use a concatenation symbol. ... -- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- - - - Jason Pyeron http://www.pyerotechnics.com - - Owner & Lead Pyerotechnics Development, Inc. - - +1 410 808 6646 (c) 500 West University Parkway #1S - - +1 410 467 2266 (f) Baltimore, Maryland 21210-3253 - - - -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- This message is for the designated recipient only and may contain privileged, proprietary, or otherwise private information. If you have received it in error, purge the message from your system and notify the sender immediately. Any other use of the email by you is prohibited. From sergeyli at pisem.net Tue Mar 25 19:47:50 2003 From: sergeyli at pisem.net (Sergey A. Lipnevich) Date: Tue, 25 Mar 2003 14:47:50 -0500 Subject: SQL call formatting style In-Reply-To: References: Message-ID: Jason, Then I respectfully disagree with this policy. I've read Myk's post before replying, and that's more of "let's fill cars with water because gasoline doesn't smell nice to a driver" kind of thing. The SQL in code is for the database and for developers, and administrators would be much better served by a tool which is adequate for their task. If a malicious user manages to squeeze destructive SQL code with newlines past Bugzilla, will a good tool show administrators entire statement or just the first line? So, administrators need to take care of statements with or without newlines, and this shouldn't be a reason for makeing developers' life difficult. By a modest estimation of mine, keeping SQL statements as sequences of characters uninterrupted by syntax alien to SQL, improves debugging by the order of mangnitude, which is 10 times. Besides, newlines mean nothing in SQL and practically nothing in HTML or XML. Developers must be allowed to use them to their advantage. I'm writing three page-long SQL statements nowadays (to be executed by DB2) with almost every significant token or condition on a separate line, and so far both administrators and testers like them very much -- they can actually read through them (in the logs or otherwise) as they would an email or a web page. Jason Pyeron wrote: > Sergey, > > I understand our policy does not explain why to use a concatenation, > but here is the gist: when new lines are embedded by formatting it is not > obvious or explicit as to their existence. It is always better to use a > "\n". > > This point was supported by Myk's statement: > > ... I find multi-line strings easier to work with, but being able to > work with the MySQL process list is really important for Bugzilla system > administrators, so I think we should use concatenation ... > > From jason at pyeron.com Tue Mar 25 20:03:56 2003 From: jason at pyeron.com (Jason Pyeron) Date: Tue, 25 Mar 2003 15:03:56 -0500 (EST) Subject: SQL call formatting style In-Reply-To: Message-ID: In those cases (3pg query) it should be put above the function not in-lined. If your structure for that query (i agree with the no foreign symbols) is sparse, then use the quoting scheme best suited. But don't then "optimize" it into the function call, mangling any hopes of speed-reading. On Tue, 25 Mar 2003, Sergey A. Lipnevich wrote: Jason, Then I respectfully disagree with this policy. I've read Myk's post before replying, and that's more of "let's fill cars with water because gasoline doesn't smell nice to a driver" kind of thing. The SQL in code is for the database and for developers, and administrators would be much better served by a tool which is adequate for their task. If a malicious user manages to squeeze destructive SQL code with newlines past Bugzilla, will a good tool show administrators entire statement or just the first line? So, administrators need to take care of statements with or without newlines, and this shouldn't be a reason for makeing developers' life difficult. By a modest estimation of mine, keeping SQL statements as sequences of characters uninterrupted by syntax alien to SQL, improves debugging by the order of mangnitude, which is 10 times. Besides, newlines mean nothing in SQL and practically nothing in HTML or XML. Developers must be allowed to use them to their advantage. I'm writing three page-long SQL statements nowadays (to be executed by DB2) with almost every significant token or condition on a separate line, and so far both administrators and testers like them very much -- they can actually read through them (in the logs or otherwise) as they would an email or a web page. Jason Pyeron wrote: > Sergey, > > I understand our policy does not explain why to use a concatenation, > but here is the gist: when new lines are embedded by formatting it is not > obvious or explicit as to their existence. It is always better to use a > "\n". > > This point was supported by Myk's statement: > > ... I find multi-line strings easier to work with, but being able to > work with the MySQL process list is really important for Bugzilla system > administrators, so I think we should use concatenation ... > > ---- To view or change your list settings, click here: -- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- - - - Jason Pyeron http://www.pyerotechnics.com - - Owner & Lead Pyerotechnics Development, Inc. - - +1 410 808 6646 (c) 500 West University Parkway #1S - - +1 410 467 2266 (f) Baltimore, Maryland 21210-3253 - - - -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- This message is for the designated recipient only and may contain privileged, proprietary, or otherwise private information. If you have received it in error, purge the message from your system and notify the sender immediately. Any other use of the email by you is prohibited. From bbaetz at acm.org Tue Mar 25 20:11:37 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Wed, 26 Mar 2003 07:11:37 +1100 Subject: SQL call formatting style In-Reply-To: <3E808B14.3000002@mozilla.org> References: <3E801434.1000406@mozilla.org> <3E808B14.3000002@mozilla.org> Message-ID: <20030325201137.GA1330@mango.home> On Tue, Mar 25, 2003 at 09:00:04AM -0800, Myk Melez wrote: > Yes, because newlines mess up the structured format in which "mysqladmin > processlist" displays queries, making it hard to identify errant > queries, especially programmatically. Otherwise I'd be on bbaetz' side, > since I find multi-line strings easier to work with, but being able to > work with the MySQL process list is really important for Bugzilla system > administrators, so I think we should use concatenation or newline > substitution where possible (queries with strings that may have newlines > in them can't be easily stripped of newlines with s/\s+/ /g). Well...... Thats a point, I guess, but I'd guess that that doens't happen too often. Doesn't mysql have a query timeout option which can replace that mysqlkill thing now, anyway? You'd be better off patching mysqladmin, I think. Our long running queries are basically only Bugzilla/Search.pm - the sort of stuff I'm using this for is just simple accessor stuff in the db. I thinkthat readability in the source code (using Q/QQ instead of concatenated strings) should take priority over that. Bradley From sergeyli at pisem.net Tue Mar 25 20:24:48 2003 From: sergeyli at pisem.net (Sergey A. Lipnevich) Date: Tue, 25 Mar 2003 15:24:48 -0500 Subject: SQL call formatting style In-Reply-To: References: Message-ID: Jason Pyeron wrote: > In those cases (3pg query) it should be put above the function not > in-lined. Absolutely. > > If your structure for that query (i agree with the no foreign symbols) is > sparse, then use the quoting scheme best suited. But don't then "optimize" > it into the function call, mangling any hopes of speed-reading. > We are in agreement then :-). I've just read latest Bradley's post and he's explained this much more concisely than I did (going as far suggesting to patch mysqladmin -- I didn't dare to use /this/ as an argument ;-) ). From jake at bugzilla.org Tue Mar 25 10:56:13 2003 From: jake at bugzilla.org (Jake) Date: Tue, 25 Mar 2003 05:56:13 -0500 Subject: SQL call formatting style In-Reply-To: <3E808B14.3000002@mozilla.org> References: <3E801434.1000406@mozilla.org> <3E808B14.3000002@mozilla.org> Message-ID: <3E8035CD.5030400@bugzilla.org> Myk Melez wrote: > (queries with strings that may have newlines in them can't be easily > stripped of newlines with s/\s+/ /g). > In my brief testing, it also collapses newlines. From justdave at syndicomm.com Tue Mar 25 23:26:11 2003 From: justdave at syndicomm.com (David Miller) Date: Tue, 25 Mar 2003 18:26:11 -0500 Subject: SQL call formatting style In-Reply-To: <3E8035CD.5030400@bugzilla.org> References: <3E801434.1000406@mozilla.org> <3E808B14.3000002@mozilla.org> <3E8035CD.5030400@bugzilla.org> Message-ID: On 3/25/03 5:56 AM -0500, Jake wrote: > Myk Melez wrote: > >> (queries with strings that may have newlines in them can't be easily >> stripped of newlines with s/\s+/ /g). >> > In my brief testing, it also collapses newlines. if you do s/\s+/ /gs (note the s on the end) it'll remove newlines. What you want to be careful of is what if you have a quoted string which contains multiple spaces within the query? you don't want to remove the spaces from the content of the string.... On the other hand, if we're using placeholders we shouldn't have to worry about that because there won't be anything quoted in the query itself. -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From gerv at mozilla.org Tue Mar 25 23:27:59 2003 From: gerv at mozilla.org (Gervase Markham) Date: Tue, 25 Mar 2003 23:27:59 +0000 Subject: SQL call formatting style In-Reply-To: References: Message-ID: <3E80E5FF.1050706@mozilla.org> If we are to allow newlines, then my proposal would become: my ($comma, $separated, $list, $of, $vars, $on, $one, $line) = $dbh->some_longwinded_methodname("SELECT some SQL variables ON multiple lines IF necessary AND enclosed in quotes BECAUSE we all understand them WITH placeholders ?, ?, ?", undef, $placeholder, $variables, $here); I believe this strikes a reasonable balance between clarity and longwindedness. It may be true that if you space the SQL out a lot, it's easier to read the SQL - but it's much harder to follow the flow of the code around it. Re: Myk's point: when do we ever do queries with newlines embedded in the query data? Gerv From jason at pyeron.com Tue Mar 25 23:36:41 2003 From: jason at pyeron.com (Jason Pyeron) Date: Tue, 25 Mar 2003 18:36:41 -0500 (EST) Subject: SQL call formatting style In-Reply-To: <3E80E5FF.1050706@mozilla.org> Message-ID: use the quoting type of prefernce... my $tmpsql="SELECT some SQL variables ON multiple lines IF necessary AND enclosed in quotes BECAUSE we all understand them WITH placeholders ?, ?, ?"; my ($comma, $separated, $list, $of, $vars, $on, $one, $line) = $dbh->some_longwinded_methodname($tmpsql,undef, $placeholder, $variables, $here); is much better to read. On Tue, 25 Mar 2003, Gervase Markham wrote: If we are to allow newlines, then my proposal would become: my ($comma, $separated, $list, $of, $vars, $on, $one, $line) = $dbh->some_longwinded_methodname("SELECT some SQL variables ON multiple lines IF necessary AND enclosed in quotes BECAUSE we all understand them WITH placeholders ?, ?, ?", undef, $placeholder, $variables, $here); I believe this strikes a reasonable balance between clarity and longwindedness. It may be true that if you space the SQL out a lot, it's easier to read the SQL - but it's much harder to follow the flow of the code around it. Re: Myk's point: when do we ever do queries with newlines embedded in the query data? Gerv ---- To view or change your list settings, click here: -- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- - - - Jason Pyeron http://www.pyerotechnics.com - - Owner & Lead Pyerotechnics Development, Inc. - - +1 410 808 6646 (c) 500 West University Parkway #1S - - +1 410 467 2266 (f) Baltimore, Maryland 21210-3253 - - - -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- This message is for the designated recipient only and may contain privileged, proprietary, or otherwise private information. If you have received it in error, purge the message from your system and notify the sender immediately. Any other use of the email by you is prohibited. From justdave at syndicomm.com Tue Mar 25 23:35:17 2003 From: justdave at syndicomm.com (David Miller) Date: Tue, 25 Mar 2003 18:35:17 -0500 Subject: SQL call formatting style In-Reply-To: <3E80E5FF.1050706@mozilla.org> References: <3E80E5FF.1050706@mozilla.org> Message-ID: On 3/25/03 11:27 PM +0000, Gervase Markham wrote: > If we are to allow newlines, then my proposal would become: > > my ($comma, $separated, $list, $of, $vars, $on, $one, $line) = > $dbh->some_longwinded_methodname("SELECT some SQL variables > ON multiple lines > IF necessary > AND enclosed in quotes > BECAUSE we all understand them > WITH placeholders ?, ?, ?", > undef, > $placeholder, $variables, $here); > > I believe this strikes a reasonable balance between clarity and > longwindedness. It may be true that if you space the SQL out a lot, it's > easier to read the SQL - but it's much harder to follow the flow of the > code around it. > > Re: Myk's point: when do we ever do queries with newlines embedded in > the query data? The example you just posted (quoted here) does. :) -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From gerv at mozilla.org Tue Mar 25 23:57:59 2003 From: gerv at mozilla.org (Gervase Markham) Date: Tue, 25 Mar 2003 23:57:59 +0000 Subject: SQL call formatting style In-Reply-To: References: <3E80E5FF.1050706@mozilla.org> Message-ID: <3E80ED07.9020508@mozilla.org> David Miller wrote: >>Re: Myk's point: when do we ever do queries with newlines embedded in >>the query data? > > The example you just posted (quoted here) does. :) No, I mean, one objection to the s/\s+/ /gs; thing was (if I read it right) that it would remove newlines embedded in e.g. search strings in queries. I was asking whether we actually do that? Gerv From justdave at syndicomm.com Wed Mar 26 00:13:03 2003 From: justdave at syndicomm.com (David Miller) Date: Tue, 25 Mar 2003 19:13:03 -0500 Subject: SQL call formatting style In-Reply-To: <3E80ED07.9020508@mozilla.org> References: <3E80E5FF.1050706@mozilla.org> <3E80ED07.9020508@mozilla.org> Message-ID: On 3/25/03 11:57 PM +0000, Gervase Markham wrote: > David Miller wrote: >>>Re: Myk's point: when do we ever do queries with newlines embedded in >>>the query data? >> >> The example you just posted (quoted here) does. :) > > No, I mean, one objection to the s/\s+/ /gs; thing was (if I read it > right) that it would remove newlines embedded in e.g. search strings in > queries. I was asking whether we actually do that? In Search.pm we do. Everything gets quoted where appropriate and appended to the appropriate section of the query. By the time we're done, we have one big long query string. -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From gerv at mozilla.org Wed Mar 26 07:23:38 2003 From: gerv at mozilla.org (Gervase Markham) Date: Wed, 26 Mar 2003 07:23:38 +0000 Subject: SQL call formatting style In-Reply-To: References: <3E80E5FF.1050706@mozilla.org> <3E80ED07.9020508@mozilla.org> Message-ID: <3E81557A.9000001@mozilla.org> >>No, I mean, one objection to the s/\s+/ /gs; thing was (if I read it >>right) that it would remove newlines embedded in e.g. search strings in >>queries. I was asking whether we actually do that? > > In Search.pm we do. Everything gets quoted where appropriate and appended > to the appropriate section of the query. By the time we're done, we have > one big long query string. Yes, I know :-) What I mean is, is there ever a reason to have a \n embedded in a query string? If not, you could do: s/\s*\n\s*/ /gs; to remove all the whitespace introduced by linebreaking the SQL, while keeping all other whitespace intact. Gerv From bbaetz at acm.org Wed Mar 26 08:15:20 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Wed, 26 Mar 2003 19:15:20 +1100 Subject: SQL call formatting style In-Reply-To: <3E80ED07.9020508@mozilla.org> References: <3E80E5FF.1050706@mozilla.org> <3E80ED07.9020508@mozilla.org> Message-ID: <20030326081520.GA1449@mango.home> On Tue, Mar 25, 2003 at 11:57:59PM +0000, Gervase Markham wrote: > No, I mean, one objection to the s/\s+/ /gs; thing was (if I read it > right) that it would remove newlines embedded in e.g. search strings in > queries. I was asking whether we actually do that? Does it matter? Where was it that you were planning on munging this, anyway? :) I'm not even sure that we care about the stuff in error messages, anyway. The only complicated ones are for Search.pm, and thats not the focus of this anyway, since that query is dynamically generated. I'm talking about the stuff which are (basically) static, simple, strings, very, very, very occasionally with |$whereClause| added in the middle. Not to mention that all of the "." stuff is annoying. Since we don't have to keep opening an closing teh quotes for SqlQuote + friends, q{} is probably nicer/simpler/etc > Gerv Bradley From gerv at mozilla.org Wed Mar 26 09:10:00 2003 From: gerv at mozilla.org (Gervase Markham) Date: Wed, 26 Mar 2003 09:10:00 +0000 Subject: SQL call formatting style In-Reply-To: <20030326081520.GA1449@mango.home> References: <3E80E5FF.1050706@mozilla.org> <3E80ED07.9020508@mozilla.org> <20030326081520.GA1449@mango.home> Message-ID: <3E816E68.5090400@mozilla.org> Bradley Baetz wrote: > On Tue, Mar 25, 2003 at 11:57:59PM +0000, Gervase Markham wrote: > >>No, I mean, one objection to the s/\s+/ /gs; thing was (if I read it >>right) that it would remove newlines embedded in e.g. search strings in >>queries. I was asking whether we actually do that? > > > Does it matter? Where was it that you were planning on munging this, > anyway? :) > > I'm not even sure that we care about the stuff in error messages, > anyway. The only complicated ones are for Search.pm, and thats not the > focus of this anyway, since that query is dynamically generated. I'm > talking about the stuff which are (basically) static, simple, strings, > very, very, very occasionally with |$whereClause| added in the middle. > > Not to mention that all of the "." stuff is annoying. You can stop making this point; I completely agree, and my most recent proposal doesn't include them. :-) But, what's the advantage of qq{FOO BAR} over "FOO BAR" ? I find the latter easier to read. Gerv From kiko at async.com.br Thu Mar 27 00:43:48 2003 From: kiko at async.com.br (Christian Reis) Date: Wed, 26 Mar 2003 21:43:48 -0300 Subject: Hall of Fame In-Reply-To: <3E7F17CF.7020605@mozilla.org>; from gerv@mozilla.org on Mon, Mar 24, 2003 at 02:35:59PM +0000 References: <3E7F17CF.7020605@mozilla.org> Message-ID: <20030326214348.C9360@blackjesus.async.com.br> On Mon, Mar 24, 2003 at 02:35:59PM +0000, Gervase Markham wrote: > Casey Gregoire wrote: > > Do you only want links directly to the company Bugzilla page? We may not be > > publicly giving access to Bugzilla at this time. I am still seeking > > permission from my employer as well for this. > > Given that we cannot list every Bugzilla or Bugzilla-using company out > there, and that this is a marketing tool, I propose the following > criteria for HOF entry: > > Either: > - Public Bugzilla we can link to (public as in "you can search it") Just as a data point, Async Open Source have bugs.async.com.br, which runs with the experimental ;) narrow layout and hosts projects like Stoq, Kiwi, IndexedCatalog and others. It also hosts a number of private or internal projects/products. Take care, -- Christian Reis, Senior Engineer, Async Open Source, Brazil. http://async.com.br/~kiko/ | [+55 16] 261 2331 | NMFL From caseyg at chsamerica.com Thu Mar 27 16:08:55 2003 From: caseyg at chsamerica.com (Casey Gregoire) Date: Thu, 27 Mar 2003 11:08:55 -0500 Subject: Hall of Fame Message-ID: And how do you suppose we talk about dinner at lunch? You are not going to lunch with me today.... Thanks, Casey Gregoire -----Original Message----- From: Christian Reis [mailto:kiko at async.com.br] Sent: Wednesday, March 26, 2003 7:44 PM To: developers at bugzilla.org Subject: Re: Hall of Fame On Mon, Mar 24, 2003 at 02:35:59PM +0000, Gervase Markham wrote: > Casey Gregoire wrote: > > Do you only want links directly to the company Bugzilla page? We may not be > > publicly giving access to Bugzilla at this time. I am still seeking > > permission from my employer as well for this. > > Given that we cannot list every Bugzilla or Bugzilla-using company out > there, and that this is a marketing tool, I propose the following > criteria for HOF entry: > > Either: > - Public Bugzilla we can link to (public as in "you can search it") Just as a data point, Async Open Source have bugs.async.com.br, which runs with the experimental ;) narrow layout and hosts projects like Stoq, Kiwi, IndexedCatalog and others. It also hosts a number of private or internal projects/products. Take care, -- Christian Reis, Senior Engineer, Async Open Source, Brazil. http://async.com.br/~kiko/ | [+55 16] 261 2331 | NMFL ---- To view or change your list settings, click here: From caseyg at chsamerica.com Thu Mar 27 16:12:09 2003 From: caseyg at chsamerica.com (Casey Gregoire) Date: Thu, 27 Mar 2003 11:12:09 -0500 Subject: Hall of Fame Message-ID: I am sorry for that last post, I replied to the wrong e-mail. I will be more careful from now on. Thanks, Casey Gregoire -----Original Message----- From: Casey Gregoire [mailto:caseyg at chsamerica.com] Sent: Thursday, March 27, 2003 11:09 AM To: developers at bugzilla.org Subject: Re: Hall of Fame And how do you suppose we talk about dinner at lunch? You are not going to lunch with me today.... Thanks, Casey Gregoire -----Original Message----- From: Christian Reis [mailto:kiko at async.com.br] Sent: Wednesday, March 26, 2003 7:44 PM To: developers at bugzilla.org Subject: Re: Hall of Fame On Mon, Mar 24, 2003 at 02:35:59PM +0000, Gervase Markham wrote: > Casey Gregoire wrote: > > Do you only want links directly to the company Bugzilla page? We may not be > > publicly giving access to Bugzilla at this time. I am still seeking > > permission from my employer as well for this. > > Given that we cannot list every Bugzilla or Bugzilla-using company out > there, and that this is a marketing tool, I propose the following > criteria for HOF entry: > > Either: > - Public Bugzilla we can link to (public as in "you can search it") Just as a data point, Async Open Source have bugs.async.com.br, which runs with the experimental ;) narrow layout and hosts projects like Stoq, Kiwi, IndexedCatalog and others. It also hosts a number of private or internal projects/products. Take care, -- Christian Reis, Senior Engineer, Async Open Source, Brazil. http://async.com.br/~kiko/ | [+55 16] 261 2331 | NMFL ---- To view or change your list settings, click here: ---- To view or change your list settings, click here: From cmason at somanetworks.com Thu Mar 27 16:44:07 2003 From: cmason at somanetworks.com (Chris L. Mason) Date: Thu, 27 Mar 2003 11:44:07 -0500 Subject: Suggesting user names Message-ID: <20030327164407.GB29163@somanetworks.com> Hi, We've implemented a parameter ("suggest-users") with the following description: "This option will perform loose name matching in case a username entered does not match, and will provide the user with a list so that they might find the username they were looking for. This is disabled by default for security and privacy reasons." We have this working (in 2.16.2) and find it very useful, for example, when a user enters a name in the cc box or the "assigned to" box, but they spell it wrong. Then they get a list of users and can back up and try again. One thing I'm not happy with, however, is the method we're using to do the pattern-matching. If this is a patch you think would be useful, I'd like to get your input on the best way to do pattern matching. The goal is to have it do a good job at finding alternatives, but not present too much of a performance penelty. Unfortunately mysql doesn't appear to have a SOUNDEX() SQL function, which would make this easier. Anyway, do you think this would be a patch you'd like to import, and if so, what sort of pattern matching would you recommend? (currently we're just using a substring of the name entered and doing a LIKE search on login_name and realname.) Thanks, Chris -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 232 bytes Desc: not available URL: From justdave at syndicomm.com Thu Mar 27 18:27:23 2003 From: justdave at syndicomm.com (David Miller) Date: Thu, 27 Mar 2003 13:27:23 -0500 Subject: Suggesting user names In-Reply-To: <20030327164407.GB29163@somanetworks.com> References: <20030327164407.GB29163@somanetworks.com> Message-ID: On 3/27/03 11:44 AM -0500, Chris L. Mason wrote: > We've implemented a parameter ("suggest-users") with the following > description: > > "This option will perform loose name matching in case a username entered > does not match, and will provide the user with a list so that they might > find the username they were looking for. This is disabled by default for > security and privacy reasons." > > We have this working (in 2.16.2) and find it very useful, for example, > when a user enters a name in the cc box or the "assigned to" box, but they > spell it wrong. Then they get a list of users and can back up and try > again. Just FYI, 2.17.1 and up already do this. :) Although having a similar patch available for 2.16.x would probably be enjoyed by a lot of users whose management won't let them upgrade to an "unstable" version. :) -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From cmason at somanetworks.com Thu Mar 27 20:04:56 2003 From: cmason at somanetworks.com (Chris L. Mason) Date: Thu, 27 Mar 2003 15:04:56 -0500 Subject: Suggesting user names In-Reply-To: References: <20030327164407.GB29163@somanetworks.com> Message-ID: <20030327200456.GB9125@somanetworks.com> On Thu, Mar 27, 2003 at 01:27:23PM -0500, David Miller wrote: ... > > Just FYI, 2.17.1 and up already do this. :) > > Although having a similar patch available for 2.16.x would probably be > enjoyed by a lot of users whose management won't let them upgrade to an > "unstable" version. :) Okay, thanks for the info. I'll take a look to see if this is easily back-portable, and if so I will do so and submit a patch. Otherwise, I'll submit my current patch. Chris -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 232 bytes Desc: not available URL: From gerv at mozilla.org Thu Mar 27 21:39:36 2003 From: gerv at mozilla.org (Gervase Markham) Date: Thu, 27 Mar 2003 21:39:36 +0000 Subject: Suggesting user names In-Reply-To: References: <20030327164407.GB29163@somanetworks.com> Message-ID: <3E836F98.8010602@mozilla.org> David Miller wrote: > Just FYI, 2.17.1 and up already do this. :) That's not quite true; I think Chris's patch does more fuzzy matching than ours, which only matches an exact substring. Chris? Gerv From cmason at somanetworks.com Fri Mar 28 13:18:20 2003 From: cmason at somanetworks.com (Chris L. Mason) Date: Fri, 28 Mar 2003 08:18:20 -0500 Subject: Suggesting user names In-Reply-To: <3E836F98.8010602@mozilla.org> References: <20030327164407.GB29163@somanetworks.com> <3E836F98.8010602@mozilla.org> Message-ID: <20030328131820.GC9125@somanetworks.com> On Thu, Mar 27, 2003 at 09:39:36PM +0000, Gervase Markham wrote: > David Miller wrote: > >Just FYI, 2.17.1 and up already do this. :) > > That's not quite true; I think Chris's patch does more fuzzy matching > than ours, which only matches an exact substring. Chris? > Yes, that's correct, although I'm not happy with the method I'm using right now. The patch is very short, so I've attached it. (Briefly, the DBNameToIdAndCheck function will show the list of suggestions if the name doesn't match, if the "suggest-users" parameter is on.) As you can see, I'm just using "substr($name,2,4)" to match the login name or $name to match the fullname. I don't really like this though. Maybe some kind of regex would be better, although then I worry about the performance. Also, I didn't really follow the whole SQL standards thread, but it *would* be a lot nicer to use something like: SELECT DISTINCT login_name, realname FROM profiles WHERE login_name LIKE :1 OR realname LIKE :2 Then bind substr($name,2,4) for :1 and $name for :2 in the execution. (Although, I realize mysql doesn't support :N, so I guess it would have to be '?') Is it permitted to use the DBI functions directly instead of using the SendSQL stuff? Chris -------------- next part -------------- ===== defparams.pl 1.1 vs edited ===== *** /tmp/defparams.pl-1.1-4134 Thu Feb 27 06:09:10 2003 --- edited/defparams.pl Thu Mar 27 07:47:52 2003 *************** *** 607,610 **** --- 607,618 ---- "t" , '1000'); + DefParam("suggest-users" , + "This option will perform loose name matching in case a username + entered does not match, and will provide the user with a list so + that they might find the username they were looking for. This is + disabled by default for security and privacy reasons." , + "b", + 0); + 1; ===== globals.pl 1.2 vs edited ===== *** /tmp/globals.pl-1.2-4134 Mon Mar 10 08:51:14 2003 --- edited/globals.pl Fri Mar 28 05:07:03 2003 *************** *** 973,982 **** return $result; } ! $name = html_quote($name); ! ThrowUserError("The name $name is not a valid username. Either you misspelled it, or the person has not registered for a Bugzilla account."); } # Use trick_taint() when you know that there is no way that the data --- 973,1009 ---- return $result; } ! if (Param('suggest-users')) { ! SuggestUsers($name); ! PutFooter(); ! exit(0); ! } else { ! $name = html_quote($name); ! ThrowUserError("The name $name is not a valid username. Either you misspelled it, or the person has not registered for a Bugzilla account."); + } + } + + sub SuggestUsers(@) { + my ($name) = (@_); + + PushGlobalSQLState(); + SendSQL("SELECT DISTINCT login_name, realname FROM profiles WHERE " . + "login_name LIKE " . SqlQuote('%' . substr($name,2,4) . '%') . + " OR realname LIKE " . SqlQuote('%' . $name . '%')); + + print "

The name you requested could not be found. Here are " . + "some possible alternatives:\n

\n" . + "\n"; + + while((my $login, my $full) = FetchSQLData()) { + print "\n"; + } + + print "
$login$full
\n

Please back up and try again.\n"; + + PopGlobalSQLState(); } # Use trick_taint() when you know that there is no way that the data -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 232 bytes Desc: not available URL: From bugreport at peshkin.net Fri Mar 28 13:26:13 2003 From: bugreport at peshkin.net (Joel Peshkin) Date: Fri, 28 Mar 2003 05:26:13 -0800 Subject: Suggesting user names In-Reply-To: <20030328131820.GC9125@somanetworks.com> References: <20030327164407.GB29163@somanetworks.com> <3E836F98.8010602@mozilla.org> <20030328131820.GC9125@somanetworks.com> Message-ID: <3E844D75.80807@peshkin.net> Chris L. Mason wrote: >On Thu, Mar 27, 2003 at 09:39:36PM +0000, Gervase Markham wrote: > > >>David Miller wrote: >> >> >>>Just FYI, 2.17.1 and up already do this. :) >>> >>> >>That's not quite true; I think Chris's patch does more fuzzy matching >>than ours, which only matches an exact substring. Chris? >> >> >> > >Yes, that's correct, although I'm not happy with the method I'm using >right now. > >The patch is very short, so I've attached it. (Briefly, the >DBNameToIdAndCheck function will show the list of suggestions if the name >doesn't match, if the "suggest-users" parameter is on.) > > > > The idea of doing a fuzzy match is kind of cool, but slipping it into DBNameToId... is not the right place for it. This needs to be an additional option in the generic user matching code (complete with confirmation). See bug 162990 From cmason at somanetworks.com Fri Mar 28 14:37:56 2003 From: cmason at somanetworks.com (Chris L. Mason) Date: Fri, 28 Mar 2003 09:37:56 -0500 Subject: Suggesting user names In-Reply-To: <3E844D75.80807@peshkin.net> References: <20030327164407.GB29163@somanetworks.com> <3E836F98.8010602@mozilla.org> <20030328131820.GC9125@somanetworks.com> <3E844D75.80807@peshkin.net> Message-ID: <20030328143756.GE13157@somanetworks.com> On Fri, Mar 28, 2003 at 05:26:13AM -0800, Joel Peshkin wrote: ... > > > The idea of doing a fuzzy match is kind of cool, but slipping it into > DBNameToId... is not the right place for it. This needs to be an > additional option in the generic user matching code (complete with > confirmation). > > See bug 162990 > I think you're right. Unfortunately my time is very limited so I may not be able to contribute too much right now. We may just live with this hack for now, then switch to using this feature if it makes it into 2.18.x as it is now targetted. At that point if it's not doing fuzzy-matching, I might try to add that to it. Thanks, Chris -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 232 bytes Desc: not available URL: From myk at mozilla.org Sat Mar 29 02:47:38 2003 From: myk at mozilla.org (Myk Melez) Date: Fri, 28 Mar 2003 18:47:38 -0800 Subject: Suggesting user names In-Reply-To: <20030327164407.GB29163@somanetworks.com> References: <20030327164407.GB29163@somanetworks.com> Message-ID: <3E85094A.5050001@mozilla.org> Chris L. Mason wrote: >Unfortunately mysql doesn't >appear to have a SOUNDEX() SQL function, which would make this easier. > > Apparently it does: http://www.mysql.com/doc/en/String_functions.html#IDX1188 The MySQL docs don't say when it was introduced--the only mention of it in the change log is the note that the "foo SOUNDS LIKE bar" syntax (equivalent to "SOUNDEX(foo)=SOUNDEX(bar)") was added to 4.1--but this article from 2001 indicates it's been around for a while: http://groups.google.com/groups?q=mysql+soundex&hl=en&lr=&ie=UTF-8&selm=Pine.LNX.4.31.0104082048320.18870-100000%40zaius.poa&rnum=1 -myk From jason at pyeron.com Sat Mar 29 04:45:47 2003 From: jason at pyeron.com (Jason Pyeron) Date: Fri, 28 Mar 2003 23:45:47 -0500 (EST) Subject: Suggesting user names In-Reply-To: <3E85094A.5050001@mozilla.org> Message-ID: mysql has supported this for a while but soundex does not function well for email addr or when the first letter is wrong you want to use a lcs function. I have attached some code, it is from a long time ago but it works well play with lcstest, you will find it very cool. http://www1.ics.uci.edu/~eppstein/161/960229.html -jason pyeron On Fri, 28 Mar 2003, Myk Melez wrote: Chris L. Mason wrote: >Unfortunately mysql doesn't >appear to have a SOUNDEX() SQL function, which would make this easier. > > Apparently it does: http://www.mysql.com/doc/en/String_functions.html#IDX1188 The MySQL docs don't say when it was introduced--the only mention of it in the change log is the note that the "foo SOUNDS LIKE bar" syntax (equivalent to "SOUNDEX(foo)=SOUNDEX(bar)") was added to 4.1--but this article from 2001 indicates it's been around for a while: http://groups.google.com/groups?q=mysql+soundex&hl=en&lr=&ie=UTF-8&selm=Pine.LNX.4.31.0104082048320.18870-100000%40zaius.poa&rnum=1 -myk ---- To view or change your list settings, click here: -- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- - - - Jason Pyeron http://www.pyerotechnics.com - - Owner & Lead Pyerotechnics Development, Inc. - - +1 410 808 6646 (c) 500 West University Parkway #1S - - +1 410 467 2266 (f) Baltimore, Maryland 21210-3253 - - - -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- This message is for the designated recipient only and may contain privileged, proprietary, or otherwise private information. If you have received it in error, purge the message from your system and notify the sender immediately. Any other use of the email by you is prohibited. -------------- next part -------------- /* * * (c) 2000 Pyerotechnics Development, Inc. * */ #include #include "lcs.h" //http://www1.ics.uci.edu/~eppstein/161/960229.html inline int max(int a,int b){return a>b?a:b;} int sl(char* s) { char* os=s; if (s) while (*s) ++s; return (int)(s-os); } int lcs0(char* A, int m, char* B, int n) { int* X; int* Y; int i; int j; //if (m=0; --i) { for(j=n; j>=0; --j) X[j]=(i==m || j==n)?0:(A[i]==B[j])?1+Y[j+1]:max(Y[j],X[j+1]); for(j=0; j<=n; ++j) Y[j]=X[j]; } i=X[0]; free(X); free(Y); return i; } /* lcs with weights, it is front heavy */ int lcsw0(char* A, int m, char* B, int n) { int* X; int* Y; int i; int j; int L=1000000000; //needs to be 8388736 or larger for 256 size strings int k=m+n-1; int M=m*n; int S=L/M; #define T(i,j) ((i)+(j)-1) #define W(i) (k-(i)+1) //if (m=0; --i) { for(j=n; j>=0; --j) X[j]=(i==m || j==n)?0:(A[i]==B[j])?1+Y[j+1]:max(Y[j],X[j+1]); for(j=0; j<=n; ++j) Y[j]=X[j]; } i=X[0]; free(X); free(Y); return (m==0)?1.0:(0.0+i)/(0.0+m); } int lcs(char * A, char * B) { return lcs0(A,sl(A),B,sl(B)); } int lcsw(char * A, char * B) { return lcsw0(A,sl(A),B,sl(B)); } double lcsp(char * A, char * B) { return lcsp0(A,sl(A),B,sl(B)); } -------------- next part -------------- /* * * (c) 2000 Pyerotechnics Development, Inc. * */ #ifndef _lcs_h_ #define _lcs_h_ int lcs0(char* A, int m, char* B, int n); int lcsw0(char* A, int m, char* B, int n); double lcsp0(char* A, int m, char* B, int n); int lcs(char * A, char * B); int lcsw(char * A, char * B); double lcsp(char * A, char * B); #endif -------------- next part -------------- /* * * (c) 2000 Pyerotechnics Development, Inc. * */ /* gcc \ -DMYSQL_SERVER -DDEFAULT_MYSQL_HOME="\"/\"" -DDATADIR="\"/var/lib/mysql\"" \ -DSHAREDIR="\"/usr/share/mysql\"" -DHAVE_CONFIG_H -DDBUG_OFF \ \ -I/usr/src/redhat/BUILD/mysql-3.23.43/bdb/build_unix \ -I/usr/src/redhat/BUILD/mysql-3.23.43/innobase/include \ -I/usr/src/redhat/BUILD/mysql-3.23.43/include \ -I/usr/src/redhat/BUILD/mysql-3.23.43/regex \ -I/usr/src/redhat/BUILD/mysql-3.23.43/sql \ -I/usr/src/redhat/BUILD/mysql-3.23.43/include \ -I/usr/src/redhat/BUILD/mysql-3.23.43 \ \ -O3 \ \ -felide-constructors \ -fno-exceptions \ -fno-rtti \ -fno-implicit-templates \ -fno-exceptions \ -fno-rtti \ \ -shared \ \ -o myfile.so src src src obj src obj etc */ #ifdef STANDARD #include #include #else #include #include #endif #include #include #include // To get strmov() #ifdef HAVE_DLOPEN #include "lcs.h" //CREATE FUNCTION diff RETURNS INTEGER SONAME "lcs.MYSQL.so"; long long diff(UDF_INIT * initid, UDF_ARGS * args, char* is_null, char* err) { if (args && args->args[0] && args->args[1]) { *is_null=0; return lcs0(args->args[0], args->lengths[0], args->args[1], args->lengths[1]); } else { *is_null=1; return 0; } } //CREATE FUNCTION diffw RETURNS INTEGER SONAME "lcs.MYSQL.so"; long long diffw(UDF_INIT * initid, UDF_ARGS * args, char* is_null, char* err) { if (args && args->args[0] && args->args[1]) { *is_null=0; return lcsw0(args->args[0], args->lengths[0], args->args[1], args->lengths[1]); } else { *is_null=1; return 0; } } //CREATE FUNCTION diffp RETURNS REAL SONAME "lcs.MYSQL.so"; double diffp(UDF_INIT * initid, UDF_ARGS * args, char* is_null, char* err) { if (args && args->args[0] && args->args[1]) { *is_null=0; return lcsp0(args->args[0], args->lengths[0], args->args[1], args->lengths[1]); } else { *is_null=1; return 0; } } my_bool init(UDF_INIT * initid, UDF_ARGS * args, char* msg) { uint i; if (args->arg_count!=2) { strcpy(msg,"diff must have 2 arguments"); return 1; } //force all arguments to strings. for (i=0 ; i < args->arg_count; i++) args->arg_type[i]=STRING_RESULT; initid->maybe_null=1; // The result may be null return 0; } my_bool diff_init(UDF_INIT * initid, UDF_ARGS * args, char* msg) { return init(initid, args, msg); } my_bool diffw_init(UDF_INIT * initid, UDF_ARGS * args, char* msg) { return init(initid, args, msg); } my_bool diffp_init(UDF_INIT * initid, UDF_ARGS * args, char* msg) { initid->decimals=6; // We want 2 decimals in the result initid->max_length=8; // 1 digit + . + 2 decimals return init(initid, args, msg); } #endif /* HAVE_DLOPEN */ -------------- next part -------------- /* * * (c) 2000 Pyerotechnics Development, Inc. * */ #include "lcs.h" int main(int argc, char* argv[]) { char* A=argc>1?argv[1]:""; char* B=argc>2?argv[2]:""; char* C=argc>3?argv[3]:""; // double m=max(sl(A),sl(B)); if (argc>3) { printf("lcs('%s','%s')=%d %d=lcs('%s','%s')\n",A,B,lcs(A,B),lcs(A,C),A,C); printf("lcsw('%s','%s')=%d %d=lcsw('%s','%s')\n",A,B,lcsw(A,B),lcsw(A,C),A,C); printf("lcsp('%s','%s')=%f %f=lcsp('%s','%s')\n",A,B,lcsp(A,B),lcsp(A,C),A,C); } else { printf("lcs('%s','%s')=%d\n",A,B,lcs(A,B)); printf("lcsw('%s','%s')=%d\n",A,B,lcsw(A,B)); printf("lcsp('%s','%s')=%f\n",A,B,lcsp(A,B)); } return lcs(A,B); } -------------- next part -------------- /* Copyright Abandoned 1996 TCX DataKonsult AB & Monty Program KB & Detron HB This file is public domain and comes with NO WARRANTY of any kind */ /* This is the main include file that should included 'first' in every C file. */ #ifndef _global_h #define _global_h #if defined( __EMX__) && !defined( MYSQL_SERVER) /* moved here to use below VOID macro redefinition */ #define INCL_BASE #define INCL_NOPMAPI #include #endif /* __EMX__ */ #if defined(_WIN32) || defined(_WIN64) || defined(__WIN32__) || defined(WIN32) #include #else #include #endif #if defined(__cplusplus) #if defined(inline) #undef inline /* fix configure problem */ #endif #endif /* _cplusplus */ /* Fix problem with S_ISLNK() on Linux */ #if defined(HAVE_LINUXTHREADS) #undef _GNU_SOURCE #define _GNU_SOURCE 1 #endif /* The client defines this to avoid all thread code */ #if defined(UNDEF_THREADS_HACK) #undef THREAD #undef HAVE_mit_thread #undef HAVE_LINUXTHREADS #undef HAVE_UNIXWARE7_THREADS #endif #ifdef HAVE_THREADS_WITHOUT_SOCKETS /* MIT pthreads does not work with unix sockets */ #undef HAVE_SYS_UN_H #endif #define __EXTENSIONS__ 1 /* We want some extension */ #ifndef __STDC_EXT__ #define __STDC_EXT__ 1 /* To get large file support on hpux */ #endif #if defined(THREAD) && !defined(__WIN__) #ifndef _POSIX_PTHREAD_SEMANTICS #define _POSIX_PTHREAD_SEMANTICS /* We want posix threads */ #endif /* was #if defined(HAVE_LINUXTHREADS) || defined(HAVE_DEC_THREADS) || defined(HPUX) */ #if !defined(SCO) #define _REENTRANT 1 /* Some thread libraries require this */ #endif #if !defined(_THREAD_SAFE) && !defined(_AIX) #define _THREAD_SAFE /* Required for OSF1 */ #endif #ifndef HAVE_mit_thread #ifdef HAVE_UNIXWARE7_THREADS #include #else #include /* AIX must have this included first */ #endif /* HAVE_UNIXWARE7_THREADS */ #endif /* HAVE_mit_thread */ #if !defined(SCO) && !defined(_REENTRANT) #define _REENTRANT 1 /* Threads requires reentrant code */ #endif #endif /* THREAD */ /* Go around some bugs in different OS and compilers */ #ifdef _AIX /* By soren at t.dk */ #define _H_STRINGS #define _SYS_STREAM_H #define _AIX32_CURSES #endif #ifdef HAVE_BROKEN_SNPRINTF /* HPUX 10.20 don't have this defined */ #undef HAVE_SNPRINTF #endif #ifdef HAVE_BROKEN_PREAD /* These doesn't work on HPUX 11.x */ #undef HAVE_PREAD #undef HAVE_PWRITE #endif #ifdef UNDEF_HAVE_GETHOSTBYNAME_R /* For OSF4.x */ #undef HAVE_GETHOSTBYNAME_R #endif #ifdef UNDEF_HAVE_INITGROUPS /* For AIX 4.3 */ #undef HAVE_INITGROUPS #endif /* Fix a bug in gcc 2.8.0 on IRIX 6.2 */ #if SIZEOF_LONG == 4 && defined(__LONG_MAX__) #undef __LONG_MAX__ /* Is a longlong value in gcc 2.8.0 ??? */ #define __LONG_MAX__ 2147483647 #endif /* egcs 1.1.2 has a problem with memcpy on Alpha */ #if defined(__GNUC__) && defined(__alpha__) && ! (__GNUC__ > 2 || (__GNUC__ == 2 && __GNUC_MINOR__ >= 95)) #define BAD_MEMCPY #endif /* In Linux-alpha we have atomic.h if we are using gcc */ #if defined(HAVE_LINUXTHREADS) && defined(__GNUC__) && defined(__alpha__) && (__GNUC__ > 2 || ( __GNUC__ == 2 && __GNUC_MINOR__ >= 95)) && !defined(HAVE_ATOMIC_ADD) #define HAVE_ATOMIC_ADD #define HAVE_ATOMIC_SUB #endif /* In Linux-ia64 including atomic.h will give us an error */ #if (defined(HAVE_LINUXTHREADS) && defined(__GNUC__) && defined(__ia64__)) || !defined(THREAD) #undef HAVE_ATOMIC_ADD #undef HAVE_ATOMIC_SUB #endif #if defined(_lint) && !defined(lint) #define lint #endif #if SIZEOF_LONG_LONG > 4 && !defined(_LONG_LONG) #define _LONG_LONG 1 /* For AIX string library */ #endif #ifndef stdin #include #endif #ifdef HAVE_STDLIB_H #include #endif #ifdef HAVE_STDDEF_H #include #endif #include #ifdef HAVE_LIMITS_H #include #endif #ifdef HAVE_FLOAT_H #include #endif #ifdef HAVE_SYS_TYPES_H #include #endif #ifdef HAVE_FCNTL_H #include #endif #ifdef HAVE_SYS_TIMEB_H #include /* Avoid warnings on SCO */ #endif #if TIME_WITH_SYS_TIME # include # include #else # if HAVE_SYS_TIME_H # include # else # include # endif #endif /* TIME_WITH_SYS_TIME */ #ifdef HAVE_UNISTD_H #include #endif #if defined(__cplusplus) && defined(NO_CPLUSPLUS_ALLOCA) #undef HAVE_ALLOCA #undef HAVE_ALLOCA_H #endif #ifdef HAVE_ALLOCA_H #include #endif #ifdef HAVE_ATOMIC_ADD #define __SMP__ #include #endif /* Go around some bugs in different OS and compilers */ #if defined(_HPUX_SOURCE) && defined(HAVE_SYS_STREAM_H) #include /* HPUX 10.20 defines ulong here. UGLY !!! */ #define HAVE_ULONG #endif #ifdef DONT_USE_FINITE /* HPUX 11.x has is_finite() */ #undef HAVE_FINITE #endif /* We can not live without these */ #define USE_MYFUNC 1 /* Must use syscall indirection */ #define MASTER 1 /* Compile without unireg */ #define ENGLISH 1 /* Messages in English */ #define POSIX_MISTAKE 1 /* regexp: Fix stupid spec error */ #define USE_REGEX 1 /* We want the use the regex library */ /* Do not define for ultra sparcs */ #define USE_BMOVE512 1 /* Use this unless the system bmove is faster */ /* Paranoid settings. Define I_AM_PARANOID if you are paranoid */ #ifdef I_AM_PARANOID #define DONT_ALLOW_USER_CHANGE 1 #define DONT_USE_MYSQL_PWD 1 #endif /* #define USE_some_charset 1 was deprecated by changes to configure */ /* my_ctype my_to_upper, my_to_lower, my_sort_order gain theit right value */ /* automagically during configuration */ /* Does the system remember a signal handler after a signal ? */ #ifndef HAVE_BSD_SIGNALS #define DONT_REMEMBER_SIGNAL #endif /* Define void to stop lint from generating "null effekt" comments */ #ifndef DONT_DEFINE_VOID #ifdef _lint int __void__; #define VOID(X) (__void__ = (int) (X)) #else #undef VOID #define VOID(X) (X) #endif #endif /* DONT_DEFINE_VOID */ #if defined(_lint) || defined(FORCE_INIT_OF_VARS) #define LINT_INIT(var) var=0 /* No uninitialize-warning */ #else #define LINT_INIT(var) #endif /* Define som useful general macros */ #if defined(__cplusplus) && defined(__GNUC__) #define max(a, b) ((a) >? (b)) #define min(a, b) ((a) (b) ? (a) : (b)) #define min(a, b) ((a) < (b) ? (a) : (b)) #endif #if defined(__EMX__) || !defined(HAVE_UINT) typedef unsigned int uint; typedef unsigned short ushort; #endif #define sgn(a) (((a) < 0) ? -1 : ((a) > 0) ? 1 : 0) #define swap(t,a,b) { register t dummy; dummy = a; a = b; b = dummy; } #define test(a) ((a) ? 1 : 0) #define set_if_bigger(a,b) { if ((a) < (b)) (a)=(b); } #define set_if_smaller(a,b) { if ((a) > (b)) (a)=(b); } #define test_all_bits(a,b) (((a) & (b)) == (b)) #define array_elements(A) ((uint) (sizeof(A)/sizeof(A[0]))) #ifndef HAVE_RINT #define rint(A) floor((A)+0.5) #endif /* Define som general constants */ #ifndef TRUE #define TRUE (1) /* Logical true */ #define FALSE (0) /* Logical false */ #endif #if defined(__GNUC__) #define function_volatile volatile #define my_reinterpret_cast(A) reinterpret_cast #define my_const_cast(A) const_cast #elif !defined(my_reinterpret_cast) #define my_reinterpret_cast(A) (A) #define my_const_cast(A) (A) #endif #if !defined(__attribute__) && (defined(__cplusplus) || !defined(__GNUC__) || __GNUC__ == 2 && __GNUC_MINOR__ < 8) #define __attribute__(A) #endif /* From old s-system.h */ /* Support macros for non ansi & other old compilers. Since such things are no longer supported we do nothing. We keep then since some of our code may still be needed to upgrade old customers. */ #define _VARARGS(X) X #define _STATIC_VARARGS(X) X #define _PC(X) X #if defined(DBUG_ON) && defined(DBUG_OFF) #undef DBUG_OFF #endif #if defined(_lint) && !defined(DBUG_OFF) #define DBUG_OFF #endif #include #ifndef DBUG_OFF #define dbug_assert(A) assert(A) #else #define dbug_assert(A) #endif #define MIN_ARRAY_SIZE 0 /* Zero or One. Gcc allows zero*/ #define ASCII_BITS_USED 8 /* Bit char used */ #define NEAR_F /* No near function handling */ /* Some types that is different between systems */ typedef int File; /* File descriptor */ #ifndef Socket_defined typedef int my_socket; /* File descriptor for sockets */ #define INVALID_SOCKET -1 #endif typedef RETSIGTYPE sig_handler; /* Function to handle signals */ typedef void (*sig_return)();/* Returns type from signal */ #if defined(__GNUC__) && !defined(_lint) typedef char pchar; /* Mixed prototypes can take char */ typedef char puchar; /* Mixed prototypes can take char */ typedef char pbool; /* Mixed prototypes can take char */ typedef short pshort; /* Mixed prototypes can take short int */ typedef float pfloat; /* Mixed prototypes can take float */ #else typedef int pchar; /* Mixed prototypes can't take char */ typedef uint puchar; /* Mixed prototypes can't take char */ typedef int pbool; /* Mixed prototypes can't take char */ typedef int pshort; /* Mixed prototypes can't take short int */ typedef double pfloat; /* Mixed prototypes can't take float */ #endif typedef int (*qsort_cmp)(const void *,const void *); #ifdef HAVE_mit_thread #define qsort_t void #undef QSORT_TYPE_IS_VOID #define QSORT_TYPE_IS_VOID #else #define qsort_t RETQSORTTYPE /* Broken GCC cant handle typedef !!!! */ #endif #ifdef HAVE_mit_thread #define size_socket socklen_t /* Type of last arg to accept */ #else #ifdef HAVE_SYS_SOCKET_H #include #endif typedef SOCKET_SIZE_TYPE size_socket; #endif #ifndef SOCKOPT_OPTLEN_TYPE #define SOCKOPT_OPTLEN_TYPE size_socket #endif /* file create flags */ #ifndef O_SHARE #define O_SHARE 0 /* Flag to my_open for shared files */ #ifndef O_BINARY #define O_BINARY 0 /* Flag to my_open for binary files */ #endif #define FILE_BINARY 0 /* Flag to my_fopen for binary streams */ #ifdef HAVE_FCNTL #define HAVE_FCNTL_LOCK #define F_TO_EOF 0L /* Param to lockf() to lock rest of file */ #endif #endif /* O_SHARE */ #ifndef O_TEMPORARY #define O_TEMPORARY 0 #endif #ifndef O_SHORT_LIVED #define O_SHORT_LIVED 0 #endif /* #define USE_RECORD_LOCK */ /* Unsigned types supported by the compiler */ #define UNSINT8 /* unsigned int8 (char) */ #define UNSINT16 /* unsigned int16 */ #define UNSINT32 /* unsigned int32 */ /* General constants */ #define SC_MAXWIDTH 256 /* Max width of screen (for error messages) */ #define FN_LEN 256 /* Max file name len */ #define FN_HEADLEN 253 /* Max length of filepart of file name */ #define FN_EXTLEN 20 /* Max length of extension (part of FN_LEN) */ #define FN_REFLEN 512 /* Max length of full path-name */ #define FN_EXTCHAR '.' #define FN_HOMELIB '~' /* ~/ is used as abbrev for home dir */ #define FN_CURLIB '.' /* ./ is used as abbrev for current dir */ #define FN_PARENTDIR ".." /* Parentdirectory; Must be a string */ #define FN_DEVCHAR ':' #ifndef FN_LIBCHAR #ifdef __EMX__ #define FN_LIBCHAR '\\' #define FN_ROOTDIR "\\" #else #define FN_LIBCHAR '/' #define FN_ROOTDIR "/" #endif #define MY_NFILE 1024 /* This is only used to save filenames */ #endif /* #define EXT_IN_LIBNAME */ /* #define FN_NO_CASE_SENCE */ /* #define FN_UPPER_CASE TRUE */ /* Io buffer size; Must be a power of 2 and a multiple of 512. May be smaller what the disk page size. This influences the speed of the isam btree library. eg to big to slow. */ #define IO_SIZE 4096 /* How much overhead does malloc have. The code often allocates something like 1024-MALLOC_OVERHEAD bytes */ #ifdef SAFEMALLOC #define MALLOC_OVERHEAD (8+24+4) #else #define MALLOC_OVERHEAD 8 #endif /* get memory in huncs */ #define ONCE_ALLOC_INIT (uint) (4096-MALLOC_OVERHEAD) /* Typical record cash */ #define RECORD_CACHE_SIZE (uint) (64*1024-MALLOC_OVERHEAD) /* Typical key cash */ #define KEY_CACHE_SIZE (uint) (8*1024*1024-MALLOC_OVERHEAD) /* Some things that this system doesn't have */ #define ONLY_OWN_DATABASES /* We are using only databases by monty */ #define NO_PISAM /* Not needed anymore */ #define NO_MISAM /* Not needed anymore */ #define NO_HASH /* Not needed anymore */ #ifdef __WIN__ #define NO_DIR_LIBRARY /* Not standar dir-library */ #define USE_MY_STAT_STRUCT /* For my_lib */ #endif /* Some things that this system does have */ #ifndef HAVE_ITOA #define USE_MY_ITOA /* There is no itoa */ #endif /* Some defines of functions for portability */ #ifndef HAVE_ATOD #define atod atof #endif #ifdef USE_MY_ATOF #define atof my_atof extern void init_my_atof(void); extern double my_atof(const char*); #endif #undef remove /* Crashes MySQL on SCO 5.0.0 */ #ifndef __WIN__ #define closesocket(A) close(A) #ifndef ulonglong2double #define ulonglong2double(A) ((double) (A)) #define my_off_t2double(A) ((double) (A)) #endif #endif #ifndef offsetof #define offsetof(TYPE, MEMBER) ((size_t) &((TYPE *)0)->MEMBER) #endif #define ulong_to_double(X) ((double) (ulong) (X)) #define SET_STACK_SIZE(X) /* Not needed on real machines */ #if !defined(HAVE_mit_thread) && !defined(HAVE_STRTOK_R) #define strtok_r(A,B,C) strtok((A),(B)) #endif #ifdef HAVE_LINUXTHREADS /* #define pthread_sigmask(A,B,C) sigprocmask((A),(B),(C)) */ /* #define sigset(A,B) signal((A),(B)) */ #endif /* Remove some things that mit_thread break or doesn't support */ #if defined(HAVE_mit_thread) && defined(THREAD) #undef HAVE_PREAD #undef HAVE_REALPATH #undef HAVE_MLOCK #undef HAVE_TEMPNAM /* Use ours */ #undef HAVE_PTHREAD_SETPRIO #undef HAVE_FTRUNCATE #undef HAVE_READLINK #endif /* This is from the old m-machine.h file */ #if SIZEOF_LONG_LONG > 4 #define HAVE_LONG_LONG 1 #endif #if defined(HAVE_LONG_LONG) && !defined(LONGLONG_MIN) #define LONGLONG_MIN ((long long) 0x8000000000000000LL) #define LONGLONG_MAX ((long long) 0x7FFFFFFFFFFFFFFFLL) #endif #if SIZEOF_LONG == 4 #define INT_MIN32 (long) 0x80000000L #define INT_MAX32 (long) 0x7FFFFFFFL #define INT_MIN24 ((long) 0xff800000L) #define INT_MAX24 0x007fffffL #define INT_MIN16 ((short int) 0x8000) #define INT_MAX16 0x7FFF #define INT_MIN8 ((char) 0x80) #define INT_MAX8 ((char) 0x7F) #else /* Probably Alpha */ #define INT_MIN32 ((long) (int) 0x80000000) #define INT_MAX32 ((long) (int) 0x7FFFFFFF) #define INT_MIN24 ((long) (int) 0xff800000) #define INT_MAX24 ((long) (int) 0x007fffff) #define INT_MIN16 ((short int) 0xffff8000) #define INT_MAX16 ((short int) 0x00007FFF) #endif /* From limits.h instead */ #ifndef DBL_MIN #define DBL_MIN 4.94065645841246544e-324 #define FLT_MIN ((float)1.40129846432481707e-45) #endif #ifndef DBL_MAX #define DBL_MAX 1.79769313486231470e+308 #define FLT_MAX ((float)3.40282346638528860e+38) #endif /* Max size that must be added to a so that we know Size to make adressable obj. */ typedef long my_ptrdiff_t; #define MY_ALIGN(A,L) (((A) + (L) - 1) & ~((L) - 1)) #define ALIGN_SIZE(A) MY_ALIGN((A),sizeof(double)) /* Size to make adressable obj. */ #define ALIGN_PTR(A, t) ((t*) MY_ALIGN((A),sizeof(t))) /* Offset of filed f in structure t */ #define OFFSET(t, f) ((size_t)(char *)&((t *)0)->f) #define ADD_TO_PTR(ptr,size,type) (type) ((byte*) (ptr)+size) #define PTR_BYTE_DIFF(A,B) (my_ptrdiff_t) ((byte*) (A) - (byte*) (B)) #define NullS (char *) 0 /* Nowdays we do not support MessyDos */ #ifndef NEAR #define NEAR /* Who needs segments ? */ #define FAR /* On a good machine */ #define HUGE_PTR #endif #ifndef STDCALL #define STDCALL #endif /* Typdefs for easyier portability */ #if defined(VOIDTYPE) typedef void *gptr; /* Generic pointer */ #else typedef char *gptr; /* Generic pointer */ #endif #ifndef HAVE_INT_8_16_32 typedef char int8; /* Signed integer >= 8 bits */ typedef short int16; /* Signed integer >= 16 bits */ #endif #ifndef HAVE_UCHAR typedef unsigned char uchar; /* Short for unsigned char */ #endif typedef unsigned char uint8; /* Short for unsigned integer >= 8 bits */ typedef unsigned short uint16; /* Short for unsigned integer >= 16 bits */ #if SIZEOF_INT == 4 #ifndef HAVE_INT_8_16_32 typedef int int32; #endif typedef unsigned int uint32; /* Short for unsigned integer >= 32 bits */ #elif SIZEOF_LONG == 4 #ifndef HAVE_INT_8_16_32 typedef long int32; #endif typedef unsigned long uint32; /* Short for unsigned integer >= 32 bits */ #else error "Neither int or long is of 4 bytes width" #endif #if !defined(HAVE_ULONG) && !defined(HAVE_LINUXTHREADS) && !defined(__USE_MISC) typedef unsigned long ulong; /* Short for unsigned long */ #endif #ifndef longlong_defined #if defined(HAVE_LONG_LONG) && SIZEOF_LONG != 8 typedef unsigned long long ulonglong; /* ulong or unsigned long long */ typedef long long longlong; #else typedef unsigned long ulonglong; /* ulong or unsigned long long */ typedef long longlong; #endif #endif #ifdef USE_RAID /* The following is done with a if to not get problems with pre-processors with late define evaluation */ #if SIZEOF_OFF_T == 4 #define SYSTEM_SIZEOF_OFF_T 4 #else #define SYSTEM_SIZEOF_OFF_T 8 #endif #undef SIZEOF_OFF_T #define SIZEOF_OFF_T 8 #else #define SYSTEM_SIZEOF_OFF_T SIZEOF_OFF_T #endif /* USE_RAID */ #if SIZEOF_OFF_T > 4 typedef ulonglong my_off_t; #else typedef unsigned long my_off_t; #endif #define MY_FILEPOS_ERROR (~(my_off_t) 0) #ifndef __WIN__ typedef off_t os_off_t; #endif typedef uint8 int7; /* Most effective integer 0 <= x <= 127 */ typedef short int15; /* Most effective integer 0 <= x <= 32767 */ typedef char *my_string; /* String of characters */ typedef unsigned long size_s; /* Size of strings (In string-funcs) */ typedef int myf; /* Type of MyFlags in my_funcs */ #ifndef byte_defined typedef char byte; /* Smallest addressable unit */ #endif typedef char my_bool; /* Small bool */ #if !defined(bool) && !defined(bool_defined) && (!defined(HAVE_BOOL) || !defined(__cplusplus)) typedef char bool; /* Ordinary boolean values 0 1 */ #endif /* Macros for converting *constants* to the right type */ #define INT8(v) (int8) (v) #define INT16(v) (int16) (v) #define INT32(v) (int32) (v) #define MYF(v) (myf) (v) /* Defines to make it possible to prioritize register assignments. No longer needed with moder compilers */ #ifndef USING_X #define reg1 register #define reg2 register #define reg3 register #define reg4 register #define reg5 register #define reg6 register #define reg7 register #define reg8 register #define reg9 register #define reg10 register #define reg11 register #define reg12 register #define reg13 register #define reg14 register #define reg15 register #define reg16 register #endif /* Defines for time function */ #define SCALE_SEC 100 #define SCALE_USEC 10000 #define MY_HOW_OFTEN_TO_ALARM 2 /* How often we want info on screen */ #define MY_HOW_OFTEN_TO_WRITE 1000 /* How often we want info on screen */ /* ** Define-funktions for reading and storing in machine independent format ** (low byte first) */ /* Optimized store functions for Intel x86 */ #ifdef __i386__ #define sint2korr(A) (*((int16 *) (A))) #define sint3korr(A) ((int32) ((((uchar) (A)[2]) & 128) ? \ (((uint32) 255L << 24) | \ (((uint32) (uchar) (A)[2]) << 16) |\ (((uint32) (uchar) (A)[1]) << 8) | \ ((uint32) (uchar) (A)[0])) : \ (((uint32) (uchar) (A)[2]) << 16) |\ (((uint32) (uchar) (A)[1]) << 8) | \ ((uint32) (uchar) (A)[0]))) #define sint4korr(A) (*((long *) (A))) #define uint2korr(A) (*((uint16 *) (A))) #define uint3korr(A) (long) (*((unsigned long *) (A)) & 0xFFFFFF) #define uint4korr(A) (*((unsigned long *) (A))) #define uint5korr(A) ((ulonglong)(((uint32) ((uchar) (A)[0])) +\ (((uint32) ((uchar) (A)[1])) << 8) +\ (((uint32) ((uchar) (A)[2])) << 16) +\ (((uint32) ((uchar) (A)[3])) << 24)) +\ (((ulonglong) ((uchar) (A)[4])) << 32)) #define uint8korr(A) (*((ulonglong *) (A))) #define sint8korr(A) (*((longlong *) (A))) #define int2store(T,A) *((uint16*) (T))= (uint16) (A) #define int3store(T,A) { *(T)= (uchar) ((A));\ *(T+1)=(uchar) (((uint) (A) >> 8));\ *(T+2)=(uchar) (((A) >> 16)); } #define int4store(T,A) *((long *) (T))= (long) (A) #define int5store(T,A) { *(T)= (uchar)((A));\ *((T)+1)=(uchar) (((A) >> 8));\ *((T)+2)=(uchar) (((A) >> 16));\ *((T)+3)=(uchar) (((A) >> 24)); \ *((T)+4)=(uchar) (((A) >> 32)); } #define int8store(T,A) *((ulonglong *) (T))= (ulonglong) (A) typedef union { double v; long m[2]; } doubleget_union; #define doubleget(V,M) { ((doubleget_union *)&V)->m[0] = *((long*) M); \ ((doubleget_union *)&V)->m[1] = *(((long*) M)+1); } #define doublestore(T,V) { *((long *) T) = ((doubleget_union *)&V)->m[0]; \ *(((long *) T)+1) = ((doubleget_union *)&V)->m[1]; } #define float4get(V,M) { *((long *) &(V)) = *((long*) (M)); } #define float8get(V,M) doubleget((V),(M)) #define float4store(V,M) memcpy((byte*) V,(byte*) (&M),sizeof(float)) #define float8store(V,M) doublestore((V),(M)) #endif /* __i386__ */ #ifndef sint2korr #define sint2korr(A) (int16) (((int16) ((uchar) (A)[0])) +\ ((int16) ((int16) (A)[1]) << 8)) #define sint3korr(A) ((int32) ((((uchar) (A)[2]) & 128) ? \ (((uint32) 255L << 24) | \ (((uint32) (uchar) (A)[2]) << 16) |\ (((uint32) (uchar) (A)[1]) << 8) | \ ((uint32) (uchar) (A)[0])) : \ (((uint32) (uchar) (A)[2]) << 16) |\ (((uint32) (uchar) (A)[1]) << 8) | \ ((uint32) (uchar) (A)[0]))) #define sint4korr(A) (int32) (((int32) ((uchar) (A)[0])) +\ (((int32) ((uchar) (A)[1]) << 8)) +\ (((int32) ((uchar) (A)[2]) << 16)) +\ (((int32) ((int16) (A)[3]) << 24))) #define sint8korr(A) (longlong) uint8korr(A) #define uint2korr(A) (uint16) (((uint16) ((uchar) (A)[0])) +\ ((uint16) ((uchar) (A)[1]) << 8)) #define uint3korr(A) (uint32) (((uint32) ((uchar) (A)[0])) +\ (((uint32) ((uchar) (A)[1])) << 8) +\ (((uint32) ((uchar) (A)[2])) << 16)) #define uint4korr(A) (uint32) (((uint32) ((uchar) (A)[0])) +\ (((uint32) ((uchar) (A)[1])) << 8) +\ (((uint32) ((uchar) (A)[2])) << 16) +\ (((uint32) ((uchar) (A)[3])) << 24)) #define uint5korr(A) ((ulonglong)(((uint32) ((uchar) (A)[0])) +\ (((uint32) ((uchar) (A)[1])) << 8) +\ (((uint32) ((uchar) (A)[2])) << 16) +\ (((uint32) ((uchar) (A)[3])) << 24)) +\ (((ulonglong) ((uchar) (A)[4])) << 32)) #define uint8korr(A) ((ulonglong)(((uint32) ((uchar) (A)[0])) +\ (((uint32) ((uchar) (A)[1])) << 8) +\ (((uint32) ((uchar) (A)[2])) << 16) +\ (((uint32) ((uchar) (A)[3])) << 24)) +\ (((ulonglong) (((uint32) ((uchar) (A)[4])) +\ (((uint32) ((uchar) (A)[5])) << 8) +\ (((uint32) ((uchar) (A)[6])) << 16) +\ (((uint32) ((uchar) (A)[7])) << 24))) <<\ 32)) #define int2store(T,A) { uint def_temp= (uint) (A) ;\ *((uchar*) (T))= (uchar)(def_temp); \ *((uchar*) (T+1))=(uchar)((def_temp >> 8)); } #define int3store(T,A) { /*lint -save -e734 */\ *((T))=(char) ((A));\ *((T)+1)=(char) (((A) >> 8));\ *((T)+2)=(char) (((A) >> 16)); \ /*lint -restore */} #define int4store(T,A) { *(T)=(char) ((A));\ *((T)+1)=(char) (((A) >> 8));\ *((T)+2)=(char) (((A) >> 16));\ *((T)+3)=(char) (((A) >> 24)); } #define int5store(T,A) { *(T)=((A));\ *((T)+1)=(((A) >> 8));\ *((T)+2)=(((A) >> 16));\ *((T)+3)=(((A) >> 24)); \ *((T)+4)=(((A) >> 32)); } #define int8store(T,A) { uint def_temp= (uint) (A), def_temp2= (uint) ((A) >> 32); \ int4store((T),def_temp); \ int4store((T+4),def_temp2); \ } #ifdef WORDS_BIGENDIAN #define float4store(T,A) { *(T)= ((byte *) &A)[3];\ *((T)+1)=(char) ((byte *) &A)[2];\ *((T)+2)=(char) ((byte *) &A)[1];\ *((T)+3)=(char) ((byte *) &A)[0]; } #define float4get(V,M) { float def_temp;\ ((byte*) &def_temp)[0]=(M)[3];\ ((byte*) &def_temp)[1]=(M)[2];\ ((byte*) &def_temp)[2]=(M)[1];\ ((byte*) &def_temp)[3]=(M)[0];\ (V)=def_temp; } #define float8store(T,V) { *(T)= ((byte *) &V)[7];\ *((T)+1)=(char) ((byte *) &V)[6];\ *((T)+2)=(char) ((byte *) &V)[5];\ *((T)+3)=(char) ((byte *) &V)[4];\ *((T)+4)=(char) ((byte *) &V)[3];\ *((T)+5)=(char) ((byte *) &V)[2];\ *((T)+6)=(char) ((byte *) &V)[1];\ *((T)+7)=(char) ((byte *) &V)[0]; } #define float8get(V,M) { double def_temp;\ ((byte*) &def_temp)[0]=(M)[7];\ ((byte*) &def_temp)[1]=(M)[6];\ ((byte*) &def_temp)[2]=(M)[5];\ ((byte*) &def_temp)[3]=(M)[4];\ ((byte*) &def_temp)[4]=(M)[3];\ ((byte*) &def_temp)[5]=(M)[2];\ ((byte*) &def_temp)[6]=(M)[1];\ ((byte*) &def_temp)[7]=(M)[0];\ (V) = def_temp; } #else #define float4get(V,M) memcpy_fixed((byte*) &V,(byte*) (M),sizeof(float)) #define float4store(V,M) memcpy_fixed((byte*) V,(byte*) (&M),sizeof(float)) #define float8get(V,M) doubleget((V),(M)) #define float8store(V,M) doublestore((V),(M)) #endif /* WORDS_BIGENDIAN */ #endif /* sint2korr */ /* Define-funktions for reading and storing in machine format from/to short/long to/from some place in memory V should be a (not register) variable, M is a pointer to byte */ #ifdef WORDS_BIGENDIAN #define ushortget(V,M) { V = (uint16) (((uint16) ((uchar) (M)[1]))+\ ((uint16) ((uint16) (M)[0]) << 8)); } #define shortget(V,M) { V = (short) (((short) ((uchar) (M)[1]))+\ ((short) ((short) (M)[0]) << 8)); } #define longget(V,M) { int32 def_temp;\ ((byte*) &def_temp)[0]=(M)[0];\ ((byte*) &def_temp)[1]=(M)[1];\ ((byte*) &def_temp)[2]=(M)[2];\ ((byte*) &def_temp)[3]=(M)[3];\ (V)=def_temp; } #define ulongget(V,M) { uint32 def_temp;\ ((byte*) &def_temp)[0]=(M)[0];\ ((byte*) &def_temp)[1]=(M)[1];\ ((byte*) &def_temp)[2]=(M)[2];\ ((byte*) &def_temp)[3]=(M)[3];\ (V)=def_temp; } #define shortstore(T,A) { uint def_temp=(uint) (A) ;\ *(T+1)=(char)(def_temp); \ *(T+0)=(char)(def_temp >> 8); } #define longstore(T,A) { *((T)+3)=((A));\ *((T)+2)=(((A) >> 8));\ *((T)+1)=(((A) >> 16));\ *((T)+0)=(((A) >> 24)); } #define doubleget(V,M) memcpy((byte*) &V,(byte*) (M),sizeof(double)) #define doublestore(T,V) memcpy((byte*) (T),(byte*) &V,sizeof(double)) #define longlongget(V,M) memcpy((byte*) &V,(byte*) (M),sizeof(ulonglong)) #define longlongstore(T,V) memcpy((byte*) (T),(byte*) &V,sizeof(ulonglong)) #else #define ushortget(V,M) { V = uint2korr(M); } #define shortget(V,M) { V = sint2korr(M); } #define longget(V,M) { V = sint4korr(M); } #define ulongget(V,M) { V = uint4korr(M); } #define shortstore(T,V) int2store(T,V) #define longstore(T,V) int4store(T,V) #ifndef doubleget #define doubleget(V,M) memcpy_fixed((byte*) &V,(byte*) (M),sizeof(double)) #define doublestore(T,V) memcpy_fixed((byte*) (T),(byte*) &V,sizeof(double)) #endif #define longlongget(V,M) memcpy_fixed((byte*) &V,(byte*) (M),sizeof(ulonglong)) #define longlongstore(T,V) memcpy_fixed((byte*) (T),(byte*) &V,sizeof(ulonglong)) #endif /* WORDS_BIGENDIAN */ /* sprintf does not always return the number of bytes :- */ #ifdef SPRINTF_RETURNS_INT #define my_sprintf(buff,args) sprintf args #else #ifdef SPRINTF_RETURNS_PTR #define my_sprintf(buff,args) ((int)(sprintf args - buff)) #else #define my_sprintf(buff,args) sprintf args,strlen(buff) #endif #endif #ifndef THREAD #define thread_safe_increment(V,L) (V)++ #define thread_safe_add(V,C,L) (V)+=(C) #define thread_safe_sub(V,C,L) (V)-=(C) #define statistic_increment(V,L) (V)++ #define statistic_add(V,C,L) (V)+=(C) #endif #endif /* _global_h */ From bbaetz at acm.org Sat Mar 29 05:06:55 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Sat, 29 Mar 2003 16:06:55 +1100 Subject: Suggesting user names In-Reply-To: <20030328131820.GC9125@somanetworks.com> References: <20030327164407.GB29163@somanetworks.com> <3E836F98.8010602@mozilla.org> <20030328131820.GC9125@somanetworks.com> Message-ID: <20030329050655.GA14050@mango.home> On Fri, Mar 28, 2003 at 08:18:20AM -0500, Chris L. Mason wrote: > Is it permitted to use the DBI functions directly instead of using the > SendSQL stuff? Not in 2.16. In 2.17, thats the preferred way to do things (although we're still discussing quoting styles... Re the soundex suggestion, I don't think that thats approriate, since 'sounds like' for an email address isn't the same as a 'normal' english word. Bradley From bbaetz at acm.org Sat Mar 29 05:19:12 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Sat, 29 Mar 2003 16:19:12 +1100 Subject: SQL call formatting style In-Reply-To: <3E816E68.5090400@mozilla.org> References: <3E80E5FF.1050706@mozilla.org> <3E80ED07.9020508@mozilla.org> <20030326081520.GA1449@mango.home> <3E816E68.5090400@mozilla.org> Message-ID: <20030329051912.GB14050@mango.home> On Wed, Mar 26, 2003 at 09:10:00AM +0000, Gervase Markham wrote: > But, what's the advantage of > qq{FOO > BAR} > over > "FOO > BAR" > ? I find the latter easier to read. well, generally we'd prefer '' (or q{}) because they don't inerpolate. The main difference is that we can include quotes in there, without extra escaping which is harder to read. Not for params (which use ? placeholders), but more for literal strings. Compare: 'SELECT id FROM flags WHERE status=\'?\' ' and q{SELECT id FROM flags WHERE status='?' } Now, we don't have many of those sorts of things (we tend to use 0/1 int flags instead), but we do have some. Flags and tokens are the main examples. (We don't want to use placeholders here, where the search terms are constant. Mainly because using bind vars would preclude the use of histograms - for example there are probably more ? flags than - ones, so the access path could change, but also because if its always constant it makes stuff clearer to read if the argument is right there with the rest of the query) It also looks a bit nicer in editors - you can distinguish it from 'regular' strings, for example based on syntax highlighting. I'm not that attached, although it is sort of the dbi standard. Bradley From gerv at mozilla.org Sat Mar 29 10:42:01 2003 From: gerv at mozilla.org (Gervase Markham) Date: Sat, 29 Mar 2003 10:42:01 +0000 Subject: SQL call formatting style In-Reply-To: <20030329051912.GB14050@mango.home> References: <3E80E5FF.1050706@mozilla.org> <3E80ED07.9020508@mozilla.org> <20030326081520.GA1449@mango.home> <3E816E68.5090400@mozilla.org> <20030329051912.GB14050@mango.home> Message-ID: <3E857879.3000800@mozilla.org> Bradley Baetz wrote: > well, generally we'd prefer '' (or q{}) because they don't inerpolate. Is this for performance reasons? Interpolation isn't bad unless you want to literally use some special characters. How common is that? The advantage of using " " is that you can embed ' without escaping. > The main difference is that we can include quotes in there, without > extra escaping which is harder to read. Not for params (which use ? > placeholders), but more for literal strings. Compare: So, basically it's a trade-off between using: " ", which allow you to embed the common ' without escaping, but if you want literal values of e.g. $, you have to escape those and ' ', which require escaping of embedded single quotes in a few rare cases and q{ }, which I find harder to read (and, perhaps, people with less Perl familiarity might not recognise), but don't require such escaping. As I've said, I find quotes around strings mentally much less jarring, because it's what I expect. q{ } always makes me think of a function call (which it is, really.) We should choose what's most readable for most people in the common case. Obviously, others will chime in, but in my view, that is either '' or " " as appropriate, with a spot of escaping in a small number of places as necessary. > It also looks a bit nicer in editors - you can distinguish it from > 'regular' strings, for example based on syntax highlighting. I'd rather it was highlighted as a regular string actually, because that's what it is :-) Gerv From chicks at chicks.net Sat Mar 29 17:02:58 2003 From: chicks at chicks.net (Christopher Hicks) Date: Sat, 29 Mar 2003 12:02:58 -0500 (EST) Subject: SQL call formatting style In-Reply-To: <3E857879.3000800@mozilla.org> Message-ID: On Sat, 29 Mar 2003, Gervase Markham wrote: > As I've said, I find quotes around strings mentally much less jarring, > because it's what I expect. q{ } always makes me think of a function > call (which it is, really.) Around simple single-line strings I tend to agree, but when you're dealing with multi-line SQL statements or other significant hunks of inline text, the q{} form isn't as grating. > We should choose what's most readable for most people in the common > case. Readability is deceptive in this case. I doubt anyone wants '' because you have to escape the 's often needs. That leaves q{} or "". "" would be more readable according to your preference, but it causes side effects that people don't expect. Numerous scripts I've read do things like: $addy = "chicks at chicks.net"; Obviously these folks were ignoring warnings and otherwise naive, but it exhibits that people new to perl don't consider the interpolation effects. In teaching Perl I've seen a lot of students, particularly those from the C and COBOL worlds trip over this. So, you're left teaching people the plethora of interpolation intricacies or "hay, these q{} things are safe quotes". Particularly if the q{} is used through the main code base, people will end up cutting and pasting it, and get used to the paradigm. I suspect that doing so would help avoid bugs. Avoiding escaping seems like a readability win too. -- The death of democracy is not likely to be an assassination from ambush. It will be a slow extinction from apathy, indifference, and undernourishment. -Robert Maynard Hutchins, educator (1899-1977) From gerv at mozilla.org Sat Mar 29 22:39:18 2003 From: gerv at mozilla.org (Gervase Markham) Date: Sat, 29 Mar 2003 22:39:18 +0000 Subject: SQL call formatting style In-Reply-To: References: Message-ID: <3E862096.9000608@mozilla.org> Christopher Hicks wrote: > Around simple single-line strings I tend to agree, but when you're dealing > with multi-line SQL statements or other significant hunks of inline text, > the q{} form isn't as grating. Perhaps not for you :-) >>We should choose what's most readable for most people in the common >>case. > > Readability is deceptive in this case. I doubt anyone wants '' because > you have to escape the 's often needs. But, as bbaetz said, it's rare that we need to embed a ' because we usually use parameters. And using ' means you don't have to escape any other character. > Obviously these folks were ignoring warnings and otherwise naive, but it > exhibits that people new to perl don't consider the interpolation effects. > In teaching Perl I've seen a lot of students, particularly those from the > C and COBOL worlds trip over this. So, you're left teaching people the > plethora of interpolation intricacies or "hay, these q{} things are safe > quotes". Particularly if the q{} is used through the main code base, > people will end up cutting and pasting it, and get used to the paradigm. I > suspect that doing so would help avoid bugs. Avoiding escaping seems like > a readability win too. I'm not convinced by any of this :-) - If people's code is creating warnings, it won't get checked in - If people don't understand interpolation in Perl, I'd be wary about letting them write code for Bugzilla - If "" was used throughout the main codebase, people would also cut and paste it, and be used to using that. - If we want to avoid escaping, let's use ' (see above.) Anyway, we are now going round in circles, and it's time for a justdave ruling :-) Dave? Pick a "standard", or say there's none and everyone can use whatever quoting style they like. Gerv From bbaetz at acm.org Sat Mar 29 22:49:44 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Sun, 30 Mar 2003 08:49:44 +1000 Subject: SQL call formatting style In-Reply-To: <3E862096.9000608@mozilla.org> References: <3E862096.9000608@mozilla.org> Message-ID: <20030329224944.GA1852@mango.home> On Sat, Mar 29, 2003 at 10:39:18PM +0000, Gervase Markham wrote: > > - If people don't understand interpolation in Perl, I'd be wary about > letting them write code for Bugzilla s/interpolation/quoting/ :) Bradley