From justdave at syndicomm.com Sun Feb 2 06:14:12 2003 From: justdave at syndicomm.com (David Miller) Date: Sun, 2 Feb 2003 01:14:12 -0500 Subject: Bugzilla->instance and friends In-Reply-To: <20030129103141.GA2390@mango.home> References: <20030129103141.GA2390@mango.home> Message-ID: On 1/29/03 9:31 PM +1100, Bradley Baetz wrote: > Currently, to get a template object, you do: > > my $template = Bugzilla->instance->template > > to get a dbh: > > my $dbh = Bugzilla->instance->dbh > > And so on. IOW, Bugzilla is a singleton class which has references to > various objects we want to keep arround for some span of time. Maybe I need some education here as to what a "singleton class" is. Does that just mean that the object is the class itself and it doesn't really instantiate anything? Or that once instantiated, any other attempt to instantiate just gives you the first one originally instantiated, so everyone's talking to the same object no matter if they used "new" or not? > My original idea for the fix was to do what I'd been planning for a > while - make the Bugzilla object creation routine take an argument > indicating what type of program it was, and then only initalise whats > required. (With defaults so that the majority of cgi files wouldn't > have to specify anything, and would get Bugzilla::TYPE_CGI > automaticaly) This sounds like a cool thing to do. Might help tinderbox from trying to open a database connection when running tests, too. :-) > Then it was mentioned that the ->instance was a bit ugly. Removing that > points to the fact that we don't really need an object, whch leaves us > with a whle set of unrelated subroutines. OK. > So, the alternative is to use Class::Singleton either directly (or just > use the standard ||= construct), possibly with Apache::Singleton too > (despite the name, this sort of works without mod_perl, although the > install tests fail). So what is Class::Singleton, and what advantage do we get from that outside of what we'd get from hanging onto the object ID of the created instance somewhere and just handing that same object ID back to any subsequent requests for a new Bugzilla object? This could be accomplished with a "my" variable within the package definition that isn't stored in the object hash, couldn't it? > To get a template, you then: > > Bugzilla::Template->instance Why not Bugzilla->template ? > and to get a cgi object: > > Bugzilla::CGI->instance or Bugzilla->cgi ? > Code which wants a cgi object then explictly asks for it, and its > created as needed. Code which doesn't never asks, and so never inits > stuff it doesn't want. This would be a good thing anyway. > (This doesn't actually solve the current problem > with processmail, because globals.pl gets the cgi object, so we probably > have to $^0 test that instead. Just forget about that for now, and > imagine a world where globals.pl is gone. :) On the other hand, turning processmail into a module is very close to landing, and that would solve that problem anyway, wouldn't it? (since the CGI stream won't be getting re-initialized after it's already been read off of standard input during a POST) -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From bbaetz at acm.org Sun Feb 2 06:21:25 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Sun, 2 Feb 2003 17:21:25 +1100 Subject: Bugzilla->instance and friends In-Reply-To: References: <20030129103141.GA2390@mango.home> Message-ID: <20030202062125.GA5363@mango.home> On Sun, Feb 02, 2003 at 01:14:12AM -0500, David Miller wrote: > On 1/29/03 9:31 PM +1100, Bradley Baetz wrote: > Maybe I need some education here as to what a "singleton class" is. Does > that just mean that the object is the class itself and it doesn't really > instantiate anything? Or that once instantiated, any other attempt to > instantiate just gives you the first one originally instantiated, so > everyone's talking to the same object no matter if they used "new" or not? The latter. You have a way of getting to the class, in such a way that its only created once, and then returned eah time its 'got'. > > > My original idea for the fix was to do what I'd been planning for a > > while - make the Bugzilla object creation routine take an argument > > indicating what type of program it was, and then only initalise whats > > required. (With defaults so that the majority of cgi files wouldn't > > have to specify anything, and would get Bugzilla::TYPE_CGI > > automaticaly) > > This sounds like a cool thing to do. Might help tinderbox from trying to > open a database connection when running tests, too. :-) Yep. > > So, the alternative is to use Class::Singleton either directly (or just > > use the standard ||= construct), possibly with Apache::Singleton too > > (despite the name, this sort of works without mod_perl, although the > > install tests fail). > > So what is Class::Singleton, and what advantage do we get from that outside > of what we'd get from hanging onto the object ID of the created instance > somewhere and just handing that same object ID back to any subsequent > requests for a new Bugzilla object? This could be accomplished with a "my" > variable within the package definition that isn't stored in the object > hash, couldn't it? Thats how Class::Singleton works - its just a semi-standard interface for it. Reusable code, etc, etc > > > To get a template, you then: > > > > Bugzilla::Template->instance > > Why not Bugzilla->template ? Because then the Bugzilla class is the central point of access. Thats what we have now, but once the template accessor sub becomes: my $_template; sub template { $_template ||= new Bugzilla::Template(); return $_template; } theres not much advantage to having it in the central package > > Code which wants a cgi object then explictly asks for it, and its > > created as needed. Code which doesn't never asks, and so never inits > > stuff it doesn't want. > > This would be a good thing anyway. Right. Currently its created at startup, and I definatley want to move away from that. > On the other hand, turning processmail into a module is very close to > landing, and that would solve that problem anyway, wouldn't it? (since the > CGI stream won't be getting re-initialized after it's already been read off > of standard input during a POST) Sure, but it doesn't solve problems like collectstats.pl trying to create one. And that CGI ENV var thing is a bit icky. Anyway, I think I'll leave it in Bugzilla.pm for now, and see what comes up later on, when I get time to do the login stuff. Bradley From justdave at syndicomm.com Sun Feb 2 06:28:47 2003 From: justdave at syndicomm.com (David Miller) Date: Sun, 2 Feb 2003 01:28:47 -0500 Subject: Bugzilla->instance and friends In-Reply-To: <20030202062125.GA5363@mango.home> References: <20030129103141.GA2390@mango.home> <20030202062125.GA5363@mango.home> Message-ID: On 2/2/03 5:21 PM +1100, Bradley Baetz wrote: >> On the other hand, turning processmail into a module is very close to >> landing, and that would solve that problem anyway, wouldn't it? (since the >> CGI stream won't be getting re-initialized after it's already been read off >> of standard input during a POST) > > Sure, but it doesn't solve problems like collectstats.pl trying to > create one. And that CGI ENV var thing is a bit icky. Well, collectstats.pl isn't trying to create one. collecstats.pl is trying to run a CGI program from the command line and capture its output. What really needs to happen there is for the meat of duplicates.cgi to get turned into a .pm file, and have both duplicates.cgi and collectstats.pl load that .pm. -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From bbaetz at acm.org Sun Feb 2 06:36:45 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Sun, 2 Feb 2003 17:36:45 +1100 Subject: Bugzilla->instance and friends In-Reply-To: References: <20030129103141.GA2390@mango.home> <20030202062125.GA5363@mango.home> Message-ID: <20030202063645.GA6682@mango.home> On Sun, Feb 02, 2003 at 01:28:47AM -0500, David Miller wrote: > On 2/2/03 5:21 PM +1100, Bradley Baetz wrote: > > Well, collectstats.pl isn't trying to create one. collecstats.pl is trying > to run a CGI program from the command line and capture its output. What > really needs to happen there is for the meat of duplicates.cgi to get > turned into a .pm file, and have both duplicates.cgi and collectstats.pl > load that .pm. > Oh, yeah. I forgot about that bit :) I meant the 'normal' part of collectstats. Ditto for attempting logins. Anyway, I'll leave the stuff in Bugzilla.pm for the moment, at least. Bradley From willy at lukoil.uu.ru Sun Feb 2 11:21:17 2003 From: willy at lukoil.uu.ru (Vitaly Fedrushkov) Date: Sun, 2 Feb 2003 16:21:17 +0500 (YEKT) Subject: Issue Nomenclature In-Reply-To: <3E357559.8060205@mozilla.org> Message-ID: Good $daytime, > Date: Mon, 27 Jan 2003 18:07:21 +0000 > From: Gervase Markham > To: developers at bugzilla.org > Subject: [Fwd: Re: Issue Nomenclature] > See the below newsgroup message. What do people think of the idea of > a patch (or incremental change) to convert all instances of "bug" > etc. to [% _bug %], [% _Bug %], [% _bugs %] etc.? > Would this just be a maintainability headache for the templates, or > would it allow people to make a much-requested customisation without > the need for custom templates and hassle? I wish things were so simple... IMHO the question is twofold: Generic Bugzilla installation can be used to handle events of a different nature: o Bugs themselves (i.e. coding errors) o Design problems/deficiencies o Requests for enhancement o Planned tasks o Support requests o Complex project-tracking nodes (a.k.a meta-bugs at mozilla.org) or what else can happen within issue/problem tracking system. For some applications (like support contract management) term 'bug' is especially unwelcome :-) These can be made configurable and engine can be written to auto-linkify all of them. One can even ignore the fact these terms are non-interchangeable and then refrain from creating 'Class' field in bugs table. Another problem is localization. Hardcoded components would require complete rewrite. With inflexional languages even regexp matches may die of complexity. For instance, paradigm of word 'bug' in Russian consists of twelve forms. Fully localized template set is unavoidable here, as well as locale-aware coding. Simply telling '[% _bug %]' from '[% _bugs %]' is not enough, more likely we need a set of 'naming operators', like bug_count(number, term). Regards, Willy. -- No easy hope or lies | Vitaly "Willy the Pooh" Fedrushkov Shall bring us to our goal, | Control Systems and Processes Division But iron sacrifice | LUKOIL Company, Chelyabinsk Branch Of Body, Will and Soul. | willy at lukoil.uu.ru +7 3512 620367 R.Kipling | VVF1-RIPE From willy at lukoil.uu.ru Sun Feb 2 09:27:55 2003 From: willy at lukoil.uu.ru (Vitaly Fedrushkov) Date: Sun, 2 Feb 2003 14:27:55 +0500 (YEKT) Subject: Impressions from LinuxWorld In-Reply-To: <3E35A171.3040704@bugopolis.com> Message-ID: Good $daytime, > Date: Mon, 27 Jan 2003 13:15:29 -0800 > From: Jim Walters > To: developers at bugzilla.org > Subject: Impressions from LinuxWorld > sort of a loss as to how to start collecting together a list of the > Bugzilla supported languages and putting them on the box. Have a look at bug 126266. Every localized Bugzilla depends on it. > Also, I'm I correct (I didn't promised this to anyone) in saying > that the TT library will switch templates based on the language of > the client browser? Apache can do it as well, rewriting URLs according to Accept-Language: field. Not sure which tool is better here. TT can be implemented directly within Bugzilla -- but it would interfere with non-bugzilla parts of a web server, already localized in a different fashion. Apache specific method is cheaper, but (a) it is Apache-specific, and (b) it deserves no more than cookbook recipe in Bugzilla Installation Guide. Regards, Willy. -- No easy hope or lies | Vitaly "Willy the Pooh" Fedrushkov Shall bring us to our goal, | Control Systems and Processes Division But iron sacrifice | LUKOIL Company, Chelyabinsk Branch Of Body, Will and Soul. | willy at lukoil.uu.ru +7 3512 620367 R.Kipling | VVF1-RIPE From chicks at chicks.net Sun Feb 2 17:35:23 2003 From: chicks at chicks.net (Christopher Hicks) Date: Sun, 2 Feb 2003 12:35:23 -0500 (EST) Subject: Bugzilla->instance and friends In-Reply-To: <20030202062125.GA5363@mango.home> Message-ID: On Sun, 2 Feb 2003, Bradley Baetz wrote: > > > To get a template, you then: > > > > > > Bugzilla::Template->instance > > > > Why not Bugzilla->template ? > > Because then the Bugzilla class is the central point of access. Thats > what we have now, but once the template accessor sub becomes: > > my $_template; > sub template { > $_template ||= new Bugzilla::Template(); > return $_template; > } > > theres not much advantage to having it in the central package I'd like to place at least one vote for the fact that Bugzilla->template looks a lot cleaner than Bugzilla::Template->instance. -- "Never offend people with style when you can offend them with substance." - Sam Brown From gerv at mozilla.org Sun Feb 2 23:06:26 2003 From: gerv at mozilla.org (Gervase Markham) Date: Sun, 02 Feb 2003 23:06:26 +0000 Subject: Bugzilla->instance and friends In-Reply-To: References: Message-ID: <3E3DA472.9070005@mozilla.org> > I'd like to place at least one vote for the fact that Bugzilla->template > looks a lot cleaner than Bugzilla::Template->instance. Seconded :-) Gerv From gerv at mozilla.org Sun Feb 2 23:19:57 2003 From: gerv at mozilla.org (Gervase Markham) Date: Sun, 02 Feb 2003 23:19:57 +0000 Subject: Issue Nomenclature In-Reply-To: References: Message-ID: <3E3DA79D.1030100@mozilla.org> > Another problem is localization. I think that there is no problem with localisation, because a localiser could, if they wished, remove this system completely (i.e. treat the template tags as text.) Gerv From stu at asyn.com Mon Feb 3 15:18:44 2003 From: stu at asyn.com (Stuart Donaldson) Date: Mon, 03 Feb 2003 07:18:44 -0800 Subject: Whats the status on customfields? Message-ID: <3E3E8854.8040909@asyn.com> As many others have pointed out, there is a need for some level of custom field support. What is the current status of custom fields, and in particular getting custom fields into a supported case? I recall someone suggesting a new approach was needed rather than the one taken in the custom fields patch. Is anything going on here? Thanks... -Stuart- From jason at pyeron.com Tue Feb 4 17:28:25 2003 From: jason at pyeron.com (Jason Pyeron) Date: Tue, 4 Feb 2003 12:28:25 -0500 (EST) Subject: API to access data In-Reply-To: Message-ID: yes we are doing some documentation On Fri, 24 Jan 2003, J. Paul Reed wrote: On Fri, 24 Jan 2003, Jeffrey Morgan wrote: > Is that a document that describes how an external system can interact > with the bugzilla data? I don't know of such a document (other than the schema diagrams, the administrator's manual, and, of course, the source code itself). But take a look at bug 188570; if they're gonna do a Java port, one of the first things they'd have to do, I'd think, is create such a document. Later, Paul ----------------------------------------------------------------------- J. Paul Reed preed at sigkill.com || web.sigkill.com/preed To hold on to sanity too tight is insane. -- Nick Falzone, Pushing Tin ---- 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 Wed Feb 5 21:40:53 2003 From: gerv at mozilla.org (Gervase Markham) Date: Wed, 05 Feb 2003 21:40:53 +0000 Subject: Whats the status on customfields? In-Reply-To: <3E3E8854.8040909@asyn.com> References: <3E3E8854.8040909@asyn.com> Message-ID: <3E4184E5.5000805@mozilla.org> Stuart Donaldson wrote: > As many others have pointed out, there is a need for some level of > custom field support. > What is the current status of custom fields, and in particular getting > custom fields into a supported case? All the details are in the bug. So I think the situation is probably the same as ever. This has not yet reached the top of anyone's priority list. > I recall someone suggesting a new > approach was needed rather than the one taken in the custom fields > patch. Is anything going on here? There's been some implementation discussion, but I don't know if a consensus was reached. Gerv From bugdude1 at syndicomm.com Thu Feb 6 03:41:06 2003 From: bugdude1 at syndicomm.com (Dave Miller) Date: Wed, 5 Feb 2003 22:41:06 -0500 Subject: [Q] Bugzilla 2.16.2 on MacOSX Server 10.2.3 In-Reply-To: References: <1fpvbcz.hovcd916hz47mN%eRrEiMcOgVoErMrE@cox.net> <1fpwwjg.1xh0hhn12rnsxyN%eRrEiMcOgVoErMrE@cox.net> Message-ID: At 2:45 PM -0500 2/5/03, Dave Miller wrote: > On 2/5/03 2:37 PM -0500, Eric wrote: > >> Dave Miller wrote: >> >>> Install DBI first. What errors did you get installing DBI? I've never had >>> any problems with it on OS X. >> >> when installing DBI, I get the following errors: > > Not sure about the taint error... bbaetz might have a better clue on that > one... For the record for anyone else reading along looking for answers later... :) Eric came into IRC and we battled this one for a while... Ended up deciding that DBI 1.32 doesn't like Perl 5.6.0. I tried it on a Linux box I had with Perl 5.6.0 on it with the same results (taint error in the t/10examp.t test). Specifically installing DBI 1.31 worked. Of course, he's using 2.16.2, so he can get away with that. Bugzilla 2.17.x requires DBI 1.32. Does this mean we now need to require Perl 5.6.1 for Bugzilla or is this something that'll get fixed in a future DBI? -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From bbaetz at acm.org Thu Feb 6 07:56:55 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Thu, 6 Feb 2003 18:56:55 +1100 Subject: [Q] Bugzilla 2.16.2 on MacOSX Server 10.2.3 In-Reply-To: References: <1fpvbcz.hovcd916hz47mN%eRrEiMcOgVoErMrE@cox.net> <1fpwwjg.1xh0hhn12rnsxyN%eRrEiMcOgVoErMrE@cox.net> Message-ID: <20030206075655.GA10805@mango.home> On Wed, Feb 05, 2003 at 10:41:06PM -0500, Dave Miller wrote: > > Ended up deciding that DBI 1.32 doesn't like Perl 5.6.0. I tried it on a > Linux box I had with Perl 5.6.0 on it with the same results (taint error in > the t/10examp.t test). Specifically installing DBI 1.31 worked. Of > course, he's using 2.16.2, so he can get away with that. Bugzilla 2.17.x > requires DBI 1.32. Does this mean we now need to require Perl 5.6.1 for > Bugzilla or is this something that'll get fixed in a future DBI? The taint tests were disabled in 1.31, so that won't help you. :) What specifically fails? I wrote the Taint patches for DBI, so... The tests are a bit hacky, though, so it could just be a test bug. Bradley From gerv at mozilla.org Thu Feb 6 10:03:23 2003 From: gerv at mozilla.org (Gervase Markham) Date: Thu, 06 Feb 2003 10:03:23 +0000 Subject: Landfill attachment uploads Message-ID: <3E4232EB.4010401@mozilla.org> Anyone else getting: Status: 400 Bad request (malformed multipart POST) in the resulting page when uploading attachments to landfill? Gerv From jussi at comlink.fi Thu Feb 6 12:02:54 2003 From: jussi at comlink.fi (Jussi Sirpoma) Date: Thu, 06 Feb 2003 14:02:54 +0200 Subject: Landfill attachment uploads In-Reply-To: <3E4232EB.4010401@mozilla.org> References: <3E4232EB.4010401@mozilla.org> Message-ID: <3E424EEE.5040905@comlink.fi> On 6.2.2003 12:03 Gervase Markham spoke: > Anyone else getting: > Status: 400 Bad request (malformed multipart POST) > in the resulting page when uploading attachments to landfill? Yes. There is a bug about it: http://bugzilla.mozilla.org/show_bug.cgi?id=187861 Jussi Sirpoma From gerv at mozilla.org Thu Feb 6 12:26:37 2003 From: gerv at mozilla.org (Gervase Markham) Date: Thu, 06 Feb 2003 12:26:37 +0000 Subject: Landfill attachment uploads In-Reply-To: <3E424EEE.5040905@comlink.fi> References: <3E4232EB.4010401@mozilla.org> <3E424EEE.5040905@comlink.fi> Message-ID: <3E42547D.8010704@mozilla.org> >> Anyone else getting: >> Status: 400 Bad request (malformed multipart POST) >> in the resulting page when uploading attachments to landfill? > > Yes. There is a bug about it: > http://bugzilla.mozilla.org/show_bug.cgi?id=187861 Oops, sorry about that, everyone. I'm even CCed. I didn't match the bug with the symptoms. Gerv From caseyg at chsamerica.com Thu Feb 6 12:34:21 2003 From: caseyg at chsamerica.com (Casey Gregoire) Date: Thu, 6 Feb 2003 07:34:21 -0500 Subject: Landfill attachment uploads Message-ID: Shadow DB: I tried turning on the shadow DB in version 2.16.2 and the page hangs and never get to the end where it loads the footer. Should I report this as a bug? Or and I doing something wrong? Are there any more steps to using a shadow db other then just turning it on in settings? Thanks, Casey Gregoire -----Original Message----- From: Gervase Markham [mailto:gerv at mozilla.org] Sent: Thursday, February 06, 2003 5:03 AM To: developers at bugzilla.org Subject: Landfill attachment uploads Anyone else getting: Status: 400 Bad request (malformed multipart POST) in the resulting page when uploading attachments to landfill? Gerv ---- To view or change your list settings, click here: From gerv at mozilla.org Thu Feb 6 13:22:48 2003 From: gerv at mozilla.org (Gervase Markham) Date: Thu, 06 Feb 2003 13:22:48 +0000 Subject: Landfill attachment uploads In-Reply-To: References: Message-ID: <3E4261A8.7020409@mozilla.org> Casey Gregoire wrote: > Shadow DB: I tried turning on the shadow DB in version 2.16.2 and the page > hangs and never get to the end where it loads the footer. Should I report > this as a bug? Or and I doing something wrong? Are there any more steps to > using a shadow db other then just turning it on in settings? Yes. I believe creating it is fairly important :-) Seriously, the 2.16.2 shadowdb is a bit of a hack, and you shouldn't need it unless your database is _very_ large (of the public Bugzillas, only b.m.o. uses it.) The new shadowdb in 2.17 uses MySQL replication, which is far better suited to the job. Gerv From caseyg at chsamerica.com Thu Feb 6 13:49:43 2003 From: caseyg at chsamerica.com (Casey Gregoire) Date: Thu, 6 Feb 2003 08:49:43 -0500 Subject: ShodawDB Message-ID: Ok thanks. Just wondering. Thanks, Casey Gregoire -----Original Message----- From: Gervase Markham [mailto:gerv at mozilla.org] Sent: Thursday, February 06, 2003 8:23 AM To: developers at bugzilla.org Subject: Re: Landfill attachment uploads Casey Gregoire wrote: > Shadow DB: I tried turning on the shadow DB in version 2.16.2 and the page > hangs and never get to the end where it loads the footer. Should I report > this as a bug? Or and I doing something wrong? Are there any more steps to > using a shadow db other then just turning it on in settings? Yes. I believe creating it is fairly important :-) Seriously, the 2.16.2 shadowdb is a bit of a hack, and you shouldn't need it unless your database is _very_ large (of the public Bugzillas, only b.m.o. uses it.) The new shadowdb in 2.17 uses MySQL replication, which is far better suited to the job. Gerv ---- To view or change your list settings, click here: From justdave at syndicomm.com Thu Feb 6 19:33:52 2003 From: justdave at syndicomm.com (David Miller) Date: Thu, 6 Feb 2003 14:33:52 -0500 Subject: [Q] Bugzilla 2.16.2 on MacOSX Server 10.2.3 In-Reply-To: <20030206075655.GA10805@mango.home> References: <1fpvbcz.hovcd916hz47mN%eRrEiMcOgVoErMrE@cox.net> <1fpwwjg.1xh0hhn12rnsxyN%eRrEiMcOgVoErMrE@cox.net> <20030206075655.GA10805@mango.home> Message-ID: On 2/6/03 6:56 PM +1100, Bradley Baetz wrote: > On Wed, Feb 05, 2003 at 10:41:06PM -0500, Dave Miller wrote: >> >> Ended up deciding that DBI 1.32 doesn't like Perl 5.6.0. I tried it on a >> Linux box I had with Perl 5.6.0 on it with the same results (taint error in >> the t/10examp.t test). Specifically installing DBI 1.31 worked. Of >> course, he's using 2.16.2, so he can get away with that. Bugzilla 2.17.x >> requires DBI 1.32. Does this mean we now need to require Perl 5.6.1 for >> Bugzilla or is this something that'll get fixed in a future DBI? > > The taint tests were disabled in 1.31, so that won't help you. :) > > What specifically fails? I wrote the Taint patches for DBI, so... > > The tests are a bit hacky, though, so it could just be a test bug. t/10examp..............ok 103/245Insecure dependency in parameter 1 of DBI::st=HASH(0x20442c)->FETCH method call while running with -T switch at t/10examp.t line 320. DBI handle cleared whilst still active. dbih_clearcom (sth 0x20563c 0x10be40, com 0x1f6c10, imp DBD::ExampleP::st): FLAGS 0x60011: COMSET Warn TaintIn TaintOut PARENT DBI::db=HASH(0x1cbeec) KIDS 0 (0 Active) IMP_DATA ARRAY(0x20587c) NUM_OF_FIELDS 3 NUM_OF_PARAMS 1 t/10examp..............dubious Test returned status 20 (wstat 5120, 0x1400) -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From justdave at syndicomm.com Sat Feb 8 21:09:45 2003 From: justdave at syndicomm.com (David Miller) Date: Sat, 8 Feb 2003 16:09:45 -0500 Subject: Patch triage Message-ID: Just FYI for everyone, I just sent Brant Gurganus off on a patch triage spree... (thanks Brant!) We appear to have somewhere between 100 and 200 patches in bugzilla.mozilla.org that kind of dropped off the radar when we switched over to using request flags, and he's going through and requesting review "from the wind" on patches that don't have comments in the bug indicating they're bad, so we can get them back in the system. -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From justdave at syndicomm.com Sat Feb 8 22:09:41 2003 From: justdave at syndicomm.com (David Miller) Date: Sat, 8 Feb 2003 17:09:41 -0500 Subject: Patch triage In-Reply-To: References: Message-ID: On 2/8/03 4:09 PM -0500, David Miller wrote: > Just FYI for everyone, I just sent Brant Gurganus off on a patch triage > spree... (thanks Brant!) We appear to have somewhere between 100 and 200 > patches in bugzilla.mozilla.org that kind of dropped off the radar when we > switched over to using request flags, and he's going through and requesting > review "from the wind" on patches that don't have comments in the bug > indicating they're bad, so we can get them back in the system. And if you see something that looks funny from what he did, by all means go and fix it. :) He's probably getting more on the radar than we need, but I'd rather get extra stuff and go back and fix it than to have things fall through the cracks. :) -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From justdave at syndicomm.com Mon Feb 10 03:29:08 2003 From: justdave at syndicomm.com (David Miller) Date: Sun, 9 Feb 2003 22:29:08 -0500 Subject: cvs updates Message-ID: The following empty directories which have never had any files checked into them on any branch have been removed from cvs: mozilla/webtools/bugzilla/DiffPrinter/ mozilla/webtools/bugzilla/PatchIterator/ mozilla/webtools/bugzilla/test.c/ If you've been using the "-dP" options with your cvs updates, as recommended, then you won't even notice. However, if you've been using "-d" without the "-dP" or you like to do updates via checking out overtop of your original, cvs will complain loudly the next time you update until you remove the entries for those directories from the CVS/Entries file. If this causes any major problems, I can put them back really easily, but I kinda doubt anyone will miss them. :) (I'm getting sick of all the "who the heck added this directory!?!?" comments, so I went and did something about it :) Oh, and if you're the person who added those, don't do it again! :) (ask first) -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From geeta.tripathi at bhartitelesoft.com Mon Feb 10 11:30:02 2003 From: geeta.tripathi at bhartitelesoft.com (geeta) Date: Mon, 10 Feb 2003 17:00:02 +0530 Subject: Regarding CVS -Urgent!!!!! References: Message-ID: <016d01c2d0f7$d0ff7fa0$800310ac@tripathi.bhartitelesoft.com> Hi David, I have to integrate the current Bugzilla 2.16.1 with CVS. How do I do? The CVS is installed in another Unix machine. Do I need to have a web interface with CVS? Is there any free CVS patch available?? How much time will it take for the integration. And how to use Bugzilla along with CVS Looking forward to hear from you. Regards, Geeta From kiko at async.com.br Mon Feb 10 11:57:56 2003 From: kiko at async.com.br (Christian Reis) Date: Mon, 10 Feb 2003 09:57:56 -0200 Subject: Regarding CVS -Urgent!!!!! In-Reply-To: <016d01c2d0f7$d0ff7fa0$800310ac@tripathi.bhartitelesoft.com>; from geeta.tripathi@bhartitelesoft.com on Mon, Feb 10, 2003 at 05:00:02PM +0530 References: <016d01c2d0f7$d0ff7fa0$800310ac@tripathi.bhartitelesoft.com> Message-ID: <20030210095756.D5114@blackjesus.async.com.br> On Mon, Feb 10, 2003 at 05:00:02PM +0530, geeta wrote: > I have to integrate the current Bugzilla 2.16.1 with CVS. > > The CVS is installed in another Unix machine. Do I need to have a web > interface with CVS? > > Is there any free CVS patch available?? > How much time will it take for the integration. Okay, I need to give a really overdue update here. I'm the one responsible for coding up a patch for CVS support, and I've actually done some work on it. It's not very difficult to code once you get the hang of writing a loginfo script. Unfortunately, the past month and the next two weeks have been/are the final period of my MSc work, and I'm hacking on the thesis like a madman. This has left me without time to work on *anything* except the damned document, and I'm getting sick of it. Anyway, this means I still don't have a patch ready. However, I *commit* myself to working on this on the weekend of Feb 22-23 as this is one day after the final schedule for delivering the aforementioned damned document, and have something ready to use on Monday. As for the web interface, no. What the script does is (roughly) look at the commit log, scan for a bug number, and make an HTTP connection to Bugzilla to submit a comment change. It's actually pretty simple, and should work well. The problems I ran into were related to using the CGI interface of Bugzilla (if you can call it an interface :) to submit, but that's solved. There are some syntax definitions to make, but that can wait till the basic infrastructure is put together. Sorry for being so late with this, I'm really embarassed. :-/ Take care, -- Christian Reis, Senior Engineer, Async Open Source, Brazil. http://async.com.br/~kiko/ | [+55 16] 261 2331 | NMFL From dswegen at software.plasmon.com Mon Feb 10 14:27:41 2003 From: dswegen at software.plasmon.com (Dave Swegen) Date: Mon, 10 Feb 2003 14:27:41 +0000 Subject: Regarding CVS -Urgent!!!!! In-Reply-To: <20030210095756.D5114@blackjesus.async.com.br> References: <016d01c2d0f7$d0ff7fa0$800310ac@tripathi.bhartitelesoft.com> <20030210095756.D5114@blackjesus.async.com.br> Message-ID: <20030210142735.GA18491@software.plasmon> On Mon, Feb 10, 2003 at 09:57:56AM -0200, Christian Reis wrote: > On Mon, Feb 10, 2003 at 05:00:02PM +0530, geeta wrote: > > As for the web interface, no. What the script does is (roughly) look at > the commit log, scan for a bug number, and make an HTTP connection to > Bugzilla to submit a comment change. It's actually pretty simple, and > should work well. The problems I ran into were related to using the CGI > interface of Bugzilla (if you can call it an interface :) to submit, but > that's solved. The way I did it at work was to use the contrib/bugzilla_email_append.pl script which picks up mails from a script called by CVSROOT/loginfo. Works like a charm. I modified it to allow appending a commit log to multiple bugs. I really don't know which approach is the cleanest. One question, though: Does the CGI approach attribute commit logs to the correct person (i.e. not in the bug comment itself, but in the bug comment header)? I'm in the process of cleaning the scripts up a tad (some other people want them too). Let me know if they would be of interest. Cheers Dave From kiko at async.com.br Mon Feb 10 14:51:21 2003 From: kiko at async.com.br (Christian Reis) Date: Mon, 10 Feb 2003 12:51:21 -0200 Subject: Regarding CVS -Urgent!!!!! In-Reply-To: <20030210142735.GA18491@software.plasmon>; from dswegen@software.plasmon.com on Mon, Feb 10, 2003 at 02:27:41PM +0000 References: <016d01c2d0f7$d0ff7fa0$800310ac@tripathi.bhartitelesoft.com> <20030210095756.D5114@blackjesus.async.com.br> <20030210142735.GA18491@software.plasmon> Message-ID: <20030210125121.B5664@blackjesus.async.com.br> On Mon, Feb 10, 2003 at 02:27:41PM +0000, Dave Swegen wrote: > The way I did it at work was to use the contrib/bugzilla_email_append.pl > script which picks up mails from a script called by CVSROOT/loginfo. > Works like a charm. I modified it to allow appending a commit log to > multiple bugs. Does it access the database directly? > I really don't know which approach is the cleanest. One question, > though: Does the CGI approach attribute commit logs to the correct > person (i.e. not in the bug comment itself, but in the bug comment > header)? Well, you in theory can't do that via CGI as it would mean impersonating a user. However, using the database directly would work. Or perhaps the new backend separation offers an easy way to create arbitrary comments; I haven't checked it out. You could access the remote database over the network, of course, though this requires opening your database server to external access. > I'm in the process of cleaning the scripts up a tad (some other people > want them too). Let me know if they would be of interest. If you'd like to take this off my hands, I'm happy to. The bug relevant in this case is: http://bugzilla.mozilla.org/show_bug.cgi?id=129765 Though it cites Bonsai I don't think bonsai is really important here - the script should work even if bonsai isn't there. Take care, -- Christian Reis, Senior Engineer, Async Open Source, Brazil. http://async.com.br/~kiko/ | [+55 16] 261 2331 | NMFL From dswegen at software.plasmon.com Mon Feb 10 15:58:16 2003 From: dswegen at software.plasmon.com (Dave Swegen) Date: Mon, 10 Feb 2003 15:58:16 +0000 Subject: Regarding CVS -Urgent!!!!! In-Reply-To: <20030210125121.B5664@blackjesus.async.com.br> References: <016d01c2d0f7$d0ff7fa0$800310ac@tripathi.bhartitelesoft.com> <20030210095756.D5114@blackjesus.async.com.br> <20030210142735.GA18491@software.plasmon> <20030210125121.B5664@blackjesus.async.com.br> Message-ID: <20030210155816.GA19235@software.plasmon> On Mon, Feb 10, 2003 at 12:51:21PM -0200, Christian Reis wrote: > On Mon, Feb 10, 2003 at 02:27:41PM +0000, Dave Swegen wrote: > > The way I did it at work was to use the contrib/bugzilla_email_append.pl > > script which picks up mails from a script called by CVSROOT/loginfo. > > Works like a charm. I modified it to allow appending a commit log to > > multiple bugs. > > Does it access the database directly? contrib/bugzilla_append_email.pl does, yes. [stuff deleted] > If you'd like to take this off my hands, I'm happy to. The bug relevant > in this case is: > > http://bugzilla.mozilla.org/show_bug.cgi?id=129765 > > Though it cites Bonsai I don't think bonsai is really important here - > the script should work even if bonsai isn't there. I'm not going to even go anywhere near venturing out on a limb on this one - I'm far too chicken :) And I know next to nothing (actually, make that nothing) about how Bonsai works. I have to say that I'm surprised that whenever the whole cvs->bugzilla questions gets raised it seems that nobody mentions the contrib/bugzilla_email_append script. Is it considered too bad, or is it just forgotten? In brief the system goes as follows: cvs->CVSROOT/script, which gets a list of bug ids to which the comment should get attached. It also grabs the comment (obviously), the old and the new version, and the branch it is on. CVSROOT/script->email->contrib/bugzilla_email_append.pl via a unqiue user on the same system as bz. This script checks that the bug-ids are valid, and that the sending user is valid, and then simply appends the comment as that user. It will however fail if the email address that the CVS system sends it as doesn't match that of the users bugzilla login. I will probably fix this by having any such non-existant users default to a bugzilla-logger user (or somesuch). Only problem with that approach is that it leaves the system even more open to abuse. The other minor problem is that if you do a mass commit which cover multiple subdirectories, CVS treats each set of sub-directory files as a separate commit, and thus you get one comment per subdirectory. Ugly, but still useable (we are thinking of patching cvs to not do this). Oh, and I have no idea if it would work on a windows system (and frankly, I don't care ;) If it sounds reasonable I'll post a patch to the contrib directory to bug 129765, and I'll let you sort it out from there. Cheers Dave From chicks at chicks.net Mon Feb 10 16:39:46 2003 From: chicks at chicks.net (Christopher Hicks) Date: Mon, 10 Feb 2003 11:39:46 -0500 (EST) Subject: Regarding CVS -Urgent!!!!! In-Reply-To: <20030210155816.GA19235@software.plasmon> Message-ID: On Mon, 10 Feb 2003, Dave Swegen wrote: > The other minor problem is that if you do a mass commit which cover > multiple subdirectories, CVS treats each set of sub-directory files as a > separate commit, and thus you get one comment per subdirectory. Ugly, > but still useable (we are thinking of patching cvs to not do this). Speaking with one of cvs' commercial competitors that does imports of cvs into their proprietary system, apparently they had good luck correlating commits that happen at close to the same time. I would think the commit messages would match too. -- "Never offend people with style when you can offend them with substance." - Sam Brown From dswegen at software.plasmon.com Mon Feb 10 17:05:19 2003 From: dswegen at software.plasmon.com (Dave Swegen) Date: Mon, 10 Feb 2003 17:05:19 +0000 Subject: Regarding CVS -Urgent!!!!! In-Reply-To: References: <20030210155816.GA19235@software.plasmon> Message-ID: <20030210170519.GA20149@software.plasmon> On Mon, Feb 10, 2003 at 11:39:46AM -0500, Christopher Hicks wrote: > On Mon, 10 Feb 2003, Dave Swegen wrote: > > The other minor problem is that if you do a mass commit which cover > > multiple subdirectories, CVS treats each set of sub-directory files as a > > separate commit, and thus you get one comment per subdirectory. Ugly, > > but still useable (we are thinking of patching cvs to not do this). > > Speaking with one of cvs' commercial competitors that does imports of cvs > into their proprietary system, apparently they had good luck correlating > commits that happen at close to the same time. I would think the commit > messages would match too. Yes, but the problem is that then instead of a nice simple script you have yucky one that has to keep state. Nope, I would say that cvs is the place to put such a fix. Btw, does anyone know how Subversion deals with this? Cheers Dave From ashutosh_jack at hotmail.com Mon Feb 10 12:08:42 2003 From: ashutosh_jack at hotmail.com (Ashutosh Tyagi) Date: Mon, 10 Feb 2003 17:38:42 +0530 Subject: bugzilla mail Message-ID: Hi Developers, Actually I am confuring Bugzilla so that it can send internal mail in an office. We have internal e-mail id's we are using Argosoft server, by which we can send internal mails. Can you please tell that which component in Bugzilla is responsible for sending mails in Linux, is it sendmail or something else With regards Ashutosh Tyagi -------------- next part -------------- An HTML attachment was scrubbed... URL: From kiko at async.com.br Mon Feb 10 17:35:52 2003 From: kiko at async.com.br (Christian Reis) Date: Mon, 10 Feb 2003 15:35:52 -0200 Subject: Regarding CVS -Urgent!!!!! In-Reply-To: <20030210170519.GA20149@software.plasmon>; from dswegen@software.plasmon.com on Mon, Feb 10, 2003 at 05:05:19PM +0000 References: <20030210155816.GA19235@software.plasmon> <20030210170519.GA20149@software.plasmon> Message-ID: <20030210153552.D6301@blackjesus.async.com.br> On Mon, Feb 10, 2003 at 05:05:19PM +0000, Dave Swegen wrote: > > Speaking with one of cvs' commercial competitors that does imports of cvs > > into their proprietary system, apparently they had good luck correlating > > commits that happen at close to the same time. I would think the commit > > messages would match too. > > Btw, does anyone know how Subversion deals with this? Subversion controls versions of directories, so it would be considerably easier to implement this, as there is something like a global changeset. Take care, -- Christian Reis, Senior Engineer, Async Open Source, Brazil. http://async.com.br/~kiko/ | [+55 16] 261 2331 | NMFL From preed at sigkill.com Mon Feb 10 18:19:31 2003 From: preed at sigkill.com (J. Paul Reed) Date: Mon, 10 Feb 2003 10:19:31 -0800 (PST) Subject: bugzilla mail In-Reply-To: Message-ID: On Mon, 10 Feb 2003, Ashutosh Tyagi wrote: > Actually I am confuring Bugzilla so that it can send internal mail in an > office. We have internal e-mail id's we are using Argosoft server, by > which we can send internal mails. Can you please tell that which > component in Bugzilla is responsible for sending mails in Linux, is it > sendmail or something else That would be the Email Notification component. Right now, sendmail is used/required, but you might be interested in bug 84876, which will be changing that. If you need to hack at things and you're using < 2.17.3, look for processmail (actually, you'll probably just want to grep for "sendmail" to make sure you get all the instances of it); if you're using CVS (not likely, but...) take a look at BugMail.pm, which just went in. =) Later Paul ---------------------------------------------------------------------- J. Paul Reed preed at sigkill.com || web.sigkill.com/preed To hold on to sanity too tight is insane. -- Nick Falzone, Pushing Tin From gerv at mozilla.org Mon Feb 10 18:30:57 2003 From: gerv at mozilla.org (Gervase Markham) Date: Mon, 10 Feb 2003 18:30:57 +0000 Subject: Regarding CVS -Urgent!!!!! In-Reply-To: <20030210153552.D6301@blackjesus.async.com.br> References: <20030210155816.GA19235@software.plasmon> <20030210170519.GA20149@software.plasmon> <20030210153552.D6301@blackjesus.async.com.br> Message-ID: <3E47EFE1.7080400@mozilla.org> > Subversion controls versions of directories, so it would be considerably > easier to implement this, as there is something like a global changeset. Yes - in Subversion, version numbers are versions of the entire repository, not of individual files. This means that checkins are atomic, and very easy to separate. Gerv From jon at vmware.com Wed Feb 12 00:47:44 2003 From: jon at vmware.com (Jonathan Schatz) Date: 11 Feb 2003 16:47:44 -0800 Subject: odd db error message Message-ID: <1045010864.25294.19.camel@jonschatz-lx.vmware.com> sometime in the past 2 weeks a cvs update caused the following error to start showing up in my error_log: [Tue Feb 11 16:20:30 2003] [error] [client 172.16.14.80] [Tue Feb 11 16:20:30 2003] editopsys.cgi: DBI::db=HASH(0x864e8b8)->disconnect invalidates 1 active statement handle (either destroy statement handles or call finish on them before disconnecting) at Bugzilla.pm line 98., referer: http://jonschatz-lx/editopsys.cgi any 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 Feb 12 00:55:37 2003 From: justdave at syndicomm.com (David Miller) Date: Tue, 11 Feb 2003 19:55:37 -0500 Subject: odd db error message In-Reply-To: <1045010864.25294.19.camel@jonschatz-lx.vmware.com> References: <1045010864.25294.19.camel@jonschatz-lx.vmware.com> Message-ID: On 2/11/03 4:47 PM -0800, Jonathan Schatz wrote: > sometime in the past 2 weeks a cvs update caused the following error to > start showing up in my error_log: > > [Tue Feb 11 16:20:30 2003] [error] [client 172.16.14.80] [Tue Feb 11 > 16:20:30 2003] editopsys.cgi: DBI::db=HASH(0x864e8b8)->disconnect > invalidates 1 active statement handle (either destroy statement handles > or call finish on them before disconnecting) at Bugzilla.pm line 98., > referer: http://jonschatz-lx/editopsys.cgi > > any ideas? Yep. See http://bugzilla.mozilla.org/show_bug.cgi?id=192531 -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From gerv at mozilla.org Wed Feb 12 08:46:38 2003 From: gerv at mozilla.org (Gervase Markham) Date: Wed, 12 Feb 2003 08:46:38 +0000 Subject: Feedback from FOSDEM Message-ID: <3E4A09EE.4050702@mozilla.org> The audience at FOSDEM gave the following feedback on Bugzilla. We had one request for the "submit only my comment" feature when you mid-air. During the post-talk discussion, there was a lot of support for a "simpler" query page - not as simple as the simple search, but more complex than the one-liner on the guided bug reporting form. This could be b.m.o.-specific, or it could be general. There was also support for changing the default query (used by those not logged in) to include UNCONFIRMED and DUPLICATE bugs (i.e. resolution = --- OR DUPLICATE). A guy from Debian said "I saw all that stuff that Bugzilla does, and thought 'We need all of that.' But there's no hope of Debian converting without a decent email interface." I said that in order to have one which is maintained, we need an organisation which is using it regularly :-) But there might be scope here to expand our empire. Someone (maybe I'll get around to it, maybe not) needs to investigate the email interface on debbugs and see how hard it would be to do something like it. Gerv From preed at sigkill.com Wed Feb 12 08:48:33 2003 From: preed at sigkill.com (J. Paul Reed) Date: Wed, 12 Feb 2003 00:48:33 -0800 (PST) Subject: Feedback from FOSDEM In-Reply-To: <3E4A09EE.4050702@mozilla.org> Message-ID: On Wed, 12 Feb 2003, Gervase Markham wrote: > Someone (maybe I'll get around to it, maybe not) needs to investigate > the email interface on debbugs and see how hard it would be to do > something like it. It's a lofty goal, I know, but after we finish getting a good foundation for email notifications in the trunk (basically after 84876 goes in), I plan on filing a "Make the Email Notifications component not suck" and work on some real redesign/cleanup. I think with 124174 and 84876 in, we'll have a good basis to start from (which is why I've waited). Later, Paul ---------------------------------------------------------------------- J. Paul Reed preed at sigkill.com || web.sigkill.com/preed To hold on to sanity too tight is insane. -- Nick Falzone, Pushing Tin From bbaetz at acm.org Wed Feb 12 09:24:22 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Wed, 12 Feb 2003 20:24:22 +1100 Subject: Feedback from FOSDEM In-Reply-To: <3E4A09EE.4050702@mozilla.org> References: <3E4A09EE.4050702@mozilla.org> Message-ID: <20030212092422.GA2029@mango.home> On Wed, Feb 12, 2003 at 08:46:38AM +0000, Gervase Markham wrote: > A guy from Debian said "I saw all that stuff that Bugzilla does, and > thought 'We need all of that.' But there's no hope of Debian converting > without a decent email interface." I said that in order to have one > which is maintained, we need an organisation which is using it regularly > :-) But there might be scope here to expand our empire. Daniel Berlin apparently has something which parses gnats mail. I don't know how 'nice' the gnats api is, though, so it may not be appropriate for our needs. Its also possible to do a lot more in bugzilla than it is in gnats (or debbugs). OTOH, if debian developers can do what they currently do via mail, then they can use the web inerface for other stuff. There are also authentication issues to worry about. My auth rewrite has some vague infrastructure to allow gpg signed authentication, but thats a long way off. Bradley From sergeyli at pisem.net Wed Feb 12 16:02:51 2003 From: sergeyli at pisem.net (Sergey A. Lipnevich) Date: Wed, 12 Feb 2003 11:02:51 -0500 Subject: Feedback from FOSDEM In-Reply-To: <20030212092422.GA2029@mango.home> References: <3E4A09EE.4050702@mozilla.org> <20030212092422.GA2029@mango.home> Message-ID: <3E4A702B.6020603@pisem.net> Bradley Baetz wrote: > There are also authentication issues to worry about. My auth rewrite has > some vague infrastructure to allow gpg signed authentication, but thats > a long way off. > Sorry to get in the middle of another subject, but I have a question about this auth rewrite. There's code in Bugzilla which I adapted to use Net-LDAP package and it seems to work OK, but my bigger goal is for Bugzilla to pick up credentials assigned by Apache, because I'm using mod_auth_pam for authentication right now. Every time I think of rewriting this I can't help but wish for an actual auth framework for me not to invent the wheel without knowing a lot about entire Bugzilla codebase. Now the question. Can I have a pick at what authentication is going to be like, and any way I could help in bringing it into the trunk? I could then proceed with my stuff. Thanks! Sergey, SourceMage GNU/Linux, http://www.sourcemage.org/. From jon at vmware.com Wed Feb 12 20:30:50 2003 From: jon at vmware.com (Jonathan Schatz) Date: 12 Feb 2003 12:30:50 -0800 Subject: Feedback from FOSDEM In-Reply-To: <20030212092422.GA2029@mango.home> References: <3E4A09EE.4050702@mozilla.org> <20030212092422.GA2029@mango.home> Message-ID: <1045081850.19538.10.camel@jonschatz-lx.vmware.com> On Wed, 2003-02-12 at 01:24, Bradley Baetz wrote: > Daniel Berlin apparently has something which parses gnats mail. I don't > know how 'nice' the gnats api is, though, so it may not be appropriate > for our needs. Its also possible to do a lot more in bugzilla than it is > in gnats (or debbugs). OTOH, if debian developers can do what they > currently do via mail, then they can use the web inerface for other > stuff. *shudder* i don't think there's a gnats api, per se. we're moving from gnats to bugzilla right now. i've got a good bit of gnats parsing code. what the debian folks probably want is an equivalent to send-pr (iirc, openbsd uses this too), so that folks can file bugs remotely without having to understand the gnats/bugzilla web interface (or without having a working network connection, since mail will just queue locally when there's no active connection to the rest of the world). this is something i've been working on here for our remote sales engineers. the authentication problem is an issue (since email can be forged relatively easily). -jon -- Jonathan Schatz Engineering System Administrator VMware, Inc "Te occidere possunt sed te edere non possunt nefas est." From bbaetz at acm.org Wed Feb 12 20:11:39 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Thu, 13 Feb 2003 07:11:39 +1100 Subject: Feedback from FOSDEM In-Reply-To: <3E4A702B.6020603@pisem.net> References: <3E4A09EE.4050702@mozilla.org> <20030212092422.GA2029@mango.home> <3E4A702B.6020603@pisem.net> Message-ID: <20030212201139.GA1305@mango.home> On Wed, Feb 12, 2003 at 11:02:51AM -0500, Sergey A. Lipnevich wrote: > Sorry to get in the middle of another subject, but I have a question > about this auth rewrite. There's code in Bugzilla which I adapted to use > Net-LDAP package and it seems to work OK, but my bigger goal is for > Bugzilla to pick up credentials assigned by Apache, because I'm using > mod_auth_pam for authentication right now. Every time I think of > rewriting this I can't help but wish for an actual auth framework for me > not to invent the wheel without knowing a lot about entire Bugzilla > codebase. Now the question. Can I have a pick at what authentication is > going to be like, and any way I could help in bringing it into the > trunk? I could then proceed with my stuff. > Thanks! > Yeah, we have this in teh rewrite. Apache stuff is harder without mod_perl, though, although ISTR that there is that env variable we can use. mod_perl is a higher priority for me than apache auth, ahtough it really should be trivial after that patch goes in. See bug 180642 for details. Bradley From gerv at mozilla.org Wed Feb 12 22:17:47 2003 From: gerv at mozilla.org (Gervase Markham) Date: Wed, 12 Feb 2003 22:17:47 +0000 Subject: Feedback from FOSDEM In-Reply-To: References: Message-ID: <3E4AC80B.5040104@mozilla.org> > It's a lofty goal, I know, but after we finish getting a good foundation > for email notifications in the trunk (basically after 84876 goes in), I > plan on filing a "Make the Email Notifications component not suck" and > work on some real redesign/cleanup. > > I think with 124174 and 84876 in, we'll have a good basis to start from > (which is why I've waited). It's not notifications that's the issue; it's submission, querying and modification. Gerv From preed at sigkill.com Wed Feb 12 23:04:14 2003 From: preed at sigkill.com (J. Paul Reed) Date: Wed, 12 Feb 2003 15:04:14 -0800 (PST) Subject: Feedback from FOSDEM In-Reply-To: <3E4AC80B.5040104@mozilla.org> References: <3E4AC80B.5040104@mozilla.org> Message-ID: On Wed, 12 Feb 2003, Gervase Markham wrote: > > I think with 124174 and 84876 in, we'll have a good basis to start from > > (which is why I've waited). > > It's not notifications that's the issue; it's submission, querying and > modification. Right... but the mail subsystem, such as it is now, needs a lot of work. There's a significant desire to integrate things like bug_email into the mainline (bug 94850), and I'm all for that, but not until we get a stable foundation to build on, which includes interfaces to do that. I consider that part of my area, but as you point out, maybe it's not. Maybe we should change the "Email Notifications" component name to the "Email-fu" component. ;-) Later, Paul ----------------------------------------------------------------------- J. Paul Reed preed at sigkill.com || web.sigkill.com/preed To hold on to sanity too tight is insane. -- Nick Falzone, Pushing Tin From geeta.tripathi at bhartitelesoft.com Tue Feb 4 10:31:35 2003 From: geeta.tripathi at bhartitelesoft.com (geeta) Date: Tue, 4 Feb 2003 16:01:35 +0530 Subject: Regarding Bugzilla 2.16.1 References: <3E4A09EE.4050702@mozilla.org> Message-ID: <001801c2cc38$9c55d1a0$800310ac@tripathi.bhartitelesoft.com> Hello All, I have a small issue. I created a Product and a product group.However the users of that product group are not able to view the product in the query page the user is logged in. When I disable the usebuggroup and usebuggroupsentry,the user of the product group is able to view the product in the query page. What could be the issue? Or is there any specfic feature/bug in version 2.16.1. Please help, I am not able to give demo to my team to use it further. Regards, Geeta Tripathi ----- Original Message ----- From: "Gervase Markham" To: Sent: Wednesday, February 12, 2003 2:16 PM Subject: Feedback from FOSDEM > The audience at FOSDEM gave the following feedback on Bugzilla. > > We had one request for the "submit only my comment" feature when you > mid-air. > > During the post-talk discussion, there was a lot of support for a > "simpler" query page - not as simple as the simple search, but more > complex than the one-liner on the guided bug reporting form. This could > be b.m.o.-specific, or it could be general. There was also support for > changing the default query (used by those not logged in) to include > UNCONFIRMED and DUPLICATE bugs (i.e. resolution = --- OR DUPLICATE). > > A guy from Debian said "I saw all that stuff that Bugzilla does, and > thought 'We need all of that.' But there's no hope of Debian converting > without a decent email interface." I said that in order to have one > which is maintained, we need an organisation which is using it regularly > :-) But there might be scope here to expand our empire. > > Someone (maybe I'll get around to it, maybe not) needs to investigate > the email interface on debbugs and see how hard it would be to do > something like it. > > Gerv > > ---- > To view or change your list settings, click here: > > From jussi at comlink.fi Thu Feb 13 12:00:14 2003 From: jussi at comlink.fi (Jussi Sirpoma) Date: Thu, 13 Feb 2003 14:00:14 +0200 Subject: Feedback from FOSDEM In-Reply-To: <3E4A09EE.4050702@mozilla.org> References: <3E4A09EE.4050702@mozilla.org> Message-ID: <3E4B88CE.7050703@comlink.fi> On 12.2.2003 10:46 Gervase Markham spoke: > During the post-talk discussion, there was a lot of support for a > "simpler" query page - not as simple as the simple search, but more > complex than the one-liner on the guided bug reporting form. This could > be b.m.o.-specific, or it could be general. There was also support for I would like to see this kind of a simpler query page also. It would help people who are just getting familiar with bz and I think it would be usefull in the public bz distribution. It could contain for example the following fields from the query page: - Summary - Product - Comment - Some predefined status/resolution sets - Some way to restrict the query with time Jussi Sirpoma From usmlekaplan at yahoo.com Thu Feb 13 19:35:10 2003 From: usmlekaplan at yahoo.com (usmle Prep) Date: Thu, 13 Feb 2003 11:35:10 -0800 (PST) Subject: Bugzilla E-mail Interface /Bugzilla_email_append.pl Message-ID: <20030213193510.55068.qmail@web20205.mail.yahoo.com> When using bugzilla e-mail interface, once a bug is created using e-mail interface,Updates are possible only to the description field through e-mail interface.Do you know if any one implemented updates to all the fields using e-mail interface(Bugzilla_email_append.pl) . Thanks a lot, Srikant Nandamuri __________________________________________________ Do you Yahoo!? Yahoo! Shopping - Send Flowers for Valentine's Day http://shopping.yahoo.com From gerv at mozilla.org Sat Feb 15 00:30:08 2003 From: gerv at mozilla.org (Gervase Markham) Date: Sat, 15 Feb 2003 00:30:08 +0000 Subject: [Fwd: [Bug 193177] new report print bug counts and changes vs. time] Message-ID: <3E4D8A10.5090403@mozilla.org> Check this out, guys. Unfortunately, the patches are against 2.10, but it is very cool. Gerv -------- Original Message -------- Subject: [Bug 193177] new report print bug counts and changes vs. time Date: Thu, 13 Feb 2003 10:24:10 -0800 (PST) From: bugzilla-daemon at mozilla.org To: gerv at mozilla.org http://bugzilla.mozilla.org/show_bug.cgi?id=193177 ------- Additional Comments From software at fdd.com 2003-02-13 10:24 ------- Created an attachment (id=114353) --> (http://bugzilla.mozilla.org/attachment.cgi?id=114353&action=view) screenshot of bug changes, priority-opens/closes/verifies vs. time ------- Additional Comments From software at fdd.com 2003-02-13 10:21 ------- Created an attachment (id=114352) --> (http://bugzilla.mozilla.org/attachment.cgi?id=114352&action=view) screenshot of open count, priority vs. week From geeta.tripathi at bhartitelesoft.com Sat Feb 15 09:59:46 2003 From: geeta.tripathi at bhartitelesoft.com (geeta.tripathi at bhartitelesoft.com) Date: Sat, 15 Feb 2003 09:59:46 GMT Subject: [Fwd: [Bug 193177] new report print bug counts and changes vs. time] Message-ID: <200302150959.h1F9xjK20379@mail.bhartitelesoft.com> Hello Grev/All, I have a query with 2.16.1. When I enable usebuggroupsentry and usebuggroups parameter. The group for the product is automatocally created. However, the user when logs in cannot see the product in the query page and also does not have access to the product for entering the bug. What could be the reason???????????? PLease help!!!!!!!! regards, Geeta > Check this out, guys. Unfortunately, the patches are against 2.10, but > it is very cool. > > Gerv > > -------- Original Message -------- > Subject: [Bug 193177] new report print bug counts and changes vs. time > Date: Thu, 13 Feb 2003 10:24:10 -0800 (PST) > From: bugzilla-daemon at mozilla.org > To: gerv at mozilla.org > > http://bugzilla.mozilla.org/show_bug.cgi?id=193177 > > ------- Additional Comments From software at fdd.com 2003- 02-13 10:24 ------- > Created an attachment (id=114353) > --> (http://bugzilla.mozilla.org/attachment.cgi? id=114353&action=view) > screenshot of bug changes, priority-opens/closes/verifies vs. time > > ------- Additional Comments From software at fdd.com 2003- 02-13 10:21 ------- > Created an attachment (id=114352) > --> (http://bugzilla.mozilla.org/attachment.cgi? id=114352&action=view) > screenshot of open count, priority vs. week > > ---- > To view or change your list settings, click here: > > ---------------------------------------------------------- http://www.bhartitelesoft.com Telecommunication and E-Commerce solutions for the future. From geeta.tripathi at bhartitelesoft.com Sat Feb 15 12:34:26 2003 From: geeta.tripathi at bhartitelesoft.com (geeta.tripathi at bhartitelesoft.com) Date: Sat, 15 Feb 2003 12:34:26 GMT Subject: A problem with accessing product by user in Product group Message-ID: <200302151234.h1FCYJK21551@mail.bhartitelesoft.com> I have been sending this issue to everyone. I think this is the tenth time I am sending please respond to my mail sent below: Forwarded Message: > To: developers at bugzilla.org, developers at bugzilla.org > From: geeta.tripathi at bhartitelesoft.com > Subject: Re: [Fwd: [Bug 193177] new report print bug counts and changes vs. time] > Date: Sat, 15 Feb 2003 09:59:46 GMT > ----- > Hello Grev/All, > > I have a query with 2.16.1. When I enable usebuggroupsentry > and usebuggroups parameter. The group for the product is > automatocally created. > > However, the user when logs in cannot see the product in > the query page and also does not have access to the product > for entering the bug. > > What could be the reason???????????? > > > PLease help!!!!!!!! > > regards, > > Geeta > > > Check this out, guys. Unfortunately, the patches are > against 2.10, but > > it is very cool. > > > > Gerv > > > > -------- Original Message -------- > > Subject: [Bug 193177] new report print bug counts and > changes vs. time > > Date: Thu, 13 Feb 2003 10:24:10 -0800 (PST) > > From: bugzilla-daemon at mozilla.org > > To: gerv at mozilla.org > > > > http://bugzilla.mozilla.org/show_bug.cgi?id=193177 > > > > ------- Additional Comments From software at fdd.com 2003- > 02-13 10:24 ------- > > Created an attachment (id=114353) > > --> (http://bugzilla.mozilla.org/attachment.cgi? > id=114353&action=view) > > screenshot of bug changes, priority- opens/closes/verifies > vs. time > > > > ------- Additional Comments From software at fdd.com 2003- > 02-13 10:21 ------- > > Created an attachment (id=114352) > > --> (http://bugzilla.mozilla.org/attachment.cgi? > id=114352&action=view) > > screenshot of open count, priority vs. week > > > > ---- > > To view or change your list settings, click here: > > user=geeta.tripathi at bhartitelesoft.com> > > > > > ---------------------------------------------------------- > http://www.bhartitelesoft.com > Telecommunication and E-Commerce solutions for the future. > > > ---- > To view or change your list settings, click here: > > ---------------------------------------------------------- http://www.bhartitelesoft.com Telecommunication and E-Commerce solutions for the future. From preed at sigkill.com Sat Feb 15 12:41:12 2003 From: preed at sigkill.com (J. Paul Reed) Date: Sat, 15 Feb 2003 04:41:12 -0800 (PST) Subject: A problem with accessing product by user in Product group In-Reply-To: <200302151234.h1FCYJK21551@mail.bhartitelesoft.com> References: <200302151234.h1FCYJK21551@mail.bhartitelesoft.com> Message-ID: On Sat, 15 Feb 2003 geeta.tripathi at bhartitelesoft.com wrote: > I have been sending this issue to everyone. I think this is the tenth > time I am sending please respond to my mail sent below: As I remember, this is how bug groups work. Read the docs. Later, Paul ----------------------------------------------------------------------- J. Paul Reed preed at sigkill.com || web.sigkill.com/preed To hold on to sanity too tight is insane. -- Nick Falzone, Pushing Tin From bbaetz at acm.org Sat Feb 15 13:45:41 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Sun, 16 Feb 2003 00:45:41 +1100 Subject: mod_perl, or Make Bugzilla Fast! Message-ID: <20030215134541.GA25702@mango.home> I decided to try to find out what mod_perl does for us ATM: [bbaetz at mango src]$ time wget -o /dev/null -O /dev/null \ http://localhost/bugzilla/ real 0m0.331s user 0m0.002s sys 0m0.004s [bbaetz at mango src]$ time wget -o /dev/null -O /dev/null \ http://localhost/bugzilla-mp/ real 0m0.015s user 0m0.006s sys 0m0.000s Yes, that is over 22 times as fast. The first number is stable, the second varies from 0.14 to 0.30. (its a bit higher for the first run - arround 0.080s. I'm obviously not preloading something which I should be, or something). Now, it doesn't actually _work_. At least not if you want to do anything apart from look at the index page while not logged in. The logged in part should be fixable once my Auth patch lands, hopefully tomorrow now that justdave reminded me to use -N. FWIW, that also implies no db login after my latest Bugzilla.pm reorg, so that hasn't been mod_perl tested yet. And it does have some patches, mostly to deal with the fact that mp2 doesn't yet chdir into the current directory before running (doing so isn't threadsafe, basically). So far I patched the config code, and the template code, and little bits in other places, too. Then I hardcoded my path into $Bugzilla::libdir. :) But it does reload params from the file when they change (via stat). So, we're getting there.... Bradley From gerv at mozilla.org Sun Feb 16 12:03:44 2003 From: gerv at mozilla.org (Gervase Markham) Date: Sun, 16 Feb 2003 12:03:44 +0000 Subject: Hold the release! Message-ID: <3E4F7E20.40507@mozilla.org> Oh dear... I've just written the test tool described in bug 192677 http://bugzilla.mozilla.org/show_bug.cgi?id=192677 and, having tested it on a single template, I've already found one instance where an incoming FORM variable is echoed directly in the template. I'm having trouble exploiting it - not enough practice, perhaps. It's the format parameter to query.cgi if anyone's interested; you need to set query_format to a valid parameter to avoid getting an error. Anyway, I think we should hold off for a couple of days while we get the test checked in, and get people to split up the work of using it to check all the templates. Gerv From usmlekaplan at yahoo.com Mon Feb 17 22:00:22 2003 From: usmlekaplan at yahoo.com (usmle Prep) Date: Mon, 17 Feb 2003 14:00:22 -0800 (PST) Subject: How to update status of a bug from "NEW" to "RESOLVED" from shell Message-ID: <20030217220022.15621.qmail@web20209.mail.yahoo.com> Hello All, I am wondering if any one can help me in fixing this issue? query="http://xyz.com/bugzilla/buglist.cgi?cmd=doit?&bug_status=NEW&?&product=dvdchips" I use the above query with wget to get a list of all bugs whose status is NEW for a product dvdchips. What I am looking for is a similar .cgi file already in bugzilla that will help me update the status from NEW to RESOLVED from shell using wget. Thanks for all your help, Srikant Nandamuri __________________________________________________ Do you Yahoo!? Yahoo! Shopping - Send Flowers for Valentine's Day http://shopping.yahoo.com From geeta.tripathi at bhartitelesoft.com Sun Feb 9 04:35:10 2003 From: geeta.tripathi at bhartitelesoft.com (geeta) Date: Sun, 9 Feb 2003 10:05:10 +0530 Subject: odd db error message References: <1045010864.25294.19.camel@jonschatz-lx.vmware.com> Message-ID: <006b01c2cff4$a62975a0$800310ac@tripathi.bhartitelesoft.com> Hello David, I have been able to successfully give a demo in my organisation regarding Bugzilla(2.16.1) and we would be using it organisation wide for our Defect tracking activities. I wanted to thank you for your support and the forum for the feedbacks that has given me lot of support all these days. Still looking forward for your support as the requirement in the organistaion increases. Thanks and regards, Geeta ----- Original Message ----- From: "David Miller" To: Sent: Wednesday, February 12, 2003 6:25 AM Subject: Re: odd db error message > On 2/11/03 4:47 PM -0800, Jonathan Schatz wrote: > > > sometime in the past 2 weeks a cvs update caused the following error to > > start showing up in my error_log: > > > > [Tue Feb 11 16:20:30 2003] [error] [client 172.16.14.80] [Tue Feb 11 > > 16:20:30 2003] editopsys.cgi: DBI::db=HASH(0x864e8b8)->disconnect > > invalidates 1 active statement handle (either destroy statement handles > > or call finish on them before disconnecting) at Bugzilla.pm line 98., > > referer: http://jonschatz-lx/editopsys.cgi > > > > any ideas? > > Yep. See http://bugzilla.mozilla.org/show_bug.cgi?id=192531 > -- > 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 jon at vmware.com Wed Feb 19 17:28:21 2003 From: jon at vmware.com (Jonathan Schatz) Date: Wed, 19 Feb 2003 09:28:21 -0800 (PST) Subject: problems using deprecated db methods Message-ID: i've got old code that uses the deprecated db methods (SendSQL, etc). On the 2.16.1 branch the code works fine. However, since i've moved to cvs head, i've had problems with the scripts. basically, on the old branch db methods would die noisily upon error. On the new branch, db errors seem to fail silently. Is there a magic configuration option i need to set to get error messages out of the old functions? -jon -- Jonathan Schatz Engineering System Administrator VMware, Inc "Te occidere possunt sed te edere non possunt nefas est." From justdave at syndicomm.com Wed Feb 19 17:30:49 2003 From: justdave at syndicomm.com (David Miller) Date: Wed, 19 Feb 2003 12:30:49 -0500 Subject: problems using deprecated db methods In-Reply-To: References: Message-ID: On 2/19/03 9:28 AM -0800, Jonathan Schatz wrote: > i've got old code that uses the deprecated db methods (SendSQL, etc). On > the 2.16.1 branch the code works fine. However, since i've moved to cvs > head, i've had problems with the scripts. basically, on the old branch db > methods would die noisily upon error. On the new branch, db errors seem to > fail silently. Is there a magic configuration option i need to set to get > error messages out of the old functions? Heh. Funny you should ask. :) See http://bugzilla.mozilla.org/show_bug.cgi?id=193985 which was just filed last night because I complained about that very thing. :) -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From jon at vmware.com Wed Feb 19 19:30:13 2003 From: jon at vmware.com (Jonathan Schatz) Date: 19 Feb 2003 11:30:13 -0800 Subject: problems using deprecated db methods In-Reply-To: References: Message-ID: <1045683013.15487.2.camel@jonschatz-lx.vmware.com> On Wed, 2003-02-19 at 09:30, David Miller wrote: > See http://bugzilla.mozilla.org/show_bug.cgi?id=193985 which was just filed > last night because I complained about that very thing. :) right on. that patch solves everything. i actually complained about this yesterday afternoon, but realized i sent the email to developers at bugzilla.com, and i'm sure that car guy really doesn't care about our db issues... -jon -- Jonathan Schatz Engineering System Administrator VMware, Inc "Te occidere possunt sed te edere non possunt nefas est." From preed at sigkill.com Wed Feb 19 19:39:09 2003 From: preed at sigkill.com (J. Paul Reed) Date: Wed, 19 Feb 2003 11:39:09 -0800 (PST) Subject: problems using deprecated db methods In-Reply-To: <1045683013.15487.2.camel@jonschatz-lx.vmware.com> References: <1045683013.15487.2.camel@jonschatz-lx.vmware.com> Message-ID: On Wed, 19 Feb 2003, Jonathan Schatz wrote: > right on. that patch solves everything. i actually complained about this > yesterday afternoon, but realized i sent the email to > developers at bugzilla.com, and i'm sure that car guy really doesn't care > about our db issues... I could ask him if he wanted to help us by setting up a forward in that domain; I actually know him... he and I helped start the Nothern Colorado Linux Users Group (small world...) Later, Paul ----------------------------------------------------------------------- J. Paul Reed preed at sigkill.com || web.sigkill.com/preed To hold on to sanity too tight is insane. -- Nick Falzone, Pushing Tin From gerv at mozilla.org Wed Feb 19 22:29:37 2003 From: gerv at mozilla.org (Gervase Markham) Date: Wed, 19 Feb 2003 22:29:37 +0000 Subject: Bugzilla E-mail Interface /Bugzilla_email_append.pl In-Reply-To: <20030213193510.55068.qmail@web20205.mail.yahoo.com> References: <20030213193510.55068.qmail@web20205.mail.yahoo.com> Message-ID: <3E540551.2000702@mozilla.org> usmle Prep wrote: > When using bugzilla e-mail interface, once a bug is > created using e-mail interface,Updates are possible > only to the description field through e-mail > interface.Do you know if any one implemented updates > to all the fields using e-mail > interface(Bugzilla_email_append.pl) . It seems that no-one knows of anyone who has done this. :-| Gerv From timeless at myrealbox.com Fri Feb 21 17:04:34 2003 From: timeless at myrealbox.com (timeless) Date: Fri, 21 Feb 2003 12:04:34 -0500 Subject: Tinderbox scripts broken? Message-ID: <3E565C21.6070706@myrealbox.com> http://tinderbox.mozilla.org/showbuilds.cgi?tree=Bugzilla Gerv wrote: > "can't open batch-1.pl" and no columns. Anyone got any ideas? http://bonsai.mozilla.org/cvsblame.cgi?file=mozilla/webtools/tinderbox/showbuilds.cgi&rev=1.179&mark=552-553,558,567,572-573#551 Justdave: can you look into the relevant files and decide what to do? otherwise i can ask leaf/mcafee. From usmlekaplan at yahoo.com Fri Feb 21 17:24:25 2003 From: usmlekaplan at yahoo.com (usmle Prep) Date: Fri, 21 Feb 2003 09:24:25 -0800 (PST) Subject: Bugzilla E-mail Interface /Bugzilla_email_append.pl In-Reply-To: <3E540551.2000702@mozilla.org> Message-ID: <20030221172425.98276.qmail@web20203.mail.yahoo.com> Hi Gervase, Thanks for responding .Now that no one has it before I am working on that and will let you folks know once I make those changes. Thanks, Srikant --- Gervase Markham wrote: > usmle Prep wrote: > > When using bugzilla e-mail interface, once a bug > is > > created using e-mail interface,Updates are > possible > > only to the description field through e-mail > > interface.Do you know if any one implemented > updates > > to all the fields using e-mail > > interface(Bugzilla_email_append.pl) . > > It seems that no-one knows of anyone who has done > this. :-| > > Gerv > > ---- > To view or change your list settings, click here: > __________________________________________________ Do you Yahoo!? Yahoo! Tax Center - forms, calculators, tips, more http://taxes.yahoo.com/ From justdave at syndicomm.com Fri Feb 21 18:14:28 2003 From: justdave at syndicomm.com (David Miller) Date: Fri, 21 Feb 2003 13:14:28 -0500 Subject: Tinderbox scripts broken? In-Reply-To: <3E565C21.6070706@myrealbox.com> References: <3E565C21.6070706@myrealbox.com> Message-ID: On 2/21/03 12:04 PM -0500, timeless wrote: > http://tinderbox.mozilla.org/showbuilds.cgi?tree=Bugzilla > > Gerv wrote: > > "can't open batch-1.pl" and no columns. Anyone got any ideas? > > >http://bonsai.mozilla.org/cvsblame.cgi?file=mozilla/webtools/tinderbox/showbuilds.cgi&rev=1.179&mark=552-553,558,567,572-573#551 > > Justdave: can you look into the relevant files and decide what to do? > otherwise i can ask leaf/mcafee. What's this have to do with Bugzilla? I've never touched Tinderbox from the server side, and would have no clue what to look for. -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From timeless at bemail.org Fri Feb 21 17:53:42 2003 From: timeless at bemail.org (Josh Soref) Date: Fri, 21 Feb 2003 09:53:42 -0800 Subject: {Spam?} [Fwd: proposal 2] Message-ID: <200302211753.h1LHrgc14576@mail6.bigmailbox.com> An embedded and charset-unspecified text was scrubbed... Name: not available URL: From gerv at mozilla.org Fri Feb 21 18:32:15 2003 From: gerv at mozilla.org (Gervase Markham) Date: Fri, 21 Feb 2003 18:32:15 +0000 Subject: {Spam?} [Fwd: proposal 2] In-Reply-To: <200302211753.h1LHrgc14576@mail6.bigmailbox.com> References: <200302211753.h1LHrgc14576@mail6.bigmailbox.com> Message-ID: <3E5670AF.3050403@mozilla.org> Er... why have you sent us this? Gerv Josh Soref wrote: > Date: Fri, 11 Jan 2002 14:57:51 -0500 > From: jesus X > To: timeless at bemail.org > >>New proposal with summary at end. I think it still needs a little >>work... > > >>1. Current "Status" and "Resolution" selections insufficent for >>effective triaging in some circumstances. >> >>UNCONFIRMED and NEW are too constrictive and too vague as to the >>real state of a bug report. There seems to be a need for >>intermediate selections based on current triage status. >> >>Possible solutions: >> >>New potential statuses. >>CANREPRO - for reproduceable bugs (is this really necessary? dosen't >>NEW handle this?) >>NOREPRO - for a failure to reproduce problem >>NEEDREPRO - alternative to NOREPRO >>NEEDINFO - for bugs where the info provided by the reporter is >>insufficient, needs testcase, etc. >>NEW may need renamed, as it is unclear as to meaning "this is a >>valid bug" or "this bug was filed by a known-good source, but >>may/maynot be a valid bug". >>Depricate UNCONFIRMED and replace with 2 new states, UNTRIAGED and >>TRIAGED. >> >>-------- >>2. ASSIGNED seems only semi-useful at best, to some. Others see it >>as very useful. >> >>Possible Solutions: >>More fine grained descriptors. Specific rules about setting a bug >>ASSIGNED. >>Renaming ASSIGNED to ACCEPTED (this seems to have some healthy >>support) >>Create UNASSIGNED state for valid bugs with little chance to be >>fixed in the near/not-so-near future. >> >>-------- >>3. LATER may need to become depricated and replaced by the >>UNASSIGNED state. >> >> >>--------- >>Proposal: >>--------- >> >>Depricate or remove the following statuses: >> UNCONFIRMED, ASSIGNED, REMIND, LATER >> >>Add the following statuses: >> UNTRIAGED, TRIAGED, ACCEPTED, UNASSIGNED >> >>and add the following Resolutions for TRIAGED bugs: >> NEEDREPRO, NEEDINFO >> >>UNTRIAGED will replace UNCONFIRMED >> >>TRIAGED will have Resolution set to NEEDREPRO, NEEDINFO >> >>ACCEPTED will replace ASSIGNED when the bug is actually being >>actively worked on. >> >>UNASSIGNED will be for bugs that are valid, but will not be worked >>on for the forseeable future. >> >> >>Process: >> >>1. Newly filed reports will start out as UNTRIAGED >> >>2. Triagers will then change the bug to TRIAGED and either NEEDINFO >>or NEDREPRO if they cannot reproduce the bug or need more info from >>the reporter or other triagers, or an UNTRIAGED bug can be changed >>directly to NEW if they can reproduce it and it does not need info. >> >>3. NEW bugs can then either be ACCEPTED by an engineer, or be >>UNASSIGNED showing it is "up for grabs" and not being worked on at >>all. >> >> >>Summary: The total possible STATUSes will then be: >>UNTRIAGED, TRIAGED, NEW, ACCEPTED, UNASSIGNED, REOPENED, RESOLVED, >>VERIFIED, and CLOSED >> >>RESOLUTIONs would be: >>NEEDREPRO, NEEDINFO, FIXED, WONTFIX, DUPLICATE, WORKSFORME > > > ------------------------------------------------------------ > Free, BeOS-friendly email accounts: http://BeMail.org/ > BeOS News and Community: http://www.BeGroovy.com/ > > > --------------------------------------------------------------------- > Express yourself with a super cool email address from BigMailBox.com. > Hundreds of choices. It's free! > http://www.bigmailbox.com > --------------------------------------------------------------------- > ---- > To view or change your list settings, click here: > From timeless at myrealbox.com Sun Feb 23 09:26:50 2003 From: timeless at myrealbox.com (timeless) Date: Sun, 23 Feb 2003 04:26:50 -0500 Subject: {Spam?} [Fwd: proposal 2] In-Reply-To: <3E5670AF.3050403@mozilla.org> References: <200302211753.h1LHrgc14576@mail6.bigmailbox.com> <3E5670AF.3050403@mozilla.org> Message-ID: <3E5893DA.7030203@myrealbox.com> Gervase Markham wrote: > Er... why have you sent us this? because it went into a mailbox i don't actively read. anyway, reorganizing the resolutions/status is something we should consider. a better question: why was it marked as spam :) From justdave at syndicomm.com Sun Feb 23 09:39:02 2003 From: justdave at syndicomm.com (David Miller) Date: Sun, 23 Feb 2003 04:39:02 -0500 Subject: {Spam?} [Fwd: proposal 2] In-Reply-To: <3E5893DA.7030203@myrealbox.com> References: <200302211753.h1LHrgc14576@mail6.bigmailbox.com> <3E5670AF.3050403@mozilla.org> <3E5893DA.7030203@myrealbox.com> Message-ID: On 2/23/03 4:26 AM -0500, timeless wrote: > a better question: why was it marked as spam :) Because the ISP you mailed it from was flagged by Spamcop as a spam source. (and in several other blacklists, too, when I looked it up). -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From justdave at syndicomm.com Mon Feb 24 20:47:07 2003 From: justdave at syndicomm.com (David Miller) Date: Mon, 24 Feb 2003 15:47:07 -0500 Subject: Documentation links Message-ID: There's been a number of bugs filed on bugzilla.mozilla.org recently suggesting that we link to points in the documentation from various places in Bugzilla to give people quicker access to the docs (and maybe even point out to some people that they're actually there). I don't think it's all that bad of an idea, but the question then comes to be what do we link it to? The docs directory is a subdirectory of the Bugzilla directory, so in theory you could link to the local copy, but the permissions won't necessarily be set so the webserver can see it... Or we could link to the online versions, but that stands a chance of the documentation being newer than the code... Comments / Suggestions? -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From bugreport at peshkin.net Mon Feb 24 21:06:44 2003 From: bugreport at peshkin.net (Joel Peshkin) Date: Mon, 24 Feb 2003 13:06:44 -0800 Subject: Documentation links References: Message-ID: <3E5A8964.1000609@peshkin.net> David Miller wrote: > >I don't think it's all that bad of an idea, but the question then comes to >be what do we link it to? > > How about a parameter called documentbase with default values "docs/" that can be cleared to disable the links or can be pointed to bmo's copy?? The biggest issue I can see is making sure that the right page/anchor combination gets loaded. Should there be a cgi that serves up the right document section for each link? Othewise, the code has to know what filename contains each anchor. From justdave at syndicomm.com Mon Feb 24 21:12:01 2003 From: justdave at syndicomm.com (David Miller) Date: Mon, 24 Feb 2003 16:12:01 -0500 Subject: Documentation links In-Reply-To: <3E5A8964.1000609@peshkin.net> References: <3E5A8964.1000609@peshkin.net> Message-ID: On 2/24/03 1:06 PM -0800, Joel Peshkin wrote: > The biggest issue I can see is making sure that the right page/anchor > combination gets loaded. Should there be a cgi that serves up the right > document section for each link? Othewise, the code has to know what > filename contains each anchor. Jake's been working on that actually... the solution to that is getting the anchor names hard-coded in the docs so it doesn't generate random ones every time the HTML gets regenerated. The OS-specific installation instructions has already had this done to it. There's a lot to do still. -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From jimw at bugopolis.com Mon Feb 24 21:24:45 2003 From: jimw at bugopolis.com (Jim Walters) Date: 24 Feb 2003 13:24:45 -0800 Subject: Documentation links In-Reply-To: <3E5A8964.1000609@peshkin.net> References: <3E5A8964.1000609@peshkin.net> Message-ID: <1046121884.6224.359.camel@control> I was mulling this over while thinking about how to use Faq-O-Matic for the contextual help. Was leaning towards making a new table in MySQL and populating it with a help_id, URL table. Seemed to be the most flexible since I could point to the html docs, URLs for the Faq-O-Matic links or even to some HTML links for help in a non-English HTML document at some point in the future without touching the perl code. Admin code update the links via the web interface eventually (perhaps even pointing to bmo hosted documentation) and though the IDs would get cast in stone, new ids are cheap. Jim's 2 cents. On Mon, 2003-02-24 at 13:06, Joel Peshkin wrote: > David Miller wrote: > > > > >I don't think it's all that bad of an idea, but the question then comes to > >be what do we link it to? > > > > > > How about a parameter called documentbase with default values "docs/" > that can be cleared to disable the links or can be pointed to bmo's copy?? > The biggest issue I can see is making sure that the right page/anchor > combination gets loaded. Should there be a cgi that serves up the right > document section for each link? Othewise, the code has to know what > filename contains each anchor. > > > > > > ---- > 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 jake at bugzilla.org Mon Feb 24 22:00:47 2003 From: jake at bugzilla.org (Jake) Date: Mon, 24 Feb 2003 17:00:47 -0500 (EST) Subject: Documentation links In-Reply-To: References: Message-ID: <25796.208.20.196.122.1046124047.squirrel@waldo.homelinux.org> > The docs directory is a subdirectory of the Bugzilla directory, so in > theory you could link to the local copy, but the permissions won't > necessarily be set so the webserver can see it... Hmm... good point. > Or we could link to the online versions, but that stands a chance of the > documentation being newer than the code... One possible solution to this is by munging the version variable. If the minor version is an odd number, link to the tip docs otherwise link to the doc tree for that particular version. Of course this means that's it's even more imperrative for me to keep docs updated for older versions, but such is life :) > Comments / Suggestions? One thing to beware, niether solution is likely to work for everybody. Some people will not want to link to the outside world and others will not be able to view the docs/ directory, Joel's suggestion seems kinda reasonable, but it does require adding yet another Param() to our already crowed interface (need that multi-pane patch :) and we would still need to decide if our default is bugzilla.org or internal. Hopefully everything I said is logical :) From gerv at mozilla.org Mon Feb 24 23:33:39 2003 From: gerv at mozilla.org (Gervase Markham) Date: Mon, 24 Feb 2003 23:33:39 +0000 Subject: Documentation links In-Reply-To: References: Message-ID: <3E5AABD3.8090602@mozilla.org> David Miller wrote: > The docs directory is a subdirectory of the Bugzilla directory, so in > theory you could link to the local copy, but the permissions won't > necessarily be set so the webserver can see it... So we make sure .htaccess is set up so that you can. We control the visibility of Bugzilla files. It seems obvious to me that the best and easiest way to provide good, consistent, correct documentation for a Bugzilla system is to link to the documentation provided with it. This is pretty much guaranteed to match the system in use better than any other method. It can also easily be updated by the local administrator if they so choose. We may want to provide a way to turn documentation links off - but why would you want to remove the help? I can see there might be a localisation issue here, though. What are the chances of the Bugzilla Guide (or at least the user-facing part) appearing in other languages? Gerv From jake at bugzilla.org Tue Feb 25 03:05:23 2003 From: jake at bugzilla.org (Jake) Date: Mon, 24 Feb 2003 22:05:23 -0500 Subject: Documentation links In-Reply-To: <3E5AABD3.8090602@mozilla.org> References: <3E5AABD3.8090602@mozilla.org> Message-ID: <3E5ADD73.2010402@bugzilla.org> Gervase Markham wrote: > So we make sure .htaccess is set up so that you can. We control the > visibility of Bugzilla files. In theory that would work, but not always. Also, what will happen in the newinstall world where it will suposedly be possible to put CGIs and static files in seperate locations. > We may want to provide a way to turn documentation links off - but why > would you want to remove the help? Some people are funny like that :) > I can see there might be a localisation issue here, though. What are > the chances of the Bugzilla Guide (or at least the user-facing part) > appearing in other languages? I know of at least one effort to translate it into German (there were actually two seperate offers that were fortunately directed to eachother). However, like the l10n of templates, these will be available as additional downloads and not added to our CVS. From gerv at mozilla.org Tue Feb 25 07:59:14 2003 From: gerv at mozilla.org (Gervase Markham) Date: Tue, 25 Feb 2003 07:59:14 +0000 Subject: Documentation links In-Reply-To: <3E5ADD73.2010402@bugzilla.org> References: <3E5AABD3.8090602@mozilla.org> <3E5ADD73.2010402@bugzilla.org> Message-ID: <3E5B2252.8010004@mozilla.org> Jake wrote: > Gervase Markham wrote: > >> So we make sure .htaccess is set up so that you can. We control the >> visibility of Bugzilla files. > > In theory that would work, but not always. Also, what will happen in > the newinstall world where it will suposedly be possible to put CGIs and > static files in seperate locations. I know nothing about that. Perhaps the person working on newinstall would like to provide a summary of its capabilities? I do hope this option is not going to become the default. I very much like having all the Bugzilla files in the same place. IMO this only causes trouble - if one file can't be certain where another one is, then it's going to cause other problems than this. >> We may want to provide a way to turn documentation links off - but why >> would you want to remove the help? > > Some people are funny like that :) Then they can edit templates to indulge their funniness. We could make the links match a regexp, to make it easier for them. >> I can see there might be a localisation issue here, though. What are >> the chances of the Bugzilla Guide (or at least the user-facing part) >> appearing in other languages? > > I know of at least one effort to translate it into German (there were > actually two seperate offers that were fortunately directed to > eachother). However, like the l10n of templates, these will be > available as additional downloads and not added to our CVS. But do we need to incorporate it into the links? My take is no - they should set up content negotiation if they want multi-language help. Gerv From bbaetz at acm.org Tue Feb 25 08:07:25 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Tue, 25 Feb 2003 19:07:25 +1100 Subject: Documentation links In-Reply-To: <3E5B2252.8010004@mozilla.org> References: <3E5AABD3.8090602@mozilla.org> <3E5ADD73.2010402@bugzilla.org> <3E5B2252.8010004@mozilla.org> Message-ID: <20030225080725.GA1461@mango.home> On Tue, Feb 25, 2003 at 07:59:14AM +0000, Gervase Markham wrote: > I do hope this option is not going to become the default. I very much > like having all the Bugzilla files in the same place. IMO this only > causes trouble - if one file can't be certain where another one is, then > it's going to cause other problems than this. Note that it will be a requirement for mod_perl that the real path of the bugzilla .pms are placed in the server statup file, and we are probably going to want to remove |use lib| calls, too. The reason for this is that changing the path at run time affects subsequent users of the mod_perl instance, and also affects stuff like Apache::Reload. Plus, with mod_perlv2, we arne't run with our directory as . Bradley From tobias.burnus at physik.fu-berlin.de Tue Feb 25 08:54:23 2003 From: tobias.burnus at physik.fu-berlin.de (Tobias Burnus) Date: Tue, 25 Feb 2003 09:54:23 +0100 (CET) Subject: Documentation links In-Reply-To: <3E5AABD3.8090602@mozilla.org> Message-ID: Hi, On Mon, 24 Feb 2003, Gervase Markham wrote: > I can see there might be a localisation issue here, though. What are the > chances of the Bugzilla Guide (or at least the user-facing part) > appearing in other languages? At least for German they are good. I started translating it (http://bugzilla-de.sourceforge.net/documentation.html) and the 5th chapter available (http://www.softwaretesting.de/article/archive/9/). At present both versions are not even complete if they are merged :-( My version can be checked out to ./docs/de/html/ (as soon as I check in the html files, at present there are only sgml files in the repository). Tobias From justdave at syndicomm.com Tue Feb 25 09:03:04 2003 From: justdave at syndicomm.com (David Miller) Date: Tue, 25 Feb 2003 04:03:04 -0500 Subject: Documentation links In-Reply-To: References: Message-ID: On 2/25/03 9:54 AM +0100, Tobias Burnus wrote: > Hi, > > On Mon, 24 Feb 2003, Gervase Markham wrote: >> I can see there might be a localisation issue here, though. What are the >> chances of the Bugzilla Guide (or at least the user-facing part) >> appearing in other languages? > > At least for German they are good. I started translating it > (http://bugzilla-de.sourceforge.net/documentation.html) > and the 5th chapter available > (http://www.softwaretesting.de/article/archive/9/). > At present both versions are not even complete if they are merged :-( > > My version can be checked out to ./docs/de/html/ (as soon as I check in > the html files, at present there are only sgml files in the repository). Interesting thought.... someone (Gerv?) mentioned earlier letting the server do content-negotiation... how about putting them right in the same directory with the English docs, but making them *.html.de ? Default Apache config recognizes that as a language code and content-negotiates appropriately. :) -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From caseyg at chsamerica.com Wed Feb 26 00:51:17 2003 From: caseyg at chsamerica.com (Casey Gregoire) Date: Tue, 25 Feb 2003 19:51:17 -0500 Subject: Changes the default screen Message-ID: Short of changing the code it self, is there a way to make the first page you see the user's current bug list? 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 Feb 26 08:07:17 2003 From: gerv at mozilla.org (Gervase Markham) Date: Wed, 26 Feb 2003 08:07:17 +0000 Subject: Changes the default screen In-Reply-To: References: Message-ID: <3E5C75B5.4000304@mozilla.org> Casey Gregoire wrote: > Short of changing the code it self, is there a way to make the first page > you see the user's current bug list? No. If you want to change the code, one way would be to change template/en/default/global/index.html.tmpl to have an IFRAME in it with the relevant search. Otherwise, you'll need to alter index.cgi to redirect to buglist.cgi with the relevant search. Gerv From preed at sigkill.com Wed Feb 26 10:44:05 2003 From: preed at sigkill.com (J. Paul Reed) Date: Wed, 26 Feb 2003 02:44:05 -0800 Subject: Changes the default screen In-Reply-To: <3E5C75B5.4000304@mozilla.org>; from gerv@mozilla.org on Wed, Feb 26, 2003 at 08:07:17AM +0000 References: <3E5C75B5.4000304@mozilla.org> Message-ID: <20030226024405.C15830@sigkill.com> On 26 Feb 2003 at 08:07:17, Gervase Markham moved bits on my disk to say: > No. > > If you want to change the code, one way would be to change > template/en/default/global/index.html.tmpl to have an IFRAME in it with > the relevant search. Otherwise, you'll need to alter index.cgi to > redirect to buglist.cgi with the relevant search. This is dumb. If there's not a bug on this, one should be filed. Casey: you might do some searching in Bugzilla to see if that's the case; if not, file one. I'd be willing to fix this myself. 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 bbaetz at acm.org Wed Feb 26 11:01:40 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Wed, 26 Feb 2003 22:01:40 +1100 Subject: Changes the default screen In-Reply-To: References: Message-ID: <20030226110140.GA2319@mango.home> On Tue, Feb 25, 2003 at 07:51:17PM -0500, Casey Gregoire wrote: > Short of changing the code it self, is there a way to make the first page > you see the user's current bug list? > Well, the problem with that is that the user may not be logged in. What you could do is write a quick script to [untested]: confirm_login(); my $bugs_url = Param('mybugstemplate'); my $username = url_quote($::COOKIE{'Bugzilla_login'}); $bugs_url =~ s/%userid%/$username/; print Bugzilla->cgi->redirect("buglist.cgi?$bugs_url"); Or something along those lines. Bradley From gerv at mozilla.org Wed Feb 26 11:06:28 2003 From: gerv at mozilla.org (Gervase Markham) Date: Wed, 26 Feb 2003 11:06:28 +0000 Subject: Changes the default screen In-Reply-To: <20030226024405.C15830@sigkill.com> References: <3E5C75B5.4000304@mozilla.org> <20030226024405.C15830@sigkill.com> Message-ID: <3E5C9FB4.3040705@mozilla.org> > This is dumb. Why? He wants "the first thing the user sees" to be their buglist (however a particular person might define "my buglist".) The obvious answer is "set your buglist as your homepage." Currently, going to the root of a Bugzilla installation gives you a page with an overview of all the stuff you can do. This is sane, and the Expected Thing. If people don't want that, there are five-lines-of-code ways to change it (adding a redirect at the top of index.cgi, with the particular business logic you want to decide what "my buglist" is.) There are a load more bugs in the database whose effects are much more dumb than this one. Gerv From preed at sigkill.com Wed Feb 26 11:15:17 2003 From: preed at sigkill.com (J. Paul Reed) Date: Wed, 26 Feb 2003 03:15:17 -0800 Subject: Changes the default screen In-Reply-To: <3E5C9FB4.3040705@mozilla.org>; from gerv@mozilla.org on Wed, Feb 26, 2003 at 11:06:28AM +0000 References: <3E5C75B5.4000304@mozilla.org> <20030226024405.C15830@sigkill.com> <3E5C9FB4.3040705@mozilla.org> Message-ID: <20030226031517.D15830@sigkill.com> On 26 Feb 2003 at 11:06:28, Gervase Markham moved bits on my disk to say: > Why? He wants "the first thing the user sees" to be their buglist > (however a particular person might define "my buglist".) The obvious > answer is "set your buglist as your homepage." Well... *maybe* I read the bug wrong, which is why I didn't respond. Initially, my thought was "templates should be able to fix this." I thought he wanted it to be: go to (for instance) bugzilla.mozilla.org, click "log in" button on toolbar, login, and have first screen that pops up be a user's bug list. Now, I actually went and did that, and I guess it's still doable with templates, assuming that there would be a way to tell the buglist.cgi template "I'm prompting you to provide me with a list of bugs, but I don't know who the user is yet, i.e. 'GoAheadAndLogin' is set, so log the person in and then deal with my request." Assuming that buglist.cgi can't do that, maybe *that's* the bug. > There are a load more bugs in the database whose effects are much more > dumb than this one. Agreed. 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 justdave at syndicomm.com Wed Feb 26 16:43:00 2003 From: justdave at syndicomm.com (David Miller) Date: Wed, 26 Feb 2003 11:43:00 -0500 Subject: Changes the default screen In-Reply-To: <20030226031517.D15830@sigkill.com> References: <3E5C75B5.4000304@mozilla.org> <20030226024405.C15830@sigkill.com> <3E5C9FB4.3040705@mozilla.org> <20030226031517.D15830@sigkill.com> Message-ID: On 2/26/03 3:15 AM -0800, J. Paul Reed wrote: > Now, I actually went and did that, and I guess it's still doable with > templates, assuming that there would be a way to tell the buglist.cgi > template "I'm prompting you to provide me with a list of bugs, but I don't > know who the user is yet, i.e. 'GoAheadAndLogin' is set, so log the person > in and then deal with my request." > > Assuming that buglist.cgi can't do that, maybe *that's* the bug. buglist.cgi can't do that. It is a bug. http://bugzilla.mozilla.org/show_bug.cgi?id=149496 -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From willy at lukoil.uu.ru Wed Feb 26 17:30:33 2003 From: willy at lukoil.uu.ru (Vitaly Fedrushkov) Date: Wed, 26 Feb 2003 22:30:33 +0500 (YEKT) Subject: Documentation links In-Reply-To: <3E5AABD3.8090602@mozilla.org> Message-ID: Good $daytime, > Date: Mon, 24 Feb 2003 23:33:39 +0000 > From: Gervase Markham > To: developers at bugzilla.org > Subject: Re: Documentation links > I can see there might be a localisation issue here, though. What are the > chances of the Bugzilla Guide (or at least the user-facing part) > appearing in other languages? I was looking at my Apache log to tell what to translate first. Now I feel that complete and consistent html documentation is not something very often referred from Bugzilla UI. But pages explaining bug lifecycle (bugwritinghelp.html, confirmhelp.html, bug_status.html) are frequent fliers. BTW, at bug 182975 I was told that help pages are to be templatised. Given that documentation is generated from SGML anyway, why not convert it into templates at build time? Advantages: o Multilanguage environment made easy; o Ability to refer to actual site environment (known to TT); o Placement and access control issues (mentioned by David at thread start) go away. Regards, Willy. -- No easy hope or lies | Vitaly "Willy the Pooh" Fedrushkov Shall bring us to our goal, | Control Systems and Processes Division But iron sacrifice | LUKOIL Company, Chelyabinsk Branch Of Body, Will and Soul. | willy at lukoil.uu.ru +7 3512 620367 R.Kipling | VVF1-RIPE From myk at mozilla.org Wed Feb 26 20:46:59 2003 From: myk at mozilla.org (Myk Melez) Date: Wed, 26 Feb 2003 12:46:59 -0800 Subject: Changes the default screen In-Reply-To: References: Message-ID: <3E5D27C3.5030909@mozilla.org> Casey Gregoire wrote: >Short of changing the code it self, is there a way to make the first page >you see the user's current bug list? > > A larger, related issue is redesigning the front page: http://bugzilla.mozilla.org/show_bug.cgi?id=130835 -myk From bbaetz at acm.org Thu Feb 27 07:44:41 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Thu, 27 Feb 2003 18:44:41 +1100 Subject: Dropping 5.6.0 support Message-ID: <20030227074440.GA1521@mango.home> The latest DBI release has as a changelog item: NOTE: Future versions of the DBI *will not* support perl 5.6.0 or earlier. : perl 5.6.1 will be the minimum supported version. This is something happening in the future, not now, and I don't know of any specific feature we'd be needing from a newer DBI, but I think tha this should probably be considered advance warning, or something. Perhaps we should document this somewhere? Bradley From gerv at mozilla.org Thu Feb 27 08:06:43 2003 From: gerv at mozilla.org (Gervase Markham) Date: Thu, 27 Feb 2003 08:06:43 +0000 Subject: Dropping 5.6.0 support In-Reply-To: <20030227074440.GA1521@mango.home> References: <20030227074440.GA1521@mango.home> Message-ID: <3E5DC713.3080804@mozilla.org> Bradley Baetz wrote: > The latest DBI release has as a changelog item: > > NOTE: Future versions of the DBI *will not* support perl 5.6.0 or earlier. > : perl 5.6.1 will be the minimum supported version. > > This is something happening in the future, not now, and I don't know of > any specific feature we'd be needing from a newer DBI, but I think tha > this should probably be considered advance warning, or something. > Perhaps we should document this somewhere? What distributions or OSes shipped with 5.6.0 as the default? Gerv From bbaetz at acm.org Thu Feb 27 08:31:05 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Thu, 27 Feb 2003 19:31:05 +1100 Subject: Dropping 5.6.0 support In-Reply-To: <3E5DC713.3080804@mozilla.org> References: <20030227074440.GA1521@mango.home> <3E5DC713.3080804@mozilla.org> Message-ID: <20030227083105.GA1602@mango.home> On Thu, Feb 27, 2003 at 08:06:43AM +0000, Gervase Markham wrote: > What distributions or OSes shipped with 5.6.0 as the default? RH7.0, 7.1 and 7.2, although 7.2 has an official errata update to 5.6.1. Debian stable has 5.6.1, and the previous release (2.2) appears to have only had 5.005. Not sure about other distributions. Bradley From geeta.tripathi at bhartitelesoft.com Tue Feb 18 12:00:49 2003 From: geeta.tripathi at bhartitelesoft.com (geeta) Date: Tue, 18 Feb 2003 17:30:49 +0530 Subject: Regarding the Login in Bugzilla References: Message-ID: <002301c2d745$6587a960$800310ac@tripathi.bhartitelesoft.com> Hello David/Gervase, There is a small issue and concern with 2.16.1. In Bugzilla after loging in as a user whenever we try to access a new webform we have to always put our login id and password. Has anybody addressed the issue of login in each web form once the user has logged in. Looking forward to hear from you at the earliest Thanks and regards, Geeta ----- Original Message ----- From: "David Miller" To: Sent: Tuesday, February 25, 2003 2:33 PM Subject: Re: Documentation links > On 2/25/03 9:54 AM +0100, Tobias Burnus wrote: > > > Hi, > > > > On Mon, 24 Feb 2003, Gervase Markham wrote: > >> I can see there might be a localisation issue here, though. What are the > >> chances of the Bugzilla Guide (or at least the user-facing part) > >> appearing in other languages? > > > > At least for German they are good. I started translating it > > (http://bugzilla-de.sourceforge.net/documentation.html) > > and the 5th chapter available > > (http://www.softwaretesting.de/article/archive/9/). > > At present both versions are not even complete if they are merged :-( > > > > My version can be checked out to ./docs/de/html/ (as soon as I check in > > the html files, at present there are only sgml files in the repository). > > Interesting thought.... someone (Gerv?) mentioned earlier letting the > server do content-negotiation... > > how about putting them right in the same directory with the English docs, > but making them *.html.de ? Default Apache config recognizes that as a > language code and content-negotiates appropriately. :) > -- > 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 geeta.tripathi at bhartitelesoft.com Tue Feb 18 12:07:00 2003 From: geeta.tripathi at bhartitelesoft.com (geeta) Date: Tue, 18 Feb 2003 17:37:00 +0530 Subject: Fw: Regarding the Login in Bugzilla Message-ID: <000801c2d746$40f9f200$800310ac@tripathi.bhartitelesoft.com> Do I need to do any customisations for the same. regards, Geeta ----- Original Message ----- From: "geeta" To: Sent: Tuesday, February 18, 2003 5:30 PM Subject: Regarding the Login in Bugzilla > Hello David/Gervase, > > There is a small issue and concern with 2.16.1. > > In Bugzilla after loging in as a user whenever we try to access a new > webform we have to always put our login id and password. > > Has anybody addressed the issue of login in each web form once the user has > logged in. > > > Looking forward to hear from you at the earliest > > > Thanks and regards, > > Geeta > ----- Original Message ----- > From: "David Miller" > To: > Sent: Tuesday, February 25, 2003 2:33 PM > Subject: Re: Documentation links > > > > On 2/25/03 9:54 AM +0100, Tobias Burnus wrote: > > > > > Hi, > > > > > > On Mon, 24 Feb 2003, Gervase Markham wrote: > > >> I can see there might be a localisation issue here, though. What are > the > > >> chances of the Bugzilla Guide (or at least the user-facing part) > > >> appearing in other languages? > > > > > > At least for German they are good. I started translating it > > > (http://bugzilla-de.sourceforge.net/documentation.html) > > > and the 5th chapter available > > > (http://www.softwaretesting.de/article/archive/9/). > > > At present both versions are not even complete if they are merged :-( > > > > > > My version can be checked out to ./docs/de/html/ (as soon as I check in > > > the html files, at present there are only sgml files in the repository). > > > > Interesting thought.... someone (Gerv?) mentioned earlier letting the > > server do content-negotiation... > > > > how about putting them right in the same directory with the English docs, > > but making them *.html.de ? Default Apache config recognizes that as a > > language code and content-negotiates appropriately. :) > > -- > > 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: > > > m> > > > > ---- > To view or change your list settings, click here: > > From preed at sigkill.com Thu Feb 27 12:09:54 2003 From: preed at sigkill.com (J. Paul Reed) Date: Thu, 27 Feb 2003 04:09:54 -0800 Subject: Regarding the Login in Bugzilla In-Reply-To: <002301c2d745$6587a960$800310ac@tripathi.bhartitelesoft.com>; from geeta.tripathi@bhartitelesoft.com on Tue, Feb 18, 2003 at 05:30:49PM +0530 References: <002301c2d745$6587a960$800310ac@tripathi.bhartitelesoft.com> Message-ID: <20030227040954.B23574@sigkill.com> On 18 Feb 2003 at 17:30:49, geeta moved bits on my disk to say: > In Bugzilla after loging in as a user whenever we try to access a new > webform we have to always put our login id and password. > > Has anybody addressed the issue of login in each web form once the user > has logged in. About a 90% chance it's the problem addressed in comment 16 on bug 165827. Read the documentation for details on the 'cookiepath' parameter. 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 sergeyli at pisem.net Thu Feb 27 14:08:21 2003 From: sergeyli at pisem.net (Sergey A. Lipnevich) Date: Thu, 27 Feb 2003 09:08:21 -0500 Subject: Fw: Regarding the Login in Bugzilla In-Reply-To: <000801c2d746$40f9f200$800310ac@tripathi.bhartitelesoft.com> References: <000801c2d746$40f9f200$800310ac@tripathi.bhartitelesoft.com> Message-ID: <3E5E1BD5.6060509@pisem.net> Just to repeat, 1) set the cookiepath properly and 2) make sure you clean the browser cookie cache before testing your changes. #2 is essential. geeta wrote: > Do I need to do any customisations for the same. > > regards, > > Geeta > ----- Original Message ----- > From: "geeta" > To: > Sent: Tuesday, February 18, 2003 5:30 PM > Subject: Regarding the Login in Bugzilla > > > >>Hello David/Gervase, >> >>There is a small issue and concern with 2.16.1. >> >>In Bugzilla after loging in as a user whenever we try to access a new >>webform we have to always put our login id and password. >> >>Has anybody addressed the issue of login in each web form once the user > > has > >>logged in. >> >> >>Looking forward to hear from you at the earliest >> >> From justdave at syndicomm.com Thu Feb 27 16:14:13 2003 From: justdave at syndicomm.com (David Miller) Date: Thu, 27 Feb 2003 11:14:13 -0500 Subject: Dropping 5.6.0 support In-Reply-To: <20030227083105.GA1602@mango.home> References: <20030227074440.GA1521@mango.home> <3E5DC713.3080804@mozilla.org> <20030227083105.GA1602@mango.home> Message-ID: On 2/27/03 7:31 PM +1100, Bradley Baetz wrote: > On Thu, Feb 27, 2003 at 08:06:43AM +0000, Gervase Markham wrote: >> What distributions or OSes shipped with 5.6.0 as the default? > > RH7.0, 7.1 and 7.2, although 7.2 has an official errata update to 5.6.1. > > Debian stable has 5.6.1, and the previous release (2.2) appears to have > only had 5.005. > > Not sure about other distributions. Mac OS X still ships with 5.6.0 as of 10.2.4. There's a rumor that 10.2.5 or 10.3 may have 5.8 on it. There is a 5.8 binary package available currently for 10.2.x from a third party. (http://www.versiontracker.com/moreinfo.fcgi?id=17636&db=mac) -- Dave Miller Project Leader, Bugzilla Bug Tracking System http://www.justdave.net/ http://www.bugzilla.org/ From tobias.burnus at physik.fu-berlin.de Thu Feb 27 17:18:53 2003 From: tobias.burnus at physik.fu-berlin.de (Tobias Burnus) Date: Thu, 27 Feb 2003 18:18:53 +0100 (CET) Subject: Dropping 5.6.0 support In-Reply-To: <3E5DC713.3080804@mozilla.org> Message-ID: Hi, On Thu, 27 Feb 2003, Gervase Markham wrote: > What distributions or OSes shipped with 5.6.0 as the default? Hewlett-Packard/Compaq's Tru64 Unix in the version 5.1a (maybe 5.1b has a newer perl ...) We use Debian Woody (and SuSE Linux 8.{1,2b1}) which have both a newer Perl, though. Besides we have our own (/usr/local/bin/perl: v5.6.1). Tobias From bbaetz at acm.org Thu Feb 27 20:11:57 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Fri, 28 Feb 2003 07:11:57 +1100 Subject: Dropping 5.6.0 support In-Reply-To: References: <20030227074440.GA1521@mango.home> <3E5DC713.3080804@mozilla.org> <20030227083105.GA1602@mango.home> Message-ID: <20030227201157.GA1222@mango.home> On Thu, Feb 27, 2003 at 11:14:13AM -0500, David Miller wrote: > > Mac OS X still ships with 5.6.0 as of 10.2.4. There's a rumor that 10.2.5 > or 10.3 may have 5.8 on it. There is a 5.8 binary package available > currently for 10.2.x from a third party. > (http://www.versiontracker.com/moreinfo.fcgi?id=17636&db=mac) Hmmm. Note that I'm not planning to do this until we actually need to, but just thought that we should document it as something likely to happen in the future. Hopefully that future is post bugzilla-2.18, though. Bradley From geeta.tripathi at bhartitelesoft.com Wed Feb 19 04:29:20 2003 From: geeta.tripathi at bhartitelesoft.com (geeta) Date: Wed, 19 Feb 2003 09:59:20 +0530 Subject: Fw: Regarding the Login in Bugzilla References: <000801c2d746$40f9f200$800310ac@tripathi.bhartitelesoft.com> <3E5E1BD5.6060509@pisem.net> Message-ID: <002101c2d7cf$b7fe65a0$800310ac@tripathi.bhartitelesoft.com> Thank you. It is now working. regards, Geeta ----- Original Message ----- From: "Sergey A. Lipnevich" To: Sent: Thursday, February 27, 2003 7:38 PM Subject: Re: Fw: Regarding the Login in Bugzilla > Just to repeat, 1) set the cookiepath properly and 2) make sure you > clean the browser cookie cache before testing your changes. #2 is essential. > > geeta wrote: > > Do I need to do any customisations for the same. > > > > regards, > > > > Geeta > > ----- Original Message ----- > > From: "geeta" > > To: > > Sent: Tuesday, February 18, 2003 5:30 PM > > Subject: Regarding the Login in Bugzilla > > > > > > > >>Hello David/Gervase, > >> > >>There is a small issue and concern with 2.16.1. > >> > >>In Bugzilla after loging in as a user whenever we try to access a new > >>webform we have to always put our login id and password. > >> > >>Has anybody addressed the issue of login in each web form once the user > > > > has > > > >>logged in. > >> > >> > >>Looking forward to hear from you at the earliest > >> > >> > > > ---- > To view or change your list settings, click here: > > From geeta.tripathi at bhartitelesoft.com Wed Feb 19 04:29:44 2003 From: geeta.tripathi at bhartitelesoft.com (geeta) Date: Wed, 19 Feb 2003 09:59:44 +0530 Subject: Regarding the Login in Bugzilla References: <002301c2d745$6587a960$800310ac@tripathi.bhartitelesoft.com> <20030227040954.B23574@sigkill.com> Message-ID: <002401c2d7cf$f48eda40$800310ac@tripathi.bhartitelesoft.com> Thank you it is now working. Regards, Geeta ----- Original Message ----- From: "J. Paul Reed" To: Sent: Thursday, February 27, 2003 5:39 PM Subject: Re: Regarding the Login in Bugzilla > On 18 Feb 2003 at 17:30:49, geeta moved bits on my disk to say: > > > In Bugzilla after loging in as a user whenever we try to access a new > > webform we have to always put our login id and password. > > > > Has anybody addressed the issue of login in each web form once the user > > has logged in. > > About a 90% chance it's the problem addressed in comment 16 on bug 165827. > > Read the documentation for details on the 'cookiepath' parameter. > > 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 > ---- > To view or change your list settings, click here: > > From geeta.tripathi at bhartitelesoft.com Wed Feb 19 09:37:07 2003 From: geeta.tripathi at bhartitelesoft.com (geeta) Date: Wed, 19 Feb 2003 15:07:07 +0530 Subject: bug life cycle in Bugzilla References: <002301c2d745$6587a960$800310ac@tripathi.bhartitelesoft.com> <20030227040954.B23574@sigkill.com> Message-ID: <000d01c2d7fa$7bb2b3a0$800310ac@tripathi.bhartitelesoft.com> Hello All, Now there are queries regarding Bugzilla which needs to be addressed very soon, please let me know incase any code changes required: 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 . Please let me know if I missing on-anything. Looking forward to hear from you. Thanks and regards, Geeta ----- Original Message ----- From: "J. Paul Reed" To: Sent: Thursday, February 27, 2003 5:39 PM Subject: Re: Regarding the Login in Bugzilla > On 18 Feb 2003 at 17:30:49, geeta moved bits on my disk to say: > > > In Bugzilla after loging in as a user whenever we try to access a new > > webform we have to always put our login id and password. > > > > Has anybody addressed the issue of login in each web form once the user > > has logged in. > > About a 90% chance it's the problem addressed in comment 16 on bug 165827. > > Read the documentation for details on the 'cookiepath' parameter. > > 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 > ---- > To view or change your list settings, click here: > > From geeta.tripathi at bhartitelesoft.com Wed Feb 19 10:13:40 2003 From: geeta.tripathi at bhartitelesoft.com (geeta) Date: Wed, 19 Feb 2003 15:43:40 +0530 Subject: Fw: bug life cycle in Bugzilla Message-ID: <001001c2d7ff$971715a0$800310ac@tripathi.bhartitelesoft.com> Hi Dave/Gervase/ 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. Do I need to make some major changes in the code. Please respond. Looking forward to hear from you. Thanks and regards, Geeta Tripathi ----- Original Message ----- From: "geeta" To: Sent: Wednesday, February 19, 2003 3:07 PM Subject: bug life cycle in Bugzilla > Hello All, > > Now there are queries regarding Bugzilla which needs to be addressed very > soon, please let me know incase any code changes required: > > 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 . > > Please let me know if I missing on-anything. > > Looking forward to hear from you. > > > Thanks and regards, > > Geeta > ----- Original Message ----- > From: "J. Paul Reed" > To: > Sent: Thursday, February 27, 2003 5:39 PM > Subject: Re: Regarding the Login in Bugzilla > > > > On 18 Feb 2003 at 17:30:49, geeta moved bits on my disk to say: > > > > > In Bugzilla after loging in as a user whenever we try to access a new > > > webform we have to always put our login id and password. > > > > > > Has anybody addressed the issue of login in each web form once the user > > > has logged in. > > > > About a 90% chance it's the problem addressed in comment 16 on bug 165827. > > > > Read the documentation for details on the 'cookiepath' parameter. > > > > 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 > > ---- > > To view or change your list settings, click here: > > > m> > > > > ---- > To view or change your list settings, click here: > > From cmason at somanetworks.com Fri Feb 28 16:08:33 2003 From: cmason at somanetworks.com (Chris L. Mason) Date: Fri, 28 Feb 2003 11:08:33 -0500 Subject: Support for multiple "Locations" in bugzilla Message-ID: <20030228160833.GH16925@somanetworks.com> [ this is a copy of my post to netscape.public.mozilla.webtools ] Hi, My company uses bugzilla extensively internally. We use it both for our product development and for IT "ticket" issues. We've done quite a lot of custom development, but a lot of it was hacked together quickly and isn't implemented in the best way. Our customized version is based on version 2.14. I'm currently reviewing these customizations and will be reimplementing them in 2.16.2. Before doing so, I thought it might be a good idea to check if any of our customizations can be generalized and submitted back to the bugzilla project. The first thing I'll be working on is support for multiple "locations". We've added the ability to set a Location for each bug. This was done so that people in one office can just check for bugs related to their office. (i.e. a desktop user with a problem in San Francisco doesn't concern an IT staffer in Toronto.) I would imagine that support for multiple locations could probably be used more generically by other people as well. So, I guess the question is, would this be of use? Would anyone be interested in a patch that adds this feature? (Obviously I'd follow the Developer Guidelines and try to submit some good clean code.) 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 preed at sigkill.com Fri Feb 28 20:21:06 2003 From: preed at sigkill.com (J. Paul Reed) Date: Fri, 28 Feb 2003 12:21:06 -0800 Subject: Support for multiple "Locations" in bugzilla In-Reply-To: <20030228160833.GH16925@somanetworks.com>; from cmason@somanetworks.com on Fri, Feb 28, 2003 at 11:08:33AM -0500 References: <20030228160833.GH16925@somanetworks.com> Message-ID: <20030228122106.G29082@sigkill.com> On 28 Feb 2003 at 11:08:33, Chris L. Mason moved bits on my disk to say: > The first thing I'll be working on is support for multiple "locations". > We've added the ability to set a Location for each bug. This was done so > that people in one office can just check for bugs related to their office. Isn't this just a special case of custom fields? Read all about *that* can of worms in bug 91037. 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 cmason at somanetworks.com Fri Feb 28 20:44:14 2003 From: cmason at somanetworks.com (Chris L. Mason) Date: Fri, 28 Feb 2003 15:44:14 -0500 Subject: Support for multiple "Locations" in bugzilla In-Reply-To: <20030228122106.G29082@sigkill.com> References: <20030228160833.GH16925@somanetworks.com> <20030228122106.G29082@sigkill.com> Message-ID: <20030228204414.GJ29540@somanetworks.com> On Fri, Feb 28, 2003 at 12:21:06PM -0800, J. Paul Reed wrote: > On 28 Feb 2003 at 11:08:33, Chris L. Mason moved bits on my disk to say: > > > The first thing I'll be working on is support for multiple "locations". > > We've added the ability to set a Location for each bug. This was done so > > that people in one office can just check for bugs related to their office. > > Isn't this just a special case of custom fields? > > Read all about *that* can of worms in bug 91037. > Interesting. Why not go further and take out the hard-coding of *all* fields and define them all in a central config file or library with all the necessary attributes required to support the current functionality. :) Anyway, I notice that this patch has been around for awhile and still hasn't been intregrated into CVS, so I wouldn't want to wait for it. My goal is to get this functionality accepted into the official bugzilla release, in order to keep our custom patches to a mininum (and reduce the effort involved in migrating to new versions.) I'd appreciate any suggestions you might have for me to help make this more likely. :) I'm currently implementing locations similarly to target_milestones so that the locations can vary by product and can be edited on the product page. Btw, another customization we have is the addition of a "system" milestone. This is because the "thing" we make has many sub-packages that each have their own numbering scheme. So that numbering scheme is handled with the versions and target milestones, and the "system" milestone is used for the "distribution" release we ship. This is similar in concept to the version of a Linux distro such as Debian 3.0 or RedHat 8.0 that contains many packages, each with it's own independant version number. Is there perhaps a better way to handle this? Can anyone else see a use for this? 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 Fri Feb 28 20:54:47 2003 From: gerv at mozilla.org (Gervase Markham) Date: Fri, 28 Feb 2003 20:54:47 +0000 Subject: Support for multiple "Locations" in bugzilla In-Reply-To: <20030228204414.GJ29540@somanetworks.com> References: <20030228160833.GH16925@somanetworks.com> <20030228122106.G29082@sigkill.com> <20030228204414.GJ29540@somanetworks.com> Message-ID: <3E5FCC97.1000700@mozilla.org> Chris L. Mason wrote: > I'd appreciate any > suggestions you might have for me to help make this more likely. :) I hope this isn't too disappointing, but I don't think there's much chance of either of these getting in. For all the fuss we are making about custom fields, that _is_ the right way to solve this problem, and I'm sure someone will get around to implementing them before too long. But in the mean time, we don't want to do anything which will make the job harder - and more fixed fields means that. Particularly given that you are the first person to ask for the two fields you are using, so they probably don't have wide applicability. Gerv From preed at sigkill.com Fri Feb 28 20:57:25 2003 From: preed at sigkill.com (J. Paul Reed) Date: Fri, 28 Feb 2003 12:57:25 -0800 Subject: Support for multiple "Locations" in bugzilla In-Reply-To: <3E5FCC97.1000700@mozilla.org>; from gerv@mozilla.org on Fri, Feb 28, 2003 at 08:54:47PM +0000 References: <20030228160833.GH16925@somanetworks.com> <20030228122106.G29082@sigkill.com> <20030228204414.GJ29540@somanetworks.com> <3E5FCC97.1000700@mozilla.org> Message-ID: <20030228125725.I29082@sigkill.com> On 28 Feb 2003 at 20:54:47, Gervase Markham moved bits on my disk to say: > I hope this isn't too disappointing, but I don't think there's much > chance of either of these getting in. For all the fuss we are making > about custom fields, that _is_ the right way to solve this problem, and > I'm sure someone will get around to implementing them before too long. Or, conversely, you, Chris, could update the custom fields patch to HEAD and get it pushed in. ;-) 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 cmason at somanetworks.com Fri Feb 28 21:09:19 2003 From: cmason at somanetworks.com (Chris L. Mason) Date: Fri, 28 Feb 2003 16:09:19 -0500 Subject: Support for multiple "Locations" in bugzilla In-Reply-To: <3E5FCC97.1000700@mozilla.org> References: <20030228160833.GH16925@somanetworks.com> <20030228122106.G29082@sigkill.com> <20030228204414.GJ29540@somanetworks.com> <3E5FCC97.1000700@mozilla.org> Message-ID: <20030228210919.GK29540@somanetworks.com> On Fri, Feb 28, 2003 at 08:54:47PM +0000, Gervase Markham wrote: > Chris L. Mason wrote: > >I'd appreciate any > >suggestions you might have for me to help make this more likely. :) > > I hope this isn't too disappointing, but I don't think there's much > chance of either of these getting in. For all the fuss we are making > about custom fields, that _is_ the right way to solve this problem, and > I'm sure someone will get around to implementing them before too long. > But in the mean time, we don't want to do anything which will make the > job harder - and more fixed fields means that. Particularly given that > you are the first person to ask for the two fields you are using, so > they probably don't have wide applicability. > Fair enough. What's the main problem with the custom fields patch? What's holding it back from being merged into the tree? Chris -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 232 bytes Desc: not available URL: From cmason at somanetworks.com Fri Feb 28 21:17:23 2003 From: cmason at somanetworks.com (Chris L. Mason) Date: Fri, 28 Feb 2003 16:17:23 -0500 Subject: Support for multiple "Locations" in bugzilla In-Reply-To: <20030228125725.I29082@sigkill.com> References: <20030228160833.GH16925@somanetworks.com> <20030228122106.G29082@sigkill.com> <20030228204414.GJ29540@somanetworks.com> <3E5FCC97.1000700@mozilla.org> <20030228125725.I29082@sigkill.com> Message-ID: <20030228211723.GA26446@somanetworks.com> On Fri, Feb 28, 2003 at 12:57:25PM -0800, J. Paul Reed wrote: > On 28 Feb 2003 at 20:54:47, Gervase Markham moved bits on my disk to say: > > > I hope this isn't too disappointing, but I don't think there's much > > chance of either of these getting in. For all the fuss we are making > > about custom fields, that _is_ the right way to solve this problem, and > > I'm sure someone will get around to implementing them before too long. > > Or, conversely, you, Chris, could update the custom fields patch to HEAD > and get it pushed in. ;-) > I'll look into this. 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 Fri Feb 28 21:19:35 2003 From: gerv at mozilla.org (Gervase Markham) Date: Fri, 28 Feb 2003 21:19:35 +0000 Subject: Support for multiple "Locations" in bugzilla In-Reply-To: <20030228210919.GK29540@somanetworks.com> References: <20030228160833.GH16925@somanetworks.com> <20030228122106.G29082@sigkill.com> <20030228204414.GJ29540@somanetworks.com> <3E5FCC97.1000700@mozilla.org> <20030228210919.GK29540@somanetworks.com> Message-ID: <3E5FD267.8040306@mozilla.org> Chris L. Mason wrote: > Fair enough. What's the main problem with the custom fields patch? > What's holding it back from being merged into the tree? Er... it's architected wrong? :-) As I understand the situation, the extensive discussion we've been having on the subject of custom field recently, here and in the newsgroup (which has been a good thing, definitely) hasn't come to a definite conclusion on exactly how to do them... but I also understand that the current patch doesn't work like any of the front-running candidate ideas. But I could be wrong about that last bit. This is getting a bit silly. We've had several discussions which have petered out. We need to sit down, either in a dedicated "we don't stop until we've designed this" email session or on IRC, and work out a detailed design. Then anyone (including Chris ;-) could go away and implement it. Gerv From cmason at somanetworks.com Fri Feb 28 21:30:30 2003 From: cmason at somanetworks.com (Chris L. Mason) Date: Fri, 28 Feb 2003 16:30:30 -0500 Subject: Support for multiple "Locations" in bugzilla In-Reply-To: <3E5FD267.8040306@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> Message-ID: <20030228213030.GN29540@somanetworks.com> On Fri, Feb 28, 2003 at 09:19:35PM +0000, Gervase Markham wrote: > Chris L. Mason wrote: > >Fair enough. What's the main problem with the custom fields patch? > >What's holding it back from being merged into the tree? > > Er... it's architected wrong? :-) > > As I understand the situation, the extensive discussion we've been > having on the subject of custom field recently, here and in the > newsgroup (which has been a good thing, definitely) hasn't come to a > definite conclusion on exactly how to do them... but I also understand > that the current patch doesn't work like any of the front-running > candidate ideas. But I could be wrong about that last bit. > > This is getting a bit silly. We've had several discussions which have > petered out. We need to sit down, either in a dedicated "we don't stop > until we've designed this" email session or on IRC, and work out a > detailed design. Then anyone (including Chris ;-) could go away and > implement it. > Okay, let me do some research on the "backstory" here. :) I'm going to check with my boss to see if he will approve my working on this (more work short-term, less work long-term) and if so, I'll try to come up with an architecture proposal and bounce it off you guys. If there's already a proposal that looks like it's just about there, I'd appreciate a pointer. 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 Fri Feb 28 21:53:56 2003 From: gerv at mozilla.org (Gervase Markham) Date: Fri, 28 Feb 2003 21:53:56 +0000 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: <3E5FDA74.4040407@mozilla.org> > Okay, let me do some research on the "backstory" here. :) > > I'm going to check with my boss to see if he will approve my working on > this (more work short-term, less work long-term) and if so, I'll try to > come up with an architecture proposal and bounce it off you guys. > > If there's already a proposal that looks like it's just about there, I'd > appreciate a pointer. If you are going to do an architecture proposal, there's discussions that happened on this mailing list, and probably in the newsgroup (netscape.public.mozilla.webtools) as well that you'll need to read. You'll need to read the bug, of course. But, I was rather hoping that bbaetz, as our database guru, would lay down the law about the right way to do this. Gerv From preed at sigkill.com Fri Feb 28 21:55:26 2003 From: preed at sigkill.com (J. Paul Reed) Date: Fri, 28 Feb 2003 13:55:26 -0800 Subject: Support for multiple "Locations" in bugzilla In-Reply-To: <3E5FD267.8040306@mozilla.org>; from gerv@mozilla.org on Fri, Feb 28, 2003 at 09:19:35PM +0000 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> Message-ID: <20030228135526.A32374@sigkill.com> On 28 Feb 2003 at 21:19:35, Gervase Markham moved bits on my disk to say: > This is getting a bit silly. We've had several discussions which have > petered out. We need to sit down, either in a dedicated "we don't stop > until we've designed this" email session or on IRC, and work out a > detailed design. Let's get ready to rumble!! ;-) I agree... I think the last time the discussion was sidetracked on whether or not we should have them at all; was there ever a clear consensus that we do? 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 bbaetz at acm.org Fri Feb 28 22:22:05 2003 From: bbaetz at acm.org (Bradley Baetz) Date: Sat, 1 Mar 2003 09:22:05 +1100 Subject: Support for multiple "Locations" in bugzilla In-Reply-To: <3E5FDA74.4040407@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> Message-ID: <20030228222205.GA1614@mango.home> On Fri, Feb 28, 2003 at 09:53:56PM +0000, Gervase Markham wrote: > > But, I was rather hoping that bbaetz, as our database guru, would lay > down the law about the right way to do this. 'database guru'? :) Anyway, I thought I'd mentioned the correct idea several times before. Use mapping tables. This (or any other solution, really) requires most of the fielddefs table hacks, such as the mailhead column, to be removed. The extra joins are very unlikly to be noticed except in buglist.cgi, and it may be possible to only do that join once, although I'd have to think about it. Hows postgres support coming along? :) > > Gerv > Bradley From gerv at mozilla.org Fri Feb 28 22:47:58 2003 From: gerv at mozilla.org (Gervase Markham) Date: Fri, 28 Feb 2003 22:47:58 +0000 Subject: Support for multiple "Locations" in bugzilla In-Reply-To: <20030228222205.GA1614@mango.home> 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> Message-ID: <3E5FE71E.6090700@mozilla.org> Bradley Baetz wrote: > On Fri, Feb 28, 2003 at 09:53:56PM +0000, Gervase Markham wrote: > >>But, I was rather hoping that bbaetz, as our database guru, would lay >>down the law about the right way to do this. > > 'database guru'? :) Well, if anyone gets that title, you do ;-) > Anyway, I thought I'd mentioned the correct idea several times before. > Use mapping tables. This (or any other solution, really) requires most > of the fielddefs table hacks, such as the mailhead column, to be > removed. A Google for 'database "mapping tables"' doesn't define the term. Do you mean lots of tables like: customers bug_id value 1 FooCo 2 FooCo 1 Bar Corp. 3 Baz Ltd. 4 Quux, Inc. 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? > 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. Gerv