Old Discussion: Python Implementation? (Bugzilla:Languages)

[Matthew Bogosian]

I have been casually fantasizing about reimplementing Bugzilla in Python. I am aware of the Bugzilla:Languages page (and corresponding discussion page), but note that it hasn't undergone any revisions for over four years, and a lot has changed since then. For example, many of the "Cons" listed under Python may have been addressed in later releases, especially v3.***

If I were to make a sincere effort, I would concentrate on producing a faithful replica of the stable branch (with refactoring where appropriate to bridge any gap between the two languages/architectures/libraries such as template engines). In other words, I would not make any attempt to fix design issues or add new features in order to preserve as much value of the existing knowledge base as possible. Obviously competent help would be greatly appreciated, as this would not be my day job.

However, I am reluctant to embark on such an undertaking without having some assurances that the effort would be useful, or that it would be a realistic candidate for becoming the adopted standard (assuming I do my part to competently comport with existing bugzilla philosophies, standards, and conventions wherever possible).

I have two questions:

1) If I was magically done today (with some approval of validity by existing developers), what is the likelihood that a Python implementation would (sometime over the following 6-12 months) be adopted as the primary architecture? I.E., how much inertia can realistically be attributed to Perl as an architecture (including comfort, emotional attachment, institutional knowledge, etc.)?

2) Assuming a 60%-or-better chance of adoption, does any onlooker have the desire and ability to assist (specifically with a Python "port")?****

Thanks, and I look forward to any discussion.


    --Matt


*** This e-mail is not directly intended to trigger a debate of tradeoffs, but such a debate may be necessary before my questions may be meaningfully answered. I'm guessing the biggest hurdle Python faces is a lack of a built-in "taint" mode which would necessitate the adoption of different coding standards/conventions. (There is an academic project which claims to provide similar functionality, but I've never used it, and I don't know anything about its performance.)

I know that the subject has been briefly broached several times on this list, but not substantively as far as I can tell.

In the interest of full disclosure, I have more experience with Python than Perl. I like Python. I do not like Perl. For the sake of this discussion, please consider this a disclosure of my emotional biases and attribute no weight to them.

**** Before answering, you should be aware that this would be a substantial undertaking. I say this only because I am trying to realistically gauge sincere interest.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.bugzilla.org/pipermail/developers/attachments/20130424/3c2d2a4a/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 4237 bytes
Desc: not available
URL: <http://lists.bugzilla.org/pipermail/developers/attachments/20130424/3c2d2a4a/attachment.p7s>