Genshi question

[anton]

Hi,

last week I had a bug which prevented me to
add an attachment on a trac wiki page.

I found out that Genshi was the reason.

The work around was to switch the prefferred language
to english insted of using the default
language (my default language is german).

Sorry ... I do not find the trac ticket for this bug actually :-(

But my question is:

Since Genshi does not seem to be developped actively anymore
( on http://genshi.edgewall.org/browser the last change is 14 months ago )
what is the future of Genshi ... and Trac of course.

Are there plans to support Jinja or some other template
engine?

I am still on python 2.7 but I will try to move to python 3.5 in the
next time, so it would be interesting if python 3.5 is supported too.

Thanks for a hint.

Anton

[Torge Riedel]

Hi,

I faced this bug too on december when upgrading trac at my company to v1.0.9 and all related stuff. The trick is to avoid downloading and installing a package of Genshi from the download page, but install from repository url http://svn.edgewall.org/repos/genshi/branches/stable/0.7.x/ which contains some additional bug fixes. Including the one you ran in. I've found this by finding the related bug ticket and commit in the svn to see, that it has been merged to this stable branch.

Too me (as project leader / software developer in business) this is confusing and not very helpful too, cause IMHO if I fix a bug causing troubles, I will release a bug fix version as soon as possible and provide a download for it. But this is as everything in the world: Just one opinion of how to work. There are other approaches we have to live with (or not).

At least with trac this went better (IMHO) the last year(s) when several bug fix versions were released. This is still ongoing and I hope this is kept in the future. Since it makes it easier to tell which version you have installed.

If a project (like Genshi) does not have much (or no) commits in the past months, this does not mean it is not actively developed. It might be the case, that this project has no enhancement / bugs waiting to be fixed. But I can't really tell for Genshi whether this is the case or not. However - it (trac+Genshi) is for free and has a big community and helpful guys (like RjOllos) and that is what counts to me and makes it worth to be used.

No knowledge on my side regarding Jinja / python 3.5

Regards
Torge

[anton]

Hi,

thanks for the info.

Hmmm just looked at

http://genshi.edgewall.org/query

and found 108 bugs open so ... it seems that there
are bugs to be fixed.

Ah, and according to

http://genshi.edgewall.org/ticket/602

python 3.5 seems to be broken (last ticket change 4 Months ago).

I will not switch to 3.5 tomorrow :-), but when I switch
to python 3.x I will switch directly to 3.5,
so I hope Trac will work with python 3.5 at that moment.

Otherwise Trac is a really great project which I like and use :-)

Cheers

Anton

[anton]

Hi,

I have to correct myself:

genshi has 108 issues open but
"only" 71 bugs.

Anton

[RjOllos]

On Sunday, January 17, 2016 at 3:50:11 AM UTC-8, anton wrote:

Hi,

I have to correct myself:

genshi has 108 issues open but
"only" 71 bugs.

Anton

There is some discussion about a Genshi release over on their mailing list:

https://groups.google.com/d/msg/genshi/Ajx0DE2urHM/44U3QvGPCAAJ

There was some activity involving moving the project to GitHub and setting up CI. I found that to be a bit unfortunate because the priority really needs to be on cutting a release.

Moving to Jinja would require a huge amount of work. It has been discussed in the past.

http://trac.edgewall.org/wiki/TracDev/Proposals/Jinja

If there is no more development on Genshi, or development is so sporadic that we can't rely on the project, then Trac will eventually need to come up with a plan for moving to another templating engine. I suppose there are alternatives, such as Trac developers contributing to Genshi, but already everyone is quite busy. I also imagine that enough other projects depend on Genshi that someone will eventually fork it, or offer to take over development, like what happened with Babel. Babel moved from Edgewall to the umbrella of the Flask project a few years ago, and seems to have maintainers and a good release cadence now.

On a related topic, I'd really like to see mature Python projects managed by a foundation that could hand off development to new maintainers when a project stalls. I had been thinking about this for a while, and then this post really resonated with my thoughts:

http://www.drmaciver.com/2015/08/what-if-we-had-more-finished-libraries/

In the Genshi mailing list thread there is a post from a member of the TurboGears project, in which they mention supporting Kajiki as well as Genshi. Kajiki appears to be a fork of Genshi, and it might be possible to migrate to that templating engine. The situation with Genshi and Kajiki was also discussed on a recent Talk Python To Me podcast:

https://talkpython.fm/episodes/show/35/turbogears-and-the-future-of-python-web-frameworks

I don't have any experience with Kajiki, but I'd be interested to hear from anyone who does have experience.

Trac should have support for Python 3.3+ starting with release 1.4. Release 1.2 should come out in Q1 2016, and we'll start development towards 1.4 after that. I imagine we might have support for Python 3.3+ on the developer-stable release (1.3.x) by the end of 2016.

https://groups.google.com/d/msg/trac-dev/Rxqr-4Y0fTU/iCrNqOFj1DgJ

Thank you for the positive words about Trac and for contributing to the community :)

- Ryan