Support 2.6

[anandology]

mynt currently works only for Python 2.7.x as it uses some latest additions like:

• OrderedDict
• flags keyword argument to re.sub
• calling function with **d when d has unicode keys

These must be fixed to be able to use mynt on older versions of Python. Python2.6 is the default version available on Debian stable, I think it'll be useful to support it.

[anandology]

from __future__ import unicode_literals doesn't work in Python 2.5. So it looks like it is a bit harder to support 2.5.

[sleepdeprecation]

Is it important to support 2.5/2.6? Why not just update your version of python?

[anandology]

Because the version of Python available on latest Debian stable is Python 2.6.

I (and many people) would like to stay with the default version available with the distribution.

[uhnomoli]

Right now with mynt still in an early stage of development with core functionality missing, I'm not really looking to support other versions of Python. For one, I don't have anything besides 2.7 to test on. I'm not really looking to take on the additional overhead that testing for other versions would require while still working on core functionality. Python 2.6/3.x support is definitely on the todo list, I'm just looking to get the core of mynt completed before I tackle that issue.

I will leave this issue and the related pull request open however, so people looking to use mynt on 2.6 will have an easier time doing so.

[anandology]

I can volunteer to test Python2.6 and maintain compatibility.

I'm trying to use mynt for Python India community website and other user groups in India. I'll be very useful if you can support Python2.6 as that is the only version of Python available on our server.

Do you have a dev plan for next version? I can take up some tasks. One feature I'll very keen to have is the ability to write regular pages in markdown, just like posts. Is that already in your task list?

[uhnomoli]

Right now, the main goal is making an official release out of the current dev version with maybe a few more fixes tossed in. The main thing holding this back is finishing the rewrite of mynt's documentation which is about half complete.

Next is pagination along with some cleanup and refactoring.

Last, which in my eyes would complete the core functionality of mynt, is the content system. This will allow 'Markdown pages'. Basically you will have the ability to define content types and configure them independently of each other (parser, renderer, URLs, etc.) Posts will just be a predefined content type. For example, for a portfolio site, you may have a projects content type.

As far as your offer to maintain Python 2.6 compatibility goes, thanks, but as I said, I'm not interested in officially supporting other Python versions right now. It is just too early in development to be worrying about compatibility. If you want to maintain your fork to be compatible with 2.6, I'd be glad to send anyone looking your way as well as leave this issue open so others can find it.

[anandology]

The only reason I'm asking to merge it back is to keep the simplicity of installing mynt using pip install. I can always fork and maintain my own pypi package, but that is unnecessary effort as I've already provided you a pull request. And you don't have to officially support older versions, even if you merge this pull request.

Are the docs available in any public repo?

[uhnomoli]

I'd actually request that you don't create your own PyPI package, though if you wanted to you could as long as you renamed the project and removed any links to mynt's site.

Git URLs are usable with pip. That's how I recommend installing the dev version. All that would be required for someone to use your fork is the following:

$ pip install git+https://github.com/anandology/mynt.git

If I merge the pull request, people are going to expect it to work with Python 2.6 and if they encounter issues on 2.6 they will open issues expecting support which I'm not in a position to give nor do I wish to right now.

The documentation is not available publicly anywhere, though I plan to do so after I finish rewriting them as I was contacted by someone interested in maintaining a Japanese translation.

[anandology]

Fine. I'll go with your suggestion for now. But I hope you'll consider Python2.6 support in future releases. As I've already mentioned earlier I'm ready to volunteer for maintaining Python2.6 support.

I can help you with writing/rewriting the documentation if you can make the documentation publicly available.

[uhnomoli]

As I said, Python 2.6/3.x support is on the todo list, I just don't think it's worth it to tackle such an issue until core functionality has been completed.

[eichin]

FWIW Debian "stable" is now wheezy, with python 2.7.3 and 3.2.3, which somewhat lowers the value of caring about older versions of python.

[Mindiell]

Hi @Anomareh, I converted mynt to python 3.8 in order to help. I don't know if you're still on it (last commit from 2014).
I did not creat any pypi package for now as I would prefer to PR into the orginial mynt project, but I'm not sure you're still there.
I'm on IRC too, on the #mynt channel if needed.

My repo is simply https://github.com/Mindiell/mynt3

thx,

[uhnomoli]

@Mindiell sorry, I was quickly going through issues and didn't see your post. I finally got some time to work on mynt and updated it to Python 3. Didn't mean to ignore your work. I just took a look at your repo and couldn't really tell what you had changed as everything was added in one commit, but if you made any other changes you'd like to see let me know and I can take a look!