Add a documentation generator that creates a PDF or HTML files locally

This would be nice to have. A very low-priority enhancement.

Maybe the first needed thing is to have http://www.djangoproject.com/documentation/ as a .rst file in SVN (index.rst ?) it's currently a hardcoded template (djangoproject.com/django_website/templates/docs/index.html). Could the index page be moved to an .rst?

#1527 has been marked as a duplicate of this ticket. That one provides a Makefile that calls rst2html.py for every file in docs/ in case you want it ;)

I've started working on this in my spare time.

FYI: I've just done a similar thing for my own projects (some of this information might be useful to you). I'm using MoinMoin? as my wiki and it has a plugin for PDF creation that uses htmldoc as the pdf generator. It provides a really nice interface and works like a dream.

I wrote a small python script in my build system that simply got the web pages with the 'CreatePdf?' action (ie after calling the wiki macro to create a PDF) and wrote it to a file which I zip up into my release.

The macro (and info) is here: http://moinmo.in/ActionMarket/PdfAction

Hope this helps :) - Tushar.

Claiming #528 and #4940 for Sprint14Sep.

Patch: a script to render HTML docs

The script I've created does not produce especially beautiful output, but the basic basic functionality is there to render the Django doc dir into one large HTML doc or many small HTML docs with an index.

Marked #4940 as a duplicate.

+1

if you want to improve the documenation, you need some way to check the wiki syntax before you create a patch.

the patch render_docs.py.diff works for me. It think it is ready for check in (it can't break anything).

This isn't ready to check in.

• There should be practically no executable code at the top level. In other words, importing the file should not execute it. At least a main() function and if __name__ == '__main__': guard are required.
• The imports should conform to PEP 8 style guidelines.
• If you don't have docutils installed (which will be noticable in the import section), a helpful error message would be good, since docutils don't come by default with Python.

If somebody wants to take this further, it's probably worth checking in with Adrian (see comment 7) to find out where he's up.

I wrote a small bit of documentation and did the requested things for improving the patch.

Apparently #6443 was created to suggest pisa from http://www.htmltopdf.org/ for conversion from HTML to PDF. I've closed it as duplicate of this and added the relevant info here.

Replying to adrian:

Replying to mtredinnick:

This isn't ready to check in. * There should be practically no executable code at the top level. In other words, importing the file should not execute it. At least a main() function and if __name__ == '__main__': guard are required. * The imports should conform to PEP 8 style guidelines. * If you don't have docutils installed (which will be noticable in the import section), a helpful error message would be good, since docutils don't come by default with Python. If somebody wants to take this further, it's probably worth checking in with Adrian (see comment 7) to find out where he's up.

This ain't a toy, sparky.

I've created a somewhat more extended version of this that produces output that looks very similar to real Django documentation. See DjangoSpecifications/Docs/ConvertingRestToOtherFormats for overview, screenshots and patch.

Not ready-for-checkin, since there doesn't appear to be any single patch to checkin, it's a large change if it's the thing you've posted to that above URL, so needs careful review by somebody other than yourself, plus, as Jacob mentioned, he's also working on a toolchain, so he'll possibly want to take bits and pieces of both approaches.

Posting the final patch attached to the wiki page here as well to avoid confusion (re: there doesn't appear to be any single patch to checkin).

Patch that implements a management command for adding docs, with documentation.