Pauli, The source code was attached and is linked in the first post of this thread.
Changes are as follows: * NumpyExt was made into a package * autosummary_genarate.py was turned into a console script with the name sphinx-autogen * module .rst generation was moved over to a jinja template, and functions, classes, and exceptions within that module are now auto- documented. I had a hard time figuring out how to use autosummary, and since I want to do a presentation at pycon about it, I'd like to have it be really easy to use. I'd also rather see it being included in sphinx, as apposed to a separate package. I'd also like to see sphinx- quickstart have an automatic option for creating an index with all of the modules in a package linked. Right now my demo will go something like: easy_install sphinx-autogen (pulls down sphinx, autogen, jinja, etc.) mkdir docs cd docs sphinx-quickstart edit conf.py to include numpyext.autogen mkdir html sphinx-build source html *edit* index.rst add in barbershop.model sphinx-autogen source/*.rst generated sphinx-build source html I'd love to have a demo that looks like this: sphinx-quickstart -autogen barbershop sphinx-build source html (lets get rid of having to create the directory if it doesn't exist) Sphinx is so very powerful, but it requires a serious amount of boilerplate to get going. I think autogen on the modules of a package would be a great start for the majority of projects. I realize sphinx does not want to be like epydocs, but providing a good starting point never hurts. cheers. -chris On Oct 24, 10:42 am, Pauli Virtanen <[EMAIL PROTECTED]> wrote: > Hi, > > Fri, 24 Oct 2008 08:47:17 -0700, percious wrote: > > > To re-iterate, I did not want to fork the entire of sphinx, rather, just > > the autosummary stuff that the folks from Numpy have created. Let me > > know if/how you want me to integrate with the sphinx codebase, and I > > will be happy to do so. > > I think it's best to avoid forking the Spinhx extensions -- they're so > easy and fun to write that I expect the 20/80 problem will be especially > bad for them. Moreover, having two extensions called "numpyext", one of > which is used in Numpy and one that's not will be a bit confusing :) > > So, (as the original author of autosummary), I'd be interested in hearing > what kind of enhancements you did to it. Is your code somewhere publicly > available? I'm definitely interested in making numpyext better and more > re-usable. Also, the Numpy manuals are still work-in-progress, so any > improvements may also benefit them. > > Cheers, > > -- > Pauli Virtanen --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "sphinx-dev" group. To post to this group, send email to sphinx-dev@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sphinx-dev?hl=en -~----------~----~----~----~------~----~------~--~---
