On Friday 21 October 2005 16:02, Phil Thompson wrote: > On Friday 21 October 2005 8:03 pm, Gustavo Sverzut Barbieri wrote: > > On 10/21/05, Gilbert Ramirez <[EMAIL PROTECTED]> wrote: > > > For me, one of the strong points of Qt is that the documentation for > > > it is great. Although the documention is for C++, it's still a handy > > > reference for PyQT programmers because the mapping from Qt/C++ to PyQt > > > is almost always straightforward. I'd like to keep the > > > non-straight-forward mappings in the documentation to a minimum, > > > because no one is going to do as good of a job as writing > > > documentation for an API that is very different from Qt's. > > > > What raises another point: PyQT should include docstrings! > > > > Phil, have you considered this? > > Yes - it's been on the TODO list for a while. > > Current PyQt documentation is effectively handwritten and (therefore) only > describes differences to Qt. I will add automatic generation of > documentation to SIP at some point which will describe what is actually > implemented using Python data types. It would then be easy enough to > generate a summary as docstrings - maybe as a configuration option. (And if > anybody has bright ideas about how to automagically generate links to the > corresponding bit of Qt documentation, I'd like to hear them.)
On the Qt3 docs I have, all of the pages are links like .../<classname>.html#<method name> basically - it looks a little trickier for ctors or overloaded methods, but at least it's easy to get close. I generate PyKDE docs (which aren't much more than method signatures) from the same code that parses h files -> sip files - it's probably easier to spit out HTML than sip syntax. KDE has made the docs more difficult to generate links for though - I think the intrapage links (#) are just numbered sequentially now rather than related to method names. Jim _______________________________________________ PyKDE mailing list PyKDE@mats.imk.fraunhofer.de http://mats.imk.fraunhofer.de/mailman/listinfo/pykde