I don't wanna sound selfish nor elude the fact that I'm grateful to Doug about his work but, speaking of work, are you still working on it ?
I'm using it but I now quite often stumble on limitations or bugs. The "-d" bug for instance (see bugtracker) looks pretty simple, but hasn't received any anwser since many monthes.
What should I think ?
I'd like to see more support and progression on this. There's many tools for Perl and such, and while happydoc is the most elaborated for python, it still has a lot of margin for progression.
I am still here, but have not had a lot of time lately to work on HappyDoc. I changed jobs in the fall, and that has changed the amount of time I have to devote to it.
I thought I had fixed the -d bug, with a special release that I asked to have tested. I'll try to go back and review the defect report and the fix this weekend.
As far as other areas for improvement, I would appreciate any feedback, and of course patches, you might like to provide.
I know everybody has a right to go on holidays ;)
But would be nice if you could quickly tell us about a schedule for a new version. Possibly with an idea of which bug/feat.request could be included.
I'm now initiating "wide" (less than 10 persons) use of StructuredText in our code and delivery in our package of the happydoc generated documentation. I'd like to know if I can allow accents for instance (badly converted to html, bug tracked) or allow images (the <img src... does not work for the same reason).
Thanks for your help
The solution to the accented characters problem appears to be to use setlocale() before importing the string module. I have updated the CVS version of the main program happydoc with this change. The 2.1 release does not include the fix, but I hope to fix a couple of other bugs over the next week and cut another release soon.
The most pressing bug right now is the -d problem. I've got what seems to be a fix, but I'm not 100% satisfied with the output (README files do not end up in the same directory as files generated from code). As soon as I have that resolved, I will be making a new release that includes it. I will then begin working on other bugs/features such as the ones you have submitted.
I'm not sure what is causing the accent problem. You might want to try grabbing the StructuredText code from the Zope CVS repository. I redistribute that version, but I have not kept it up to date lately. I'm not sure if upgrading will fix anything, but you could test it pretty easily by replacing the files from your HappyDoc installation with those from the Zope CVS tree.
I'm not sure what you mean by "<img src... does not work for the same reason". Are you trying to embed images in your docstrings, or in external files?
Remember that I've tested sucessfully (or seems to be) the dos-bug release that fixed the bug. Probably you're only 90% satisfied with this fix ;)
I'm gonna look into Zope.
The <img src : I've added a comment to 504810 regarding embedded HTML. Two examples :
<i>Foo</i> in docstring ends up as <i>Foo</i> in the HTML (doesn't display as I expected.
The img src with double quotes are worse because <img src="foo.gif"> ends up like this : <img src="foo.gif">
Is this really the bug described in 504810 ? I'm not sure myself.
Honestly, I don't feel confident enough with Happydoc source tree, nor with Zope's one to test the latest code against the 504810 and the "accents" stuff.
But reading latest description of ST "classical" http://dev.zope.org/Members/jim/StructuredTextWiki/ClassicStructuredTextRules
I think it would already be a good idea to take it into account in next release so that we can use new features (I like the document references  This doc), get all the bugfixes, and hopefully solve a couple of Happydoc bugs that are ST related.
I hope this is as easy as you told me to do.
I suck :-(
I was using since the beginning the StructuredText desciprtion from Python.org (sigs/doc-sig) which does'nt mention latest additions to ST definition.
While reading at the code, I realisezed that the "new" items I was talking about are already available in Happydoc.
Shame on me.
However, I've looked for a workaround to the "accents" stuff and I've found one. See bug tracker for the description. Would be nice anyway to check if Zope's implementation could take this into account as well (maybe they already did ?)
Second time today : shame on me.
The <img src=" docstring being converted to <img src" is the regular action of the htmlQuoteText=true parameter. Setting it to false passes through the HTML tags without touching them.
Forget about this.
Well, I have to appologize for 2 things. First, for forgetting that the htmlQuoteText option was in there, and second for making it difficult for you to discover on your own.
I guess after the -d bug is rolled into a release, I'll work on documentation next. :-)
Log in to post a comment.