Man, I wish I could script Mozilla extensions in Python. OK, actually I can (or with this). I wish it was easy; easy to set up as a developer, easy to distribute the results. Not as an applet or anything, but either (a) as a stand-alone application which used Mozilla as its runtime, or (b) as an extension (though the UI experience doesn't seem as good, because what seems like a independent application turns into a menu item in a browser). Actually, I really want both, as they both have use cases, especially when you consider Thunderbird and Nvu.
I really really really wish I could do this. Every so often I forget about how much I want this, then I am reminded of it for some reason.
I know, so contribute, but let's be realistic. But man, I really wish someone would make it happen. I think they already want this anyway.
Or is it already possible? Or at least close? If so, correct me!
The PyXPCOM documentation dates to 2001. Maybe it worky-worky, maybe not.# Chris Smith
I make the same dream often, I think xft+pyxpcom can leverage scripting in python. A pity I have no time getting involved in this !# anonymous
This is not exactly what you want, and it is rather old, but at least you don't need to start from scratch. http://www.datalab.uci.edu/people/xge/python-glade/pymoz/ http://www.datalab.uci.edu/people/xge/python-glade/pymoz/README
Or may be this one? http://www.thomas-schilz.de/MozPython/README.html
+1. We need to build bridges between Python and Mozilla. PyXPCom in particular needs to be fixed. For one thing, the build process is excessively difficult, since the instructions suggest you have to successfully build the whole Mozilla beast in order to build PyXPCom. So just last night I decided to work on this. It turns out someone has already written a rough draft of a setup.py and patches for the unicode incompatibility problems (Mozilla always uses 2-byte characters, but my build of Python uses 4-byte characters.) With some hacking, the setup.py and patches worked.
I plan to finish modernizing the patches and write the build instructions within a few days. Then, if I decide to be ambitious, I'm thinking about making the Mozilla suite usable as a standard Python GUI, using a Wax-like interface. That would give me something to present at PyCon 2006, if someone else doesn't beat me to it. :-)
That would be sweet (and save me from having to code in javascript ever :-) Right now part of the reason I still use Safari is that I can manipulate it via it's OSA dictionary ("applescript" the interface, not the language) from python ... so I have things like SessionSaver that actually work, and such.