On Monday 15 August 2005 15:36, Chris Curvey wrote:
> 
>  What I need to do is control a Mozilla browser via a Python script. In
>  other words, I want my Python script to start Mozilla, tell it to navigate
>  to a certain page, click on a link, fill out a form, click on a button,
>  etc. I want Mozilla to continue to handle JavaScript, cookies, and the rest
>  of it. Maybe a better way to describe it is that I want to replace my mouse
>  and keyboard with a Python script.

I've been wanting to do similar things.

>  PyXPCOM looks like what I need to do this, but after reading the docs, I'm
>  still unclear (and this seems like a pretty deep pool to jump into
>  unawares). So my questions are:
> 
>  Is PyXPCOM what I'm looking for?

Yes, I think so, but with various caveats. I believe that to control Mozilla 
and to examine documents, you need to have your PyXPCOM code running inside 
the Mozilla process; this means that it has to be installed as a component 
and "bootstrapped" from JavaScript, although I'd love to hear that this isn't 
the case in one way or another.

>  Do I need to build my own version of Mozilla or Firefox in order to make
>  this happen?

At this stage, I think you do need to build Mozilla or Firefox with the 
PyXPCOM extension enabled. Various helpful people have recently made this 
process a lot easier than it was just a few weeks ago.

>  It looks like I need the Mozilla or Firefox source in any case, just to
>  compile PyXPCOM. Is this correct?

Yes, I think so. I did hear stuff about a standalone PyXPCOM but I don't think 
this is relevant to what you want to do.

>  Can anyone point me to any doco for the interfaces that Mozilla/Firefox
>  provides thru XPCOM?

I've found this resource quite useful:

http://xulplanet.com/references/xpcomref/

>  I'll stop asking questions now :) Thanks in advance, and if PyXPCOM is the
>  right thing for me, then I look forward to working with you all.

As I said, I've been wanting to look at browser automation, too. 
Unfortunately, I haven't managed to be able to get any sensible document 
content from loaded pages, or at least not HTML pages (in contrast to 
XML/XHTML pages). Whether my issues are related to documentation or a lack of 
understanding is not yet clear, and I haven't been inclined to do much more 
with this in the past few weeks.

Paul

P.S. I may have mentioned this already, but some interesting browser 
automation is also possible with Konqueror/KHTML given enough additional 
components and packages:

http://www.boddie.org.uk/david/Projects/Python/KDE/index.html

(See the kpartplugins package, probably requiring fairly recent PyQt and PyKDE 
releases.)
_______________________________________________
pyxpcom mailing list
pyxpcom@[...].com
To unsubscribe: http://listserv.ActiveState.com/mailman/mysubs