Re: [orca-list] Getting started with Orca scripting



Hi Nolan:

I'm CC'ing Li Yuan, who is expert in all things related to AT-SPI.

How do I set up Orca for debugging and logging its output? I've changed
the debugging level, which gives lots of useful output, and I've seen
lines in user-settings.py for opening a file. Seems that setting the
debug level only sends the data to stdout, though, which isn't quite
what I want. I can certainly redirect the output from a terminal, but
since there appears to be a logging mechanism available, I should
probably use that. And it'd appear that I've uncommented one of the
output file creation lines. Is this all that is necessary to start
logging?

Putting this line in your ~/.orca/user-setting.py or ~/.orca/orca-customizations.py file should work:

orca.debug.debugFile = open('debug.out', 'w', 0)

It's preferable to modify your ~/.orca/orca-customizations.py file since it will not be overwritten when you adjust your preferences via the preferences GUI. With the line above, debug.out will be created in the directory from where you launched Orca. If you want an exact path, you could use '/tmp/debug.out' or something like it.

Now for the actual scripting questions. My first challenge is to
actually pronounce the source column name correctly. Debugging seems to
indicate that the column name is being sent as "S_ource" which isn't
particularly aesthentically pleasing. :)

Strange. The underscore character is usually used in labels to indicate the mnemonic to be used to give something focus. I'm not sure it should be exposed in the name string sent over the AT-SPI. But, maybe there's a bug here (i.e., possibly column headers don't really support mnemonics?).

What would be the best way to
change this?

Ah..yech. Table row and col headers are treated as part of the contextual information for a table cell. Right now, the logic for obtaining them is buried in default.py:locusOfFocusChanged, though this could be much cleaner. If you have ideas for how to refactor this, please let us know.

Also, here's something else that confuses me. I'm trying to hack the
rhythmbox source to display more useful information on podcast feeds,
but I'm getting some disparity between what Orca speaks and what's
displayed when I flat review the screen. Specifically, when I view my
list of podcast feeds, I first get a column containing the feed URL
titled "Feed," then get an identically-named column containing the
podcast title. When I navigate this area with flat review, I don't see
the first "Feed" column containing the URL.

I haven't looked at RhythmBox in a long long time, so I'm not sure what the UI looks like. But, it might be possible that you're running into a situation where a non-text object is being used to present the "Feed" info. If this is the case, the flat-review stuff might be missing it.

Will




[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]