Thank you all for your input. To address some of the points and
> How complicated would it be to make shift+middleclick == plaintext
Not at all complicated. I think this would be a great compromise that
would be uninvasive and pretty much hidden from anybody who doesn't care
about this functionality.
Andrew Sayman wrote:
> Can this kind of thing be done as a plugin?
I tried to figure out a way to do this initially and failed. From what
I can tell, the plugin API itself lacks the hooks to do this, so it
couldn't be done in perl, for example. (Which would be ideal as far as
a plugin goes because it wouldn't require any recompiling between
versions of gaim.) I've had some brief flashes of insight, though, and
I intend to try to proof the idea of a plugin that grant this control.
I think I can have a plugin that would disconnect the callback from the
button click signal in all the conversation window entry widgets and
connect my own replacement of that function (and then undo that on
unload of the plugin.) It ain't pretty, and it would likely require
ongoing maintenance after each release of gaim, but it's interesting
enough that I intend to try it. Frankly, it'd be easier to embed the
behavior in gaim proper and only expose checkbox to control in a plugin.
Sean Egan wrote:
> This is a textbook case where we should use a sane default rather than
> introduce a mostly useless preference.
Sane defaults are good. So are ways to override them (otherwise they
wouldn't be "defaults".) I'm inclined to submit a patch making
shift-middleclick paste unformatted and leave it at that.
(Notwithstanding my intention to play with the plugin idea above for my