auctex-devel
[Top][All Lists]
Advanced

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

[AUCTeX-devel] Re: [AUCTeX] Inverse Search with Evince & DBUS


From: Tassilo Horn
Subject: [AUCTeX-devel] Re: [AUCTeX] Inverse Search with Evince & DBUS
Date: Sun, 06 Feb 2011 17:41:37 +0100
User-agent: Gnus/5.110011 (No Gnus v0.11) Emacs/24.0.50 (gnu/linux)

Ralf Angeli <address@hidden> writes:

Hi Ralf,

>> Sure, if you give me a pointer to where it should be placed, I can
>> provide a patch.
>
> It's not really obvious to me where this could be integrated, at least
> with respect to the call to `dbus-register-signal' and its parameters.

Yes, I've thought that.

> Ideally the function should only be called if a Dbus-aware viewer is
> selected and be fed with the parameters which are specific to that
> viewer.

I guess, all the options are specific to Evince.

> That would mean that these details could be stored in
> `TeX-view-program-list{-builtin}'.  Indeed, the original intention of
> these variables was to have a sort of database of viewers including
> all their capabilities and invocation options.  But I'm not sure how
> good the Dbus stuff would fit in there.  (And we'd probably have to
> reserve a place for forward search via Dbus, see the discussion at
> <URL:http://thread.gmane.org/gmane.emacs.auctex.devel/2490>)

AFAIKS, in that alist only the options for forward search are added.

> An option which would require less coding effort would be to hardcode
> the call and put it into `TeX-source-correlate-mode'.  If the name of
> the Dbus signal is not changed in the (near) future and there are not
> many other viewers for which this would have to be done, then it might
> be a pragmatic solution.

Right now, I only know of Evince implementing that stuff via DBUS.

> Do you know if listening to Dbus signals is resource intensive or has
> any security implications?  If that's the case we might at least have
> to provide a way to switch it off.

I don't know.  At least I cannot notice any performance issues on my
system.  However, I don't know if it can get resource intensive to
dispatch between many messages and signals to pick the one(s) emacs has
registered for.  You might want to use `dbus-monitor' to see what goes
over the bus.

WRT security, what goes over the session bus is only the basename of the
TeX document and a line number (and a column number that is always -1).

I Cc-ed Michael Albinus, the DBUS expert.  I guess he can provide more
insights.

Bye,
Tassilo



reply via email to

[Prev in Thread] Current Thread [Next in Thread]