From: Dan B. <db...@gm...> - 2012-02-24 17:14:02
|
I was thinking that as long as you only set breakpoints from within emacs, you wouldn't need to query matlab, and that would make things simpler. That thread mentions a couple pages of code, which doesn't sound too bad too me ... but that doesn't mean I'm volunteering to write said code ;-) But I've never done any real emacs lisp programming, so maybe writing something to hook into GUD would be a good first project. I will email the list if I come up with anything. Thanks for the replies! Dan On Thu, Feb 23, 2012 at 5:11 AM, David Chappaz <dav...@fr...> wrote: > >>> -----Original Message----- >>> From: Dan Becker [mailto:db...@gm...] >>> >>> Even if you set breakpoints through dbstop >>> commands in the matlab shell, it should still be able to keep track. >>> >>> Is there something about this that makes it much harder than it seems? > > This was shortly discussed in this message: > http://sourceforge.net/mailarchive/message.php?msg_id=28513597 > > > For matlab-shell to keep track of the breakpoints is not impossible but a > little tricky, because it would have to constantly query matlab with > "dbstatus". > > Everything would be much easier if matlab was itself sending a message to > matlab-emacs whenever the list of breakpoints changes (e.g. all breakpoints > are cleared, or a new one is added). This feature used to be present with > EmacsLink, but has now disappeared. > > My personal take on this is that if enough people complain to MathWworks > directly (not Eric), then they'll (perhaps) do something about it. > As I can't remember where my breakpoints are... I just stick with R2008 and > EmacsLink for now... > > -- Dan Becker |