Incorrect activation of mark upon isearch exit

Incorrect activation of mark upon isearch exit

[Nordlöw]

Hey!

I'm trying to find the reason for a non-critical but still disturbing
bug in my emacs setup.
Once in while after I do isearch suddenly the mark always becomes
active (mark-active) after I finish my isearch.
And if I do isearch again the behaviour remains until I restart Emacs.
So I guess that question now becomes; How do I trace the change of
"mark-active" from nil to t? Is this possible?
I couldn't find anything in edebug that does this right? Is there some
other function that is called when mark is activated?

As always with these tricky bug it never occurrs in vanilla emacs.
I'm quite sure this state-change is triggered by some incorrect? key-
combo during isearch but I'm not sure how.

Thanks in advance,
Per Nordlöw

RE: Incorrect activation of mark upon isearch exit

[Drew Adams]

> I'm trying to find the reason for a non-critical but still disturbing
> bug in my emacs setup.
> Once in while after I do isearch suddenly the mark always becomes
> active (mark-active) after I finish my isearch.
> And if I do isearch again the behaviour remains until I restart Emacs.
> So I guess that question now becomes; How do I trace the change of
> "mark-active" from nil to t? Is this possible?
> I couldn't find anything in edebug that does this right? Is there some
> other function that is called when mark is activated?
> 
> As always with these tricky bug it never occurrs in vanilla emacs.
> I'm quite sure this state-change is triggered by some incorrect? key-
> combo during isearch but I'm not sure how.

Hi Per,

Are you using Isearch+?  If so, check the value of option
`isearchp-set-region-flag'.  During isearch, `C-SPC' toggles the value.  If for
some reason you saved the value change then that could explain why the region
was set initially in subsequent Emacs sessions.