Graham Percival <[email protected]> writes:
> On Thu, Dec 03, 2009 at 10:00:13AM +0100, David Kastrup wrote:
>> the current version of the patch set
>> <URL:http://codereview.appspot.com/160048> has been sitting there on
>> Rietveld for about a week.
>
> Marek: I guess it's time to add this as a PATCH item to the google
> tracker.
>
>> For the current version, checked in a week ago, I did not get
>> any feedback from testers or reviews or comments from the people
>> who know more about the background of the original code design.
>
> Are you aware that the world does not revolve around you? For the
> past few days, the git tree has been broken -- I cannot compile
> the regression tests or documentation. Not just "the regtests
> look like garbage", but "there is a segfault when attempting to
> compile a regtest and the docs".
>
> IMNSHO, fixing that problem takes priority over new code.
Ok, here is your fix. Can you now check the memory leak?
diff --git a/lily/script-column.cc b/lily/script-column.cc
index 23e1b69..e001529 100644
--- a/lily/script-column.cc
+++ b/lily/script-column.cc
@@ -140,8 +140,7 @@ Script_column::order_grobs (vector<Grob*> grobs)
ss = scm_stable_sort_x (ss, ly_grob_script_priority_less_proc);
Grob *last = 0;
- Grob *first = unsmob_grob (scm_car (ss));
- SCM default_outside_staff = first->get_property("outside-staff-priority");
+ SCM default_outside_staff = SCM_EOL;
for (SCM s = ss; scm_is_pair (s); s = scm_cdr (s))
{
Grob *g = unsmob_grob (scm_car (s));
@@ -176,6 +175,8 @@ Script_column::order_grobs (vector<Grob*> grobs)
else
Side_position_interface::add_support (g, last);
}
+ else
+ default_outside_staff = g->get_property("outside-staff-priority");
last = g;
}
}
--
David Kastrup
_______________________________________________
lilypond-devel mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/lilypond-devel
