It's actually on both PPC and Intel, and it's from forgetting to update
gramps.accel after upgrading Gtk past 2.24.10, which changed the mapping
of alt/option from Mod5 to Mod1.
This allows the bundle to be sanitized with
`find Gramps.app -name '*.py[oc]' -delete` if there are code-signing
errors when moving the bundle from one machine to another.
svn: r22375
Make Arial Unicode MS the default face for OSX
Supports East Asian, Semitic, and European scripts, but not Indic
(Hindi, Telugu, Tamil, Thai, etc.). Since Pango's ATSUI backend doesn't
support fallback, users of these scripts will have to override the
default face in their ~/.gtkrc-2.0
svn: r18790
Importing gramps at the end of rungramps causes threading deadlocks because of a lock (held by the main thread as long as Gramps is running) on import.
svn: r17706
Rewrote the launcher in python with much improved language handling;
renamed it to rungramps.py, and replaced launcher.sh with a minimal
shell launcher.
Any english selection will now have $LANG set to "C", while $LC_ALL will
be set to whatever english variant the user selected (and en_US if she
specified just "en"). For languages other than English, $LC_ALL will be
set to xx_XX if the language selection had no region (e.g., if the
language selection is de, $LANG will be de and $LC_ALL will be de_DE.
$LANG will be truncated if the language selection has a region that
Gramps doesn't support, but does support the bare language (e.g., if the
language selection is de_DE, $LANG will be de because there is no de_DE
translation. Gettext is smart enough to get this right without the help,
but the check is necessary to ensure that we don't skip over a supported
generic language.
The language settings can now be overridden for gramps by using the
defaults system.
svn: r17342
