While introducing Calendar2 and the XCalendar3 interface anyway, as well the
"one letter" narrow month names sometimes needed in calendaring can be added.
These are not always the first capitalized letters of the months in all
locales, and might even not necessarily be just one character.
Introduces a new struct css::i18n::Calendar2 that is not derived from
css::i18n::Calendar because it uses a new struct css::i18n::CalendarItem2 to
hold the NarrowName elements.
In locale data the elements DaysOfMonth, MonthsOfYear and GenitiveMonths now
have an optional element DefaultNarrowName. If not specified, the first
character of DefaultFullName is taken.
LocaleDataWrapper and CalendarWrapper use and return the new Calendar2 and
sequences of CalendarItem2 structs. Application code adapted.
The number formatter now displays narrow month names (genitive if applicable)
for the MMMMM code, previously it displayed the first 16 bit code point of the
full name (not even utf-16 aware).
Narrow day names currently are not used, except in svtools' Calendar control.
* Switching between noun and genitive case month names is very simply done:
* If the format contains a day of month (D or DD) code, genitive case is used.
* If there is no day of month, the month noun is used. For example for
a standalone MMMM code or a combined MMMM-YYYY code.
* For input both, noun and genitive case, forms are recognized.
Passing the implementation instead of an uno reference may result in
lifetime problems if the passed interface is distroyed while the second
one is still valid
see rhbz#579817 with memcheck
Add possibility to start strace tracing the process. Unfortunately it
this doesn't seem to work as nicely as one might have hoped. If the
process crashes, the last strace output does not show up. Some
buffering that gets abruptly discarded when the straced process dies?
Add a function to the native code to look up an archive member in the
.apk and return a pointer to it. To be used for non-compressed
resources only. We mmap the whole .apk. The Zip format parsing code is
borrrowed from Mozilla's APKOpen.cpp.
Correspondingly, add to the local build.xml a re-definition of the
"-package-resources" target from Ant's build.xml modified to not
compress resources.
Improved the Makefile a lot. New target "install" to build the apk and
install it without constructing the apk twice.
Other minor changes here and there.
When filtering by equality, we need to have at least one value to
compare to, or else bad things may happen.
The pivot table OTOH can handle empty set. So over there, it's okay
to allow no items to be selected.