* Changed Calc's internal definition of points per inch from TeX points (72.27
points per inch) to PostScript points (72 points per inch) as used by the
rest of the world (except TeX).
* Aligns between internal representation and API that at various places
assumed PS points already.
* Fixes mis-positioning and bad sizing of drawing layer objects relative to
column widths and row heights.
* More precise printout of column widths and row heights.
* UI dialogs such as Column Width always assumed PS points, so earlier
versions actually displayed wrong metric (and imperial) units.
* Printouts now are slightly larger due to 0.27 points less in an inch, but
more accurate in reflecting the actual UI values, and drawing layer objects'
position and size match the row heights and column widths.
- Problems with drawing layer objects and zoomed view versus Page Preview
versus printout that don't align still persist.
- Problems with resized rows and misplaced drawing layer objects are also not
magically healed.
Wraps this
/usr/include/neon/ne_locks.h:125:51: warning: 'int ne_lock(ne_session*, ne_lock*)'
hides constructor for 'struct ne_lock' [-Wshadow]
into a system header guard.
xsec_fw is always built, so always register and install
xsec_xmlsec is built when ENABLE_XMLSEC is true, match
register and install
plugin is built when WITH_MOZILLA is not NO, match
register
Introduce gb_Helper_LIBRARY_PATH_VAR that is the name of the
environment variable used to look up shared libraries, which differs
depending on OS_FOR_BUILD.
Set that environment variable instead of a hardcoded LD_LIBRARY_PATH
in gb_Helper_set_ld_path.
Make all the PRECOMMAND macros actually use gb_Helper_set_ld_path
instead of duplicating it.
This fixes cross-compilation to HOST platforms that use unxgcc.mk (for
instance Android) from BUILD platforms that don't (for instance
MacOSX).
Slight tweak of d0d62edf3f - getPixel()
and getPixelData() are complementary functions, similar in spirit
to const and non-const getters. Added unit test for it to avoid
flagging it for removal again.
I removed 2 unused headers.
I also stopped delivering a lot of headers that no one outside of basebmp
cared about.
I also removed the unused methods:
basebmp::BitmapDevice::getPaletteEntryCount() const
basebmp::BitmapDevice::getPixelData(basegfx::B2IPoint const&)
We get some assertions in the smoketest with --enable-debug, looking
closer I think to use a single bIsGrey flag we...
a) would need to check after every modification of a Palette that mbIsGrey is
still valid and update it accordingly
b) There are a lot of routes outside of direct assignation to the stock Grey
Palettes that can result in Grey palettes in e.g. the pngreader, they would
all need updating :-(
So, how about just refactoring the original HasGreyPalette and IsGreyPalette.
still needs work in:
- RdbTarget_sc_filters_test -- missing component files to NSS (need to be
conditional on nss)
- nssinitializer.cxx:306 -- new environment variable BRAND_PROGRAM_DIR for
$(BRAND_BASE_DIR)/program that is overriden in the unit test to
solver/(platform)/lib
- make the testPassword in filters-test.cxx conditional on NSS
- LD_LIBRARY_PATH needs an additional entry for
solver/(platform)/lib/sqlite
password.ods has password test and has been created Nov, 3rd 2011 with a
master build and should be ODF1.2 conform
