Two things were wrong with the old code:

1) it didn't check for out-of-bound access, and
2) used SC_DP_MAX_FIELDS as the upper bound of dimension members, when
   in fact SC_DP_MAX_FIELDS was the upper bound of dimensions themselves,.
   not their members.  We shouldn't impose such upper bounds actually,
   at least not in the core implementation.  We should do that in the
   UI code if we really have to.

Change-Id: I307d4f34d2b0be84f0fd2b93a6270ffce2448bdf

Change-Id: I562ad42818fa7a6bfc765b7484a28ff1800b2949

It caused breakage of the existing slide-show full-screen mode. So
let's revert for now.

Pivot table output applies several pre-defined styles, which also include
number formats.  But, the source number format, or one set by the pivot
table should take precedence over the one specified by the style, or else
it would lead to a very peculiar behavior.

This is especially the case when the documnet has its 'Default' style
specify a number format that's not 'General'.

Also, we should take all number formats into account, including the ones
that are 0.

Change-Id: Ifa794314a2881c5146e24eb7be892585a823048d

Change-Id: I33a5dad31372b3add97ce31f6f9488a13f944f51

Change-Id: I6c031e9d2e9211409b28ce0c20a3cd11119ab86e

They should be the same for Mac OS X and iOS, so keep just one copy,
in com_GCC_class.mk. (Actually the iOS one was missing the
gb_VISIBILITY_FLAGS, which caused some linker warnings.)

Change-Id: I340fd7e262459bcc2b083c807b783d64d52040ee

Change-Id: I3190fba16638512dbc4247fa3fa95c39fa377b34

The Viewer app is intended to eventually resemble the experimental
Android DocumentLoader app.

Build using the gbuild mechanism, which is also invoked from an Xcode
project. This seems to work out fine, the resulting app installs at
least on the simulator, and you can debug all the LO code involved
even if Xcode (obviously) has no knowledge of the LO source
files/classes/etc.

Change-Id: Ic96178d80b8d6467cac969b29e37f0d39513acf9

The ones that are included in an app bundle, that is. Link them with
all our (static) libraries.

Change-Id: I36262dc32d09d28b43fcd702e9af1666c1daee3a

Change-Id: Ice30031754e00958e9b1906ed767f56d481889c3

Change-Id: I4c376f53af0a20203a4c8ea34abbf607ac7e987e

Change-Id: I46e4081375e2f7eda0c9f443cf28dcfe575a7dba

Change-Id: I2588b69280abbc510ee2fdc9845df56616db8b95

Change-Id: I1820711664a6aa18587e1e4a3d74a8e4cbfcc07e
Reviewed-on: https://gerrit.libreoffice.org/1537Reviewed-by: Luboš Luňák <l.lunak@suse.cz>
Tested-by: Luboš Luňák <l.lunak@suse.cz>

No need to two strings here, since just destFolder is used.

Change-Id: Ie1bfd5744113764caaec11d9fb9c771f9c67e1c4
Signed-off-by: Marcos Paulo de Souza <marcos.souza.org@gmail.com>
Signed-off-by: Luboš Luňák <l.lunak@suse.cz>

Change-Id: I179b5bfd1d85152a53bb821ab9eeb78efb8a6083
Signed-off-by: Marcos Paulo de Souza <marcos.souza.org@gmail.com>
Reviewed-on: https://gerrit.libreoffice.org/1516Reviewed-by: Luboš Luňák <l.lunak@suse.cz>
Tested-by: Luboš Luňák <l.lunak@suse.cz>

Change-Id: Icac8ec992d993748a063aa95cc6f58c24fa87444
Signed-off-by: Marcos Paulo de Souza <marcos.souza.org@gmail.com>
Reviewed-on: https://gerrit.libreoffice.org/1515Reviewed-by: Luboš Luňák <l.lunak@suse.cz>
Tested-by: Luboš Luňák <l.lunak@suse.cz>

It is already set by gb_CppunitTest_CPPTESTPRECOMMAND (or should be),
and doing it here again causes problems avoided
by 265feeb3 , or when ICECREAM_RUN is set.

Change-Id: I5ebe8ad735ea6ff8cb1c205c55c45b93550c8b55

This is the first part of the fix for gnome#627150

Change-Id: I02cd8c553dd6b9a026c0ea38d77c6831fef6498e

and restore some of the old optimizations

Change-Id: I1b42ed135bd3f8dfb663740a1bd2dbfa1216a388

Change-Id: I011a946311498cbe42ebcd5cd6517f855ea5e8d8

... caused by missing new-line after "\"

Change-Id: Ibdccf7f18e909308fbc1e58b6ddd3301e08a2223

The test just loads 4 documents and has been disabled for months.
Also remove the test documents because they are duplicates and are
also stored in writerfilter/qa/documents/.

Change-Id: I9f51240483d800bc104c3842b8158dec90b3e69b

Also fix the wrong check in cppunit/ExternalProject_cppunit.mk
which caused cppunit to always be built without -D_GLIBCXX_DEBUG.

Change-Id: Ia247dcd84e2c6fa0e9384fd27643537984d980b5

Change-Id: I67e38db4b41849ac4be4ad20825391c84c7ed412

Change-Id: I08ab04005720e0de826d6a0cd33211e3a70d5965

Change-Id: I927f7ea5cae396bb10ffc2568ec68ae69b8f3f82

adds another tweak that had been removed with some print related rewrite

Change-Id: If63072c9b2c6158d46f26933fd54766fd808569c

Change-Id: I774af9ca0631e931ddc8c5025981dabca557e325

pread / pwrite don't look as a Linux and Solaris privilege.

Change-Id: Ifb2e88445d4064c13a406007bfd523ae0caa38e5

This is a big commit, so thanks for your time reviewing this :)

Conflicts:
	basic/source/sbx/sbxscan.cxx

Change-Id: Ib9bc710b87475b5695764557321e5dcef25e5933
Signed-off-by: Miklos Vajna <vmiklos@suse.cz>

Or else Calc would crash while loading the xls document from
gnome519788.

As an aside, this xls document appears to be corrupt. Trying to
open it in Excel (XP and 2007) causes Excel to offer to repair it,
while fails in both versions.  Excel XP is somehow able to open it
with some content preserved. No such luck with Excel 2007.

Change-Id: I04616a4c926862461a2efdd99ccabe36122d6825

Or else it would crash when loading the doc from fdo#55174.

Change-Id: I21eb77f97dcab467c185dacaf9f00753bb4fc18c

Not sure what's wrong exactly, but on one tinderbox cl fails because
of "unknown" argument (unix path to the source file). Work it around
by explicitly converting the path to windows path.

This should fix the crash with gnome#627420.

Change-Id: Ibfff498282dc1c6fe9124ced645392107ef8829f

This should fix the crash with kde#245919

Change-Id: I7c3af01e27233d2f94d5585247c59e7d5b6ea8ca