Change-Id: I798376268b15f8b2f340d48fd2e38d68e53dbc6f

Change-Id: Ibdd144158f1484cf0e8c3861b6f3758a03638434

It doesn't really matter, just saves some cycles looking up the system
locale.

Change-Id: I8895593b445c49f167f26bcbf510c12a3c25e1d4

Change-Id: I289900b998c5a25307b76efb723aaf12dd6431b1

_DelBookmarks() will actually delete marks that exactly match both start
and end position of the range, so restrict the call to only the
fully-deleted nodes that will be moved to the Undo-array.

(regression from 370febbf)

Change-Id: Icf5097774aa55ee152a1e20c0c7264330955c615

These call malloc so often that valgrind forgets where the things i'm
looking for were freed, because the address has already been re-used.

Change-Id: I8e1a0f739a774972d3afe750250ee79262c79b85

Change-Id: I78efb852e5a240c90d3f8e70a3c0a358030c0dde
Reviewed-on: https://gerrit.libreoffice.org/15578Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Jenkins <ci@libreoffice.org>

Change-Id: I32368329edfff18ea2a73de9a4cf60b2f734a54d

CPPUNIT_TEST_NAME is the environment variable that needs to be set and
contain the name of the tests. The test names need to be fully qualified
to be recognized.

Examples:

CPPUNIT_TEST_NAME="ScFiltersTest::testOutlineODS ScFiltersTest::testRangeNameXLS" make CppunitTest_sc_subsequent_filters_test

CPPUNIT_TEST_NAME="ScFiltersTest" make CppunitTest_sc_subsequent_filters_test

Change-Id: I78a8a076638e19c918ca1c411086bb353f5100a2
Reviewed-on: https://gerrit.libreoffice.org/15579Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

AppCompat 22.1 is stricter in regards to what flags it it accepts

Change-Id: If9b77e59604147cb7050b5721380ab213fcdb9a6
Reviewed-on: https://gerrit.libreoffice.org/15571Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>

Change-Id: I65c20f75c4086dc5167c6fa41b0171f07c1419bb

(and MenuButton::GetSelectHdl was unused)

Change-Id: I94df06bcda9e041fde65553e30247874b20ff74b

...as it will be used to implement a function that needs to be compatiblie with
some specific Link<Arg,Ret> type.

Change-Id: Ieabd0ee499850a304e0c6e625953532be47ac784

Change-Id: Id26adce17a2479eead935958556012f13f2821b7

Change-Id: I56d61b577df00855a49dd618e9dafcdb86cd7ca4

Change-Id: Id85f33eab2767961f24688fffac84416b1cb22d8

(fully qualify ::tools::detail::makeLink to avoid clashes with namespace
basegfx::tools)

Change-Id: I52818d0a47c78f1af7673ff91eb6517bc0820655

Change-Id: I1fd2e76c098d0bffb446a1b2133d3392bfbb3de7

Change-Id: Iaa6f353fb4819bc238f2a914303feb162e7f4eba

Change-Id: I8dbfdf4bb2eceac0b5afbddd3f35e1dcde2db68b
Reviewed-on: https://gerrit.libreoffice.org/14611Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Eike Rathke <erack@redhat.com>

In case of embedding a Base document, there is no root storage and
attachResource() is not called with a parameter to be able to make one,
all we can (and have to) do is to store the component to the target
store given as a parameter to storeToStorage().

With this, a .odb embedded in a .odt no longer lost on export. Import is
still broken, though.

Change-Id: I1d92a1d28f27f49609ef70fbf16effb592b9edf5

Change-Id: I0ab8f98b64ac56b1f7b76ad99b433b6a2c370e27

Change-Id: I4dff83941df62bcd5ec5a70a87f59e9ce17c979d

The "own" filter of Base has FileFormatVersion=6200.

Change-Id: I3bbd28ebcf4d517ea9cefe692b01b688358c51e8

For us our own configured locale is even more important.

Change-Id: Idfab979e868f49e2323dc9f516996393112b0689

Problem Description:
- CharKerning i.e spacing between characters/letter is not exported
after roundtrip.

XML Difference:
Original:
<a:rPr lang="en-IN" sz="6000" b="1" kern="0" spc="2000" baseline="0">

After Roundtrip:
<a:rPr b="1" lang="en-IN" sz="6000" strike="noStrike">

spc is missing after RT.

Change-Id: I8bcefd6f0d0e3a1825618bece6f563792633a272
Solution: added support for spc.
Reviewed-on: https://gerrit.libreoffice.org/15575Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Problem Description :
The xlsx file has the extended document property <TotalTime>#n</TotalTime> in /docProps/app.xml
LibreOffice appears to read this as a a total editing time of in second_ but according to ECMA-376 2nd edition:
22.2.2.27 TotalTime (Total Edit Time Metadata Element)
Total time that a document has been edited.

The default time unit is minutes.
(ECMA-376 1st edition has the same statement in  7.2.2.27)

XML difference: None
After RT : None

Solution :
As LO uses the TotalTime in seconds, while importing the TotalTime,
the TotalTime mentioned in the app.xml [which is in minutes] converted it to seconds.

TODO: Writing a UT seems to be tricky for this change. 
Need to analyze and will raise the UT in separate CL.

Change-Id: Icbe998b9881ac241a9147628573bab6ef63582c4
Reviewed-on: https://gerrit.libreoffice.org/15479Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

This patch fixes only a part of the issue. It respects the header if
chosen by the user in the autofilter dialog showing when the first row
contains empty cells or numbers. With minimal effort a lot is fixed
(80-20-rule). With headers, the autofilter sorting works with empty or
number cells in the header row. (The main part of the issue.)

What this patch does not solve: If there is no header row, but only data
rows. If the user's choice in the dialog is no header for the autofilter,
then the first row is sorted too. The autofilter buttons will move
toghether with the first row. The autofilter buttons should stay at the
first row, even if the data of the first row changes.

In conclusion, this patch improves the autofilter sorting code, though it
does not address everything of the issue.

Change-Id: I9cc56577a90586bf8575c9630e5c03871198c04c
Reviewed-on: https://gerrit.libreoffice.org/15277Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Change-Id: Id34a655c554e78476e3bd83c87de3cbe1b557980

Change-Id: I87311d8989c23538983d45ad9b12a64080441d78
Reviewed-on: https://gerrit.libreoffice.org/15569Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>

Change-Id: I7a7d9cc7948ee08211cbfaa3df8197c1ede148b7

...in light of the various UNO integer types

Change-Id: Ie525e8cdd1d155446985ddfb2667714be2dd4904

Change-Id: I29ade324b4bb5eb34ce05107794f0ac9a4834c6e

Change-Id: Ic9eebca9d8fe3e9173b1873bdc7800e5161f9999

Change-Id: Ib03c6ba0eac9400eba89674ca92f638baf65b203

Change-Id: Ifc82f42ff8d82cfe094701ace57ff7cadabc59d9

Change-Id: I6af270f1c7ae5348b96a1fb2948b6aa675d39ace

Problem Description:
After RT 'rtl' tag of <a:pPr> are not exported

XML Diff:
Original :
<a:pPr marL="457200" indent="-457200" algn="r" rtl="1">
RT :
<a:pPr algn="r">

Change-Id: I307bc748c73127c311cabd127bb98e43bd298271
Reviewed-on: https://gerrit.libreoffice.org/15428Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

In case of 'no fill' for vml textboxes, MS office rendered as
transpanrent textbox.
When opened same file on LO, it rendered it as textframe with white
background. (Even if the fillstyle is type of 'no fill' and 'backcolortransparency
' is 100%)
So, Fixing this issue by applying 100% 'FillTransparency' if it gets above situation.

Change-Id: Ied2d3855354ada994288c29c78d9ccb4e5b0b4d6
Reviewed-on: https://gerrit.libreoffice.org/15390Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

If a line happens to be invisible (e.g. in too thin cell of a table)
then aLine.FormatLine(nStart) returns nStart, and
aLine.Insert( new SwLineLayout() ) is executed until OOM.

Change-Id: I9b12cf2f1f3f1987cd8ce380b1d85196ffe95e5e
Reviewed-on: https://gerrit.libreoffice.org/15542Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>