Public Git Hosting - LibreOffice.git/log

fix sometimes data corruption with xlsx import ( with scenario sheets )

formula import is buffered 'till the end of import. The processing of the
formula data need to happen before the scenario import happens. This is
necessary because sheet numbers stored in the formula addresses can become
invalid as scenario import insertes new hidden sheets upsetting the previous
table order

Change-Id: I9357f028f31bec1b1504ca991f5534f80d79c9bc

fdo#58142: fix frame selection in SwPageFrm::GetCrsrOfst:

The function does not handle the case where a paragraph was hit but no
background frame was hit properly, which may or may not (depending on
the platform) cause it to return a position of a frame when that is
undesirable, such as when called from SwWrtShell::UnSelectFrm().

When UnselectFrm() leaves a frame selected pasting multiple pictures
leads to a crash.

(regression from e8fbe97900f13305b17015d9044993bde4adab36)

Change-Id: I3604825f654c523a37f47a9fc660647d8b3e7077
(cherry picked from commit 2fc4af311656e7d019ca802d23444e5161f820ee)
Reviewed-on: https://gerrit.libreoffice.org/1314
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Resolves: fdo#49587 apparent loop in style hierarchy

Change-Id: I32317a1a6500377f7e7bb98d1dcc65ad86feb9a2
(cherry picked from commit 7e3923520c27ff139bc3abab35a1c01e1a785fb1)

Signed-off-by: Michael Stahl <mstahl@redhat.com>

fdo#57780: sw: lcl_ArrangeLowers: register drawing ...

... objects at the right page. This was effectively disabled for all
anchor types except FLY_AT_PARA by a misplaced ! in the conditional,
and C++ not being a type safe language.

Change-Id: I64af2141e577ac896292eecd9f8c46c62a0548db
(cherry picked from commit c479ccc438edeb7197a2d05bdb99067365d7bac0)
Reviewed-on: https://gerrit.libreoffice.org/1290
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Eike Rathke <erack@redhat.com>

fdo#52240 added [hr-HR] date acceptance patterns D/M/Y;D/M

As per https://bugs.freedesktop.org/show_bug.cgi?id=52240#c59

(cherry picked from commit b1184f8a4e4a3949d38407c5f1d5155c2c2b42d5)

Change-Id: I83afc59e4ac71b18219ace60dcbcabafebd5b525
Reviewed-on: https://gerrit.libreoffice.org/1288
Reviewed-by: Kohei Yoshida <kohei.yoshida@gmail.com>
Tested-by: Kohei Yoshida <kohei.yoshida@gmail.com>

fdo#52240 added [no-NO] date acceptance patterns D.M;D/M/Y;D/M;D/M Y

Also referenced by nb-NO and ny-NO
Changes as per http://erack.org/blog/archives/18-comments.html#c351

Change-Id: I9e07f4307800e99fdff21bccde91af4178b244f3
(cherry picked from commit 638993d9b80b3793dfd31db0dacba3746ceb98fd)
Reviewed-on: https://gerrit.libreoffice.org/1287
Reviewed-by: Kohei Yoshida <kohei.yoshida@gmail.com>
Tested-by: Kohei Yoshida <kohei.yoshida@gmail.com>

Resolves: fdo#51638 In "unopkg gui" dispose component context from DeInitVCL

...the same way it is done in soffice.bin. framework's Desktop::dispose()
requires the solar mutex to be still alive, which is destroyed in DeInitVCL, so
if the component context/service manager is only disposed afterwards, the solar
mutex is already gone.

This required moving disposeBridges() around, but it allowed to get rid of
DisposeGuard.

(cherry picked from commit 37cc83e594fa8ca131fc5fb98506287b7daedffd)

Conflicts:
desktop/source/pkgchk/unopkg/unopkg_app.cxx
desktop/source/pkgchk/unopkg/unopkg_shared.h

Change-Id: Ibec3d19040fdae23f492cd1e29084e673403e00b
Signed-off-by: Stephan Bergmann <sbergman@redhat.com>

fdo#55289: fix crash in SwXShape::setPropertyValue:

In the libreoffice-3-6 branch the docx import crashes here when setting
the anchor position of a shape, because the previous setting of the
AnchorType was not done properly: the position of the anchor in the
SwFmt did not match the actual position of the text attribute, so we get
these assertions: "Missing FlyInCnt-Hint." and crash.

This cannot be reproduced with the docx import on master, because on
master a different intermediate AnchorType is set, and transitioning
from that happens not to cause the crash. But it can be reproduced with
a unit test.

The regression was introduced in libreoffice-3-6 with
backport commit bbbb10a0774b9da546557632cd09c4f6958f8939.

Change-Id: I394643e11862543177ba57958a26cd8ce06dc09c
(cherry picked from commit 01cd24df1ec7eaa53e9e2156d4d6890a7cc5996a)
Reviewed-on: https://gerrit.libreoffice.org/1264
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

handle properly anchor transition at page->at paragraph->as character

Otherwise setting text range afterwards crashes with pHnt being NULL.

Change-Id: Ib4ac2712c061605dcaaa262280c0307f7a9af2ce
(cherry picked from commit 5de61f7a4ddb502730454e4bf3232f7f4b4175e1)

Signed-off-by: Michael Stahl <mstahl@redhat.com>

fix incorrect OSL_ENSURE -> SAL_WARN_IF conversion (condition not inverted)

Change-Id: I57dfa64c55e046ed51797b2abeb7ba3242e637da
(cherry picked from commit dfc4832049c140b7442c085062599aef1953e8fd)

Signed-off-by: Michael Stahl <mstahl@redhat.com>

Resolves: fdo#57532 restrict page sanity check to non-storage chunks

Change-Id: I47c24131e95b9a8a691dbff4710869ba91df8306
(cherry picked from commit 4a1044229afd6b52ca10e5c55f1957b153a8c6ea)
Reviewed-on: https://gerrit.libreoffice.org/1245
Reviewed-by: Michael Meeks <michael.meeks@suse.com>
Tested-by: Michael Meeks <michael.meeks@suse.com>

fdo#52640 fix right-aligned tabstops

It is no good to change temporary object...

Change-Id: Ib1178e154341dd0216fd965d3ed6c264de40d634
Signed-off-by: Kohei Yoshida <kohei.yoshida@gmail.com>

fdo#54473 fdo#57133 fix RTF import of character styles

Regression from commit d2d77fd86e6cb7341e113557e2cfe3d7948759bd

(cherry picked from commit 41890bf714803969515578b47e63fa9730ea7090)

Change-Id: I7accc5c383de3fa53c12527766c3fc155bfa9863
Reviewed-on: https://gerrit.libreoffice.org/1235
Reviewed-by: Fridrich Strba <fridrich@documentfoundation.org>
Tested-by: Fridrich Strba <fridrich@documentfoundation.org>

Resolves: fdo#56549 crash in format->autocorrect->apply and edit changes

regression from cad9afa15f53d547733fa55f1353772f6d696611

Change-Id: I1aa36bf86969a5a90456817d360d59125948e519
(cherry picked from commit 32c26ffb1f7585318a11f684a9886b58cea5b307)
Reviewed-on: https://gerrit.libreoffice.org/1218
Reviewed-by: Miklos Vajna <vmiklos@suse.cz>
Tested-by: Miklos Vajna <vmiklos@suse.cz>

resolved fdo#57668 additional date formats using '/' separator

Change-Id: Ib71fb25b8d5411b83f55649d9a18833470647eb3
Reviewed-on: https://gerrit.libreoffice.org/1196
Reviewed-by: Björn Michaelsen <bjoern.michaelsen@canonical.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Fix lacking "/" on .icedove

It would prevent to find icedove addressbook if there's no link .thunderbird to .icedove
I already did the same on master sources, see:
https://gerrit.libreoffice.org/gitweb?p=core.git;a=commitdiff;h=ea5a96a5c771ee2c2ebbe4c12c4e7f3b1cba29a7

Change-Id: I24d1ff9b4af327a810997b97b0cf4112eba21e64

fdo#57611 do not crash when setting too small size

setWidth/setHeight's exception specifications are lacking,
but fixing this in a stable release line is too invasive:
they are "inherited" from the specification in
offapi/com/sun/star/report/XReportComponent.idl
and are thus part of our public API.

Change-Id: I8ae2048a6c0acc6b6cf2beece9e630c87caac39f
Signed-off-by: Stephan Bergmann <sbergman@redhat.com>

lo-commit-stat: allow to filer cherry-picked commits

Add --cherry option that filters commits using the "git cherry" command.

Note that you need to pass git arguments for "git cherry". It means
"old-branch-or-tag" "new-branch-or-tag".

Change-Id: Iea67d0ead205c66112791cb0444fa183c7fa6e9b

lo-commit-stat: allow to generate more logs at once

It takes ages to get bug summary from bugzilla => it is worh to
generate normal and wiki logs at the same time. Why not produce
all logs with one call

Add --commits option. Also rename --wikibugs to --bugs-wiki.

Use .wiki suffix for wiki logs. Also rename "commit-log" to "commits".
It is enough to use ".log" as the suffix.

Change-Id: I92bbc4d56a0ae9e23401be0677256059c777d712

Make commit stat script recognize opensuse fate entries, too

Change-Id: Ia6b3e6459c5bda7ea24091024cf3dabd19746237
Signed-off-by: Petr Mladek <pmladek@suse.cz>

Slightly more robust removal of bug title prefix.

Change-Id: Ic37589222831d03ec48689a077b1eb16a9199385
Signed-off-by: Petr Mladek <pmladek@suse.cz>

fix bug 53508

In the format paintbrush function :
Apply the paragraph automatic attributes to all the nodes in the
selection instead of just the last node.

Change-Id: I655f00cbf44d3d80c19a7ef623bc1c7cb505ead9
Reviewed-on: https://gerrit.libreoffice.org/1180
Reviewed-by: Norbert Thiebaud <nthiebaud@gmail.com>
Tested-by: Norbert Thiebaud <nthiebaud@gmail.com>

this method actually can throw, related fdo#57451

Change-Id: I26b79829029e127eb8b9ab4fbea3e0f02035ee41
Signed-off-by: Michael Meeks <michael.meeks@suse.com>

we need to catch the exception here, fdo#57451

Change-Id: I203aa640e7d4373f3a090a4988c28c6059b93064
Signed-off-by: Michael Meeks <michael.meeks@suse.com>

vml import: only apply width-percent attribute if it's != 0

This fixes an issue with a shape defined with these attributes:
mso-width-percent:0;mso-height-percent:0 and
mso-width-relative:page;mso-height-relative:page;
where all points were then located in (0,0)

Change-Id: I51070ad2b2e4e05ab64f16813472ca1d7099fb09
Reviewed-on: https://gerrit.libreoffice.org/775
Reviewed-by: Petr Mladek <pmladek@suse.cz>
Tested-by: Petr Mladek <pmladek@suse.cz>

fdo#52182: Fixed click in frames located in header/footer

Using a distance to click to select the best object to select between
normal text and background object.

(cherry picked from commit e8fbe97900f13305b17015d9044993bde4adab36)

Conflicts:
sw/source/ui/docvw/edtwin.cxx
sw/source/ui/inc/edtwin.hxx

Change-Id: Ib5b53161c7af2c16f4df379382f2e53fc6d8092b
Signed-off-by: Michael Stahl <mstahl@redhat.com>

oox: v:textbox's inset attribute should be imported in Hmm, not in Emu

Trivial reproducer: http://people.freedesktop.org/~vmiklos/2012/simple-textbox.docx

Change-Id: I030f7153326affc5a01e7ede1ddf1164fa500071
(cherry picked from commit 5bc835d48f1e5373b3a31bd166fadfdf19b3588f)

.dot files sometimes rejected by writer, bnc#793422

because CWW8 appears twice, once for normal .doc and once for .dot. So a .dot
files is tested twice if it's supported by the WW8 filter. Depending on the
(effectively arbitrary) order they appear in the list .dots may fail the "is a
normal non template .doc" test after the "is a template .dot" test and get
rejected as a CWW8 candidate.

Change-Id: I88aec29fdd5f9ec4dd4ad2813ff3c6b8fa5c5461
Signed-off-by: Tor Lillqvist <tlillqvist@suse.com>

optional ppFlt argument never used

Change-Id: Ib2c68091f14d04818ae5e7db0ad7b8a1ec151f4f
Signed-off-by: Tor Lillqvist <tlillqvist@suse.com>

Simplify SwIoSystem::IsFileFilter

bRet = bRet && (pFltr->GetUserData().equals(rFmtName)); is already inside an if
bRet = bRet && (pFltr->GetUserData().equals(rFmtName)); condition

bRet = xStg.Is() && IsValidStgFilter( *xStg, *pFltr ); is already inside an
if ( xStg.Is() ) condition

Change-Id: I2115041d99b6fcbae7a829557c37627d03b5ee24
Signed-off-by: Tor Lillqvist <tlillqvist@suse.com>

bnc#791952: Setting properties for all data labels.

You can set different properties to individual data labels. But setting
new properties to all data labels should overwrite the individual settings.

Change-Id: I934bad89326115aa83feb86275cf380d701a5686

bnc#791952: Use "center" label placement for "Best fit" in pie charts.

It works better this way for Excel interoperability. In Excel, Best fit
is slightly different than the center placement, but the two are pretty
close in pie charts.

Change-Id: I7f04a1babb04fd488d8cf90df247c294fa9d7b67

bnc#791952: Better placement of 3D pie chart data labels.

This value appears to control the logical vertical offset of data labels
along the z-axis for 3D pie charts (no effect in 2D pie charts). With
-0.5, the data labels appear lifted from the top surface of the pie chart.
With -1.0, the labels get placed right at the top surface, which improves
the precision of data label positions.

With this change, data labels that are centered are really centered inside
their respective pie segments.

Change-Id: I26d5d7cb7f68bd09131b70bcb4fc030ac924bcce

fdo#57497: Remove group table data when all group dimensions are gone.

Otherwise ungrouping date grouped dimension may crash, or produce
incorrect results.

Change-Id: I3634e3c0bf8336fc1221f5d234cb7e01eb1f07c6

use snap rectangle instead of logic rectangle

- so that pictures inserted into rotated placeholders are placed
in the transformed rectangle and don't endup at wrong position
e.g. outside of slide

Change-Id: I2f3d206b43639c094f7dfb1d25e42b9cbf7dd102

make presentation spanning all screens work (bnc#758138)

This reverts an API "improvement" from 6ee5dfa150d408179e20a5525ff7ec46649e0e58
that tried to use -1 as the current screen/display but failed. The check
for nDisplayScreen being in range broke all-screens (which is
represented as last_screen+1 value, which is broken in its own way, as e.g.
the screen count can change). The same way -1 as the default value is broken,
as everywhere else invalid invalid screen number means spanning all screens.
The extra overloads without a default value should take care of that.

Conflicts:
vcl/source/window/wrkwin.cxx

Change-Id: Ie118038eacacebc007c25860732d5904ad0de2b9
Signed-off-by: Thorsten Behrens <tbehrens@suse.com>

STARTCENTER_HIDE_EXTERNAL_LINKS 1

Change-Id: Ifa0b49d713648fcc0ce4569a6eaf89c79117086c

remove obsolete configure switch --enable-activex-component

Change-Id: Ib5e2e5f2b44af830e85d966be1d1b1dc857f6f2a

Release SUSE 3.6-3, build id 303, Linux version 3.6.3.2.3, tag suse-3.6-3

Change-Id: Ic9aed7618934ff534e3b166c1b0b47a38ed744f1

bump product version to 3.6-3, release number to 303

Change-Id: Id4e078c4055c71589dd70e2fe1d7898b99ac3853

Resolves: fdo#53909 STG_FREE sector locations rejected

(cherry picked from commit ef9f9b750eac2966634c385a2021716c4de0e4cd)

Conflicts:
sot/source/sdstor/stgelem.cxx

Change-Id: I6f71c7e3cfeeba8a76096d5be68838b8d401c580
Reviewed-on: https://gerrit.libreoffice.org/1150
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
Tested-by: Stephan Bergmann <sbergman@redhat.com>

I hate sal_Int32, episode #1326.

Change-Id: Ibfda2b553bae1ad040f722dde46a352a313218fc

no matching function for call to 'max(int, long int)'

Change-Id: I7533098781b0b6987cd852d74a688aac6e0af056

fdo#56584: delete also note caption.

Change-Id: I1d792082e79d6bb68004a84c172cc3bb5c194826
Signed-off-by: Kohei Yoshida <kohei.yoshida@gmail.com>

WaE: fix windows build of npshell.
(cherry picked from commit 655ef3e507d1bc41e67cd7529f827ac58fb9bab0)

Signed-off-by: Stephan Bergmann <sbergman@redhat.com>

WaE: duplicated variable name

Change-Id: Id42116f2ddfe679b1a995590e35a619ff8d82d7e
(cherry picked from commit fdc76a03f50e1291a90035befac67b8205c7e770)

Signed-off-by: Stephan Bergmann <sbergman@redhat.com>

fdo#55034 Update default Hebrew font from Lucida Sans to Lucida Sans Unicode

Change-Id: Ia2204dda0cb452a4ce38a44db61d17d6802b5a38
(cherry picked from commit a40f08c1ac97078fa501da7c6dc456aa12146f25)

Signed-off-by: Lior Kaplan <kaplanlior@gmail.com>

fdo#54898: Test equality by order index (integer) which is more stable.

At the point where std::unique is called, we can use order indices to
determine whether the two items are equal. This should be more stable
than using CaseInsEqual() to assess equality.

Change-Id: I88310fc7beede19fb1c629b9b7e3cb9a069b2b23
Reviewed-on: https://gerrit.libreoffice.org/1013
Reviewed-by: Tor Lillqvist <tml@iki.fi>
Tested-by: Tor Lillqvist <tml@iki.fi>

remove conditional formats withiout range after import, related fdo#56983

(cherry picked from commit eb240530443b964ea45bf80a5e3711e11a3f09c5)

Change-Id: I9a8c02369c3e5a12310a46229bdc1671f75593ad

export embedded spreadsheet(s), fixes n#780830

Change-Id: Iee8328a695b554d97f6805b64b98706b9e3d7ab9

limit buSzPct range

Change-Id: I41411301ee131f92b9725208e53133ee0f8a6f82

only set entries for existing sort entries, related fdo#56958

Change-Id: Ib48606402fb90b259f705badeac294d7c336cb9c
(cherry picked from commit c45b6496e87fe74570a41821bc5f4fbdb3d3cf8e)

Signed-off-by: David Tardon <dtardon@redhat.com>

make sure that the sort vector is large enough, fdo#56958

Change-Id: I7099e984a8b05ccd3324f3b9231c193e716eaf83
(cherry picked from commit cd545dcf511daedd548c11159b8c244f272e9cb6)

Signed-off-by: David Tardon <dtardon@redhat.com>

Fix MinGW build.

Change-Id: I48bad55cc982fba6fb48a921746801de9993bff1
Signed-off-by: Luboš Luňák <l.lunak@suse.cz>

fix for fdo#55875 numbers as text converted strangely

old code used to use XCell->setString, new code uses rDoc.SetString which by default tries to detect number formats. The ScColumn::SetString that eventually
gets called seems to do lots of additional checks ( and apparently even if
an ScSetStringParam instance with mbDetectNumberFormat ( false ) was passed
it seems that it will still try to detect decimal number formats. With that
in mind I restore and un-unoified version of what XCell->setString used do

Change-Id: Ifaef74c78b198f492a390a3d5dc1721622a01ea4
Reviewed-on: https://gerrit.libreoffice.org/1020
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

fdo#51550 fix DOCX export dataloss on non-math/chart OLE export

This is still not complete, but having the replacement graphic only is
far better than having nothing.

(cherry picked from commit bd6ae389008e110be62a335dfcd82c655d512e63)

Conflicts:
sw/source/filter/ww8/docxattributeoutput.cxx

Change-Id: I141a3de1a449f4261c7086e10f2c141b3f6cdb10
Reviewed-on: https://gerrit.libreoffice.org/1017
Reviewed-by: Norbert Thiebaud <nthiebaud@gmail.com>
Tested-by: Norbert Thiebaud <nthiebaud@gmail.com>

fix build with icu-4.0

use functions already provided by icu-4.0

Change-Id: I238a499ae1bc2a7efeb4d61fbe780e24d7d36a7a
(cherry picked from commit 524461f1721d0fb7c5bd115d72ca82711e0ff950)

Signed-off-by: Caolán McNamara <caolanm@redhat.com>

Enable NPP_Initialize/Shutdown again

...calling which had erroneously been dropped in
bdead69dd3c947743907681c8a2f16ca1f50cd9c "Add switch for system/internal
npapi.h." This makes the npsoplugin browser plugin work again.

(Also, init write_fd to -1 instead of 0, so that a missing do_init_pipe does not
accidentally cause writing to write_fd succeed in writing to stdin.)

(cherry picked from commit 9081e07c9c92267d86a90752b28bd42b592782a8 plus follow-
up 6ce21d57d9a627a95bfb2f10519ae3cd032f521b "Typo")

Conflicts:
extensions/Library_npsoplugin.mk

Change-Id: Iced734512a43c8dfea46088f7ac16f159e7dd807
Reviewed-on: https://gerrit.libreoffice.org/1042
Reviewed-by: Tor Lillqvist <tml@iki.fi>
Tested-by: Tor Lillqvist <tml@iki.fi>

don't always show integer part for fraction formats with hard denom fdo#56205

fixes bug when you have a fraction format with a forced denominator e.g. '?/5'
which showed the integer part of the fraction ( as if the format was '# ?/5' )
but even without the space
e.g. before fix
pi 3.14159265358979 with format '?/8' would be shown as "31/8"

after the fix
3.14159265358979 with format '?/8' would be shown as "25/8"

Change-Id: I1feb8b78af94b90db1bcc30da248077243dd8dfc
Reviewed-on: https://gerrit.libreoffice.org/1047
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Eike Rathke <erack@redhat.com>
Reviewed-on: https://gerrit.libreoffice.org/1064

fdo#56776: Change in formula options should cause repaint.

Else Calc would fail to update the column headers due to A1 to R1C1
change (and vise versa).

Change-Id: I2c403212f5bc539aea92370e60da96f480549d6d
Reviewed-on: https://gerrit.libreoffice.org/1049
Reviewed-by: Noel Power <noel.power@suse.com>
Tested-by: Noel Power <noel.power@suse.com>

resolved fdo#56278 broadcast deletion to dependent formula cells

Consolidated changes for fdo#56278 fdo#54074 fdo#53364 fdo#55059 (and
earlier i#99844 and i#101869). The replacement cell for the broadcaster
to inform listeners, if any, is transferred the broadcaster and owns it,
the remembered formula cell never does. When broadcasting the change use
the replacement cell instead of the remembered formula cell. Don't
fiddle around with ReleaseBroadcaster() at the remembered cell, we don't
delete twice and now also don't leak.

(cherry picked from commit 8165bfa2afbe828caf37df57b83cb31bef2559cb)

Change-Id: Ia11b6d3d751c784af64aaed05f8508887f869a80
Reviewed-on: https://gerrit.libreoffice.org/1069
Reviewed-by: Norbert Thiebaud <nthiebaud@gmail.com>
Tested-by: Norbert Thiebaud <nthiebaud@gmail.com>

resolved fdo#55369 accept fraction input if preset

Accept fraction input without integer portion if the format was preset
as fraction, e.g. 1/5 instead of 0 1/5 and don't force to date.

(cherry picked from commit 5391bd1e54157457abe1098b29f067d6e0059113)

Change-Id: I326ec85b163962425efa074119405ec1395af481
Reviewed-on: https://gerrit.libreoffice.org/1045
Reviewed-by: Petr Mladek <pmladek@suse.cz>
Tested-by: Petr Mladek <pmladek@suse.cz>

n#780645 DOCX import: add support for fixed column width

In general, DOCX tables provide a grid for the table, then use spans in
case different number of cells are used in different rows. In this case,
the cell width is ignored, as the wished width can be counted from the
grid and span values.

However, it's also valid to simply provide more cells then the amount
defined by the grid, and in this case the cell width should decide the
final width of the cells. This commit adds support for this later case.

(cherry picked from commit ad913681c0e0cdaaa706329f518f59ee54b86d43)

Conflicts:
writerfilter/source/dmapper/DomainMapperTableManager.cxx

Change-Id: I0dd6c1f0c06f81c2afa00489b7ad1f33ff300a7c

n#780277: select field when clicking it instead of having cursor before it

Change-Id: I93d623b62e9357a71bfca3403cc98092ef58c7e4

n#782833: Rotated text in emf images looks thicker.

Any rotated text in an emf image (image size != original size),
has thicker fonts.
This is kind of a regression caused by commit
8c2902ecbf135fae070ea032fa10a59bb76ec1b0
So, partially reverted it.
The code was kind of a workaround/copy-paste, so I am assuming
it is safe to revert.

Side-effects: The dotted borders (in calc) look a
little lighter and not so very visible in the page-preview.

Windows: Enable nicer toolbars even for XP, not only Vista and higher.

Change-Id: I26837290d1416637032d2dfcfa45b4003c874309

n#789482 DOCX: export track change data after w:hyperlink

(cherry picked from commit eac3e6e746300df379226941ba75c4e0ce1feb7a)

Conflicts:
sw/source/filter/ww8/docxattributeoutput.cxx

Change-Id: If204523d7da544b11b2d809993ada180476104ef

bnc#681110: Fix squashed glyphs with Cairo canvas.

I suppose it is the same misconception as in fdo#55931 - I am convinced that
the font metrics should not be used here.

Looks to me as if in the past, the font width computation was broken somewhere
deep inside OOo, and everyone instead of fixing the root cause was just
working that around by using the FontMetric; and then one day the root cause
was fixed, and all the workarounds broke ;-) - but that is just a theory.

Change-Id: I741bf8f4eaea4f7d8bc698dc9a8124109dfb8c20

Resolves: fdo#47432 Improve the "faux bold" in cairo canvas

The "faux bold" make the bold font in Asian ugly & unacceptable,
Adjust & improve the "faux bold" to make the bold font better.

Change-Id: I2e02156ffc30629476b80134ffb0b50e793f7ead
Signed-off-by: Jan Holesovsky <kendy@suse.cz>

[SUSE only] update EULA

Change-Id: Ia55ff9b54202dad6c0df46a69235543d88f43e61

implement inset attribute of <v:textbox> (part of bnc#773061)

Change-Id: I1ec9b5d9fed86c44dc0a412c2323033ea67985c9

bump buildid to 302 and version suffix to -2

The build is based in the upstream 3.6.3 and will be called SUSE 3.6-rc2

Change-Id: I04337c2d4001eec123cfb1fe1043b4955a5bba2d

fdo#55138: Do not double-expand bootstrap variables

For some reason that escapes me now, the old code double-expanded URLs in
CONFIGURATION_LAYERS and in extension layer's configmgr.ini SCHEMA/DATA. That
produced wrong results when the pathnames represented by those URLs contain
any "$" characters, as apparently routinely happens on Windows with roaming
profiles.

Change-Id: Iff149e7d8736cbcda579376cdc89e24cf99ccc97
(cherry picked from commit 8246bdb500a6d83ea9b61f58fb74e051432e32d7)
Reviewed-on: https://gerrit.libreoffice.org/994
Reviewed-by: Michael Meeks <michael.meeks@suse.com>
Tested-by: Michael Meeks <michael.meeks@suse.com>

add license info

Change-Id: I56653d99e1fac5e1b7cd2a69b4b160df4b75bb5d
(cherry picked from commit a7d38322d6211863024456602da1dff23a8316dd)

Signed-off-by: Caolán McNamara <caolanm@redhat.com>

fdo#52989 ignore provided size of JPEG files during RTF import

See commit 66fa759b for rationale, this fix just does the same for JPEG
files as well.

(cherry picked from commit ebc61e11cdb02f5cc33aeabead3d191eaf0d23d3)

Conflicts:
sw/qa/extras/rtfimport/rtfimport.cxx

Change-Id: I6a9ead5f4b4b9c4661cae99501bd7c14a2a79409
Reviewed-on: https://gerrit.libreoffice.org/991
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

fdo#48442 fix default hori/vert frame anchor during RTF import

We used to send nothing when we got nothing, but this is not correct:
\pvmrg and \phmrg is the default in RTF, but not in Writer.

(cherry picked from commit 7b7bee4ed5722ce59ffd3394a0330d71d69d66b3)

Conflicts:
sw/qa/extras/rtfimport/rtfimport.cxx

Change-Id: I9f69e282e68f0828c8b5ba98657cad1dd0715eb3
Reviewed-on: https://gerrit.libreoffice.org/993
Tested-by: Caolán McNamara <caolanm@redhat.com>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>

Fix fdo#56603 by NULL pointer checking in SwDrawTextShell::Execute

Pushed in master as c954e6e1317f82e06fac66556674bc066b429469

Change-Id: I98f246b2bafcf2669a8f6c3ec4c86ccee56519bf
Reviewed-on: https://gerrit.libreoffice.org/984
Reviewed-by: Norbert Thiebaud <nthiebaud@gmail.com>
Tested-by: Norbert Thiebaud <nthiebaud@gmail.com>
Reviewed-on: https://gerrit.libreoffice.org/985
Tested-by: Miklos Vajna <vmiklos@suse.cz>
Reviewed-by: Miklos Vajna <vmiklos@suse.cz>

fix ooxml color export

Change-Id: I4b94883f594e13e9923bdd05593447a152fa8b0a
Signed-off-by: Kohei Yoshida <kohei.yoshida@gmail.com>

export dxfs at the correct place

Change-Id: I2ab31904169adad4f173d4655d0511855de77464
Signed-off-by: Kohei Yoshida <kohei.yoshida@gmail.com>

fdo#53531: Time field can be represented with an extended time field type.

And Calc's header footer currently don't distinguish between time and
extended time. Time fields in the header / footer are always dynamic
and never static.

Change-Id: If628d179968c270a63d38cdc3bf6018c01241ac1
Reviewed-on: https://gerrit.libreoffice.org/992
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

halt corrupted styles with multi-saved xlsx document fdo#55418

Change-Id: I2d53c88ab121edeec2bf53f7fd177a0edcc05a25
Reviewed-on: https://gerrit.libreoffice.org/958
Reviewed-by: Kohei Yoshida <kohei.yoshida@gmail.com>
Tested-by: Kohei Yoshida <kohei.yoshida@gmail.com>

fix copy paste error, fdo#55673

Change-Id: I1e31fe05e00ed7e298642da9c0e35ae834dfa74f

Signed-off-by: Kohei Yoshida <kohei.yoshida@gmail.com>

only delete cond formats if the correct attributes are passed, fdo#56316

Change-Id: If9859b5021c532daacccfaf386e0489c134e4afb

Signed-off-by: Kohei Yoshida <kohei.yoshida@gmail.com>

[mono] install mono before finalizing filelist

it was mistake created by porting mono stuff for LO-3.6

Change-Id: I498f7cb7f6317dabd94798dc94db57adf304e5f1

[mono] fix build in gbuildised scp2

Change-Id: Ieee849e3e3b8bda5674027af9c6c2993b045d71a

[mono] fix build in gbuildised unoil

Change-Id: I9c717b067e067338160a38c88d935187db8cc04d

Correctly handle cell note shifting when immediate row/column is deleted.

Currently, having a note e.g. at D5, and deleting cell D4 and shifting
the cells below upward will remove the note at D5. But the correct behavior
is to shift that note up to D4. This change fixes it.

Change-Id: Ia37f1ce67a003deab424f2b805a2ce333fc10ed4
(cherry picked from commit d3344dd85ee31b195a3709d16e734245e1d0a4b6)

Signed-off-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

Fix incorrect shifting of cell notes upon cell insertion / deletion.

Steps to reproduce:

1) Insert a comment at D5.
2) Move cursor to C4.
3) Right-click and select Insert.
4) Choose shift cells down.
5) The comment gets shifted down but it shouldn't.

The same thing happens when deleting a cell and shifting content.

Change-Id: I5a71845cca6abde6b7c940e152e155da26343cef
Signed-off-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

Resolves: fdo#55367 wrong convertion factor for "tspm"

Change-Id: I80fd050e1bbdc0386c13e0afbeeb404e14b2694b
Reviewed-on: https://gerrit.libreoffice.org/849
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Eike Rathke <erack@redhat.com>
Reviewed-on: https://gerrit.libreoffice.org/855

Resolves: fdo#35991 rowspan is of size sal_Int32, not size long

Change-Id: I347672a2175cf7a0304cf5d6c0f8f3f6afa92cd4
(cherry picked from commit 8bd3bf0ff20fc18a0a6358e36b3f8a7e3b34a2bb)
Reviewed-on: https://gerrit.libreoffice.org/919
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Eike Rathke <erack@redhat.com>

fdo#49517: Revert "fdo#46102: Load Java scripts with class loaders [...]"

This reverts commit dd6c4f4db1d62268d73e09ae52d23f760a967dcc "fdo#46102: Load
Java scripts with class loaders that actually find them."  That commit broke
support for macros embedded in documents (as
new java.net.URL("vnd.sun.star.tdoc:...") throws a MalformedURLExcetpion), and
it looks like that commit was not necessary after all -- or rather that what it
tried to work around must have been some other problem that has been fixed
meanwhile.  "It is unclear to me how the Java script provider shall ever have
found the script jars in the past" indicates that something must have been
fishy, and what I failed to notice back then is that createURL creates
java.net.URL instances with a UCBStreamHandler that does allow to obtain content
from weird-looking URLs.

Anyway, with that reverted, all three following scenarios work on both current
master (towards LO 3.7) and libreoffice-3-6 (towards LO 3.6.4); I haven't yet
come around to test on libreoffice-3-5:

1  Stock macros, "Tools - Macros - Run Macro... - LibreOffice Macros -
HelloWorld", running all of the four "helloworld.bsh", "helloworld.js",
"HelloWorldPyhton", and
"org.libreoffice.example.java_scripts.HelloWorld.printHW".

2  Per-document macros, loading test.odt attached to fdo#49517, then "Tools -
Macros - Run Macro... - test.odt - HelloWorld", running
"org.libreoffice.example.java_scripts.HelloWorld.printHW".

3  Extension macros, installing ScriptDispatch.oxt attached to fdo#46012 as
shared extension, then loading StartScriptDispatch.odt attached to fdo#46012 and
pressing the "Start Java via ScriptProvider" button.

Change-Id: I31cd16b3720ffeb1058722d4d1fdffb773f8a067
(cherry picked from commit 7ea7fb009ddcfb0723e88ba0c5778b5fdbe2b553)
Reviewed-on: https://gerrit.libreoffice.org/922
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

resolved rhbz865058 retard overflow of internal table of number formats

Due to an insane amount of ridiculous "user-defined" number formats the
internal table may overflow during import, stave off and increase the
number of available slots to another arbitrary limit.
(cherry picked from commit 075e9ca0b96f37b3561824c23af60872d6a9d80a)

Conflicts:

svl/inc/svl/zforlist.hxx

Change-Id: I2fb85c83a65d6c5b3183aeb3a3bda88828473a3c
Reviewed-on: https://gerrit.libreoffice.org/951
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>

Remove the correct range, or else maItems would end up with invalid pointer.

nStopPos is non-inclusive, and STL's erase() method also expects a
non-inclusive end position (like any other STL methods do). It's wrong
to -1 here which would end up not erasing the last element containing
a pointer to the deleted cell instance.

Change-Id: Ic09ab4a6bb03d0f56bb854a91bf93a99be867116

Related: fdo#47250 fdo#56430 catch harder

so duden extension doesn't blow us up

Change-Id: Ie6b26670a6809fc07b344660111be44cb3e6a011
Reviewed-on: https://gerrit.libreoffice.org/939
Reviewed-by: Tor Lillqvist <tml@iki.fi>
Tested-by: Tor Lillqvist <tml@iki.fi>

Related: fdo#51247 bandaid to not crash

Change-Id: I1f98a6a1e5ede6fcd9a3570788969edcb251384b
(cherry picked from commit 63f00d37a20169743d9709977b382c379560ef9e)
Reviewed-on: https://gerrit.libreoffice.org/918
Reviewed-by: Michael Stahl <mstahl@redhat.com>
Tested-by: Michael Stahl <mstahl@redhat.com>

Resolves: fdo#47291 polypoly[line|gon] points are signed

regression from f6a34255af1339cd7132b7527dc0c10c10d38249

Change-Id: Iabfaf92629cd4d53ab7af5f3e3013eb81bb8104d
(cherry picked from commit 2536fae7b8565b5dd9f09bb3dc015576fafe4031)

Signed-off-by: Michael Stahl <mstahl@redhat.com>

Resolves: fdo#50285 merging props unsafe with intermediate charstyle prop

Its a nice idea to merge properties of the same type and value if one starts
and the same place the last one ended. But character styles are properties as
well, so if we have character-style+superscript on range a, and
character-style+superscript on range b and merge these so that we end up as...

character-style on range a, superscript on range a+b, and character-style on
range b then that clearly gives the wrong result if applied in that order.

So its only safe to merge if there are no intermediate properties that can
affect the merge candidates.

A regression from b3cee382f449aa69213dc21f7b1ba6a5356d2865

Change-Id: I541563d11265426736b840de068922eef8d45573
(cherry picked from commit 08ffb7bc5ec4472126762f4cb9677349b61122f6)

Signed-off-by: Michael Stahl <mstahl@redhat.com>