Change-Id: I1aad3e40ad61a1397b619bdc6e0f578e2c9c141e

Change-Id: Ice59382e13ca91f37f7140ee5bc5b5841a61acce

Project: dictionaries  59ff183f473301efcdf7d5b85f3f5e2769d030ac

Revert Spanish dictionary’s description to allow for its localization

Thanks to András for noticing.

While adding check that the temp file move succeeds, overlooked that the
file handle is actually leaked, which means the move will fail on WNT
because the file is already open.

(regression from ef490df2)

Change-Id: I779ccde79871930fef3bad7e17b633fb62e36a46

Besides, wtf is it using -fast by default so the binaries don't run on any
system other than the one that did the build?

Change-Id: I5adc26fb6370a5292721df3167afc5485c15cb0a

HTMLMODE_SOME_STYLES is enabled if we are in HTML mode (it is
enabled for all HTML modes). These checks for conditionally disabling
some table background controls are not needed anymore.

Change-Id: I10fb73a67ac4bba250da4cea79ff6bae427a1613
Reviewed-on: https://gerrit.libreoffice.org/15264Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Change-Id: Ic8274e7d060f315d7891f8f5cf576f834ca89117

Change-Id: I533d603ac1f073336eda99de57892ff5e12132d4
Reviewed-on: https://gerrit.libreoffice.org/15288Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Michael Stahl <mstahl@redhat.com>

Change-Id: I1460131d25212266b910d76a7713d1e973f7c415

Change-Id: Ib30a0e3d56ca9ebd07ee55e0aee9b13c0474a060

Change-Id: I81dcd2c9f13a93e15689fed6491883e7d828ca99

Change-Id: Ia8c331f33bc81e7bdabeea649a47c73930dae206

Dxf font export needs a lot of special handling to detect which parts
are set in the itemset and only export these. The old code only checked
for font name, font family and text encoding and exported the whole font
as dxf font in that case. The new approach just exports the necessary
bits.

Change-Id: Ic9a9c547723a65f26ebb6c88040a51f04f3cc42b

Change-Id: Ic9e2a947805f4aad987160c52ffb5ecf4a7cb54b

Change-Id: I3158762e7da10eacaf2616a7d868531ad7fa7a51

cut and paste error I guess from the block above it

Change-Id: I121e1d1a9f08ea3731bb7f65c385f264b3c2d2f4

This fixes a fallout from bf8b86e9.

Change-Id: Ie2d2a1329df0ec36f21a07938af142c32c74277a

The bugdoc has:

<style:page-layout-properties fo:background-color="#007f7f" >
    <style:background-image
     xlink:href="Pictures/2000003900002FA3000055F455EC9F63.svm"/>
</style:page-layout-properties>

In this case it's clearly better to prefer the Graphics.

The old code could actually store both in the model but the UI only
allows setting either Color or Graphic, so this should be good enough.

(regression from 7d9bb549)

Change-Id: Id33231ad90815d9791f90289fb0b55455d334c14

(regression from 7d9bb549)

Change-Id: I67ec4516ff431efd47451ff07fc261ba32e80385

Add the fill style properties so background images etc. are not lost.

(regression from 7d9bb549)

Change-Id: I517350eec2f32e4b4bb80427666aad23a31d666a

Change-Id: I7e5f28f6cc8010ac233d58cfa6a2158e3f7be90b

Somehow this gets called in a different order than in 4.3; with the
bugdoc from tdf#89478 i get this assert on master because of the
uninitizlized / LONG_MAX.

Change-Id: Iacce21571246c96276bd3c126f982282df7fc491

see also on export of ooo103014-1.odt to doc and many more

easiest thing is to move the restoration of the redline mode upwards for
all filters to the central SwWriter::Write after all pams have been
destroyed.

Change-Id: Ieb747ad2ed465d3fd3de81db0ef820073bc3ab55

Change-Id: Ib43b844ead1720df3c93b2d2371d310af400fb29

it breaks sorting of conditional formatting manager table.
This reverts part of commit f92ab4da

Change-Id: I22d4b939cf888451bff6500fdf0d6b900903deb1

Removes dynamic loading logic described in tdf#84315, similar thing
removed in swdbtoolsclient.

Change-Id: I8762102a7263e6933354c2ff6f9978929b760f6e
Reviewed-on: https://gerrit.libreoffice.org/15147Reviewed-by: David Tardon <dtardon@redhat.com>
Tested-by: David Tardon <dtardon@redhat.com>

Change-Id: I3c1163f9757a4592692bb8a920d1dc679bf53699

Change-Id: I5bd6965bde1a5dcbe26abc197a67e09c06a2bb21

Change-Id: I91478cf0a9b128f96dab5be67d018af46d75bd4f

Change-Id: I80af505d0dd9a71eda5d640dd336a034288a574d

word has a limit of MAXTABLECELLS cells in a row, we were clipping
the rowspan container to that max, but not the cell container

Change-Id: I5274dc7febd47e6c8c8e5337f638972b65239922

Change-Id: Ib2e969ca54da16b29ec20338148ee51810cb3c47

Change-Id: Ie60c09d7b7c626a0a7624e4c833d72eb5d0d5965

Change-Id: Ica21cefb705f4140b9f298e09be04b4950407b78

Change-Id: I61f7eaaad92b2a7b543fbb8f16f5b695700e8073

Change-Id: If36ec5ed78a3e0b7a9037eaa1903772f1d228bce

Change-Id: Ifdf3ef84cd886b338536629fb36ebba694916d6a

Change-Id: I228c0bca3133e94b28db2540dbd41e8b4fd1eee9

Setting bInSelection in MouseButtonUp() feels wrong, and against of its other
handling elsewhere.  Also it breaks the selection handling in the tiled
rendering case - on further redraw, selections may disappear.

But it has been so since the initial check-in of the OOo code, so I have no
idea what corner case I might break by deleting this, so let's guard the
behavior by isTiledReneding() for now.

Change-Id: I7f7b237f9f0f427e317266e05481bac439c3c8ad

Normally, the text overflowing from other cells is completely ignored by Calc,
and the user always works with the underlying cells.

On Android / mobile, it is though more natural to be able to select the text
directly; so implement a compromise:

* tap places the text cursor, so that the user can write into the
  cells hidden by the text too
* long-tap selects the word in the text, even if the text 'just' overflows
  from another cell

Change-Id: Ibe8666301ff1df0414c0206c1f3336842485433b