Change-Id: I86e2095bee82a2a6a28b8750e5a328c4d395e2aa

Change-Id: Ib765ce5a61916b0598ebb42cb9445a5f09c8d37f

Change-Id: I60678dfd1bd0d2e4754dbbe6ba053e90d5870780

Change-Id: I64fdb27a7f83f6417a9cd67ed5d2c44072ec7f2e

Change-Id: I3d61c1c64f7e781c032a77d5d35998e953bf5c0c

Change-Id: Ie6867e04156f40cacee275d5bbf3d63ac4965b30

Change-Id: I384a5e60f4b7b2f479c89ef97630519059ab720f

Change-Id: Idfd6368eba8e638cf9e6adf998b19ccd1f5e876c

Change-Id: Iea68c4e466dd8e1cb7164959a680e5e52542166b

Change-Id: Id775680d6b2aec1fd84981a72a94fe4934bef003
Reviewed-on: https://gerrit.libreoffice.org/15632Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Andras Timar <andras.timar@collabora.com>

Change-Id: I0de1cfaa0df460c7bc8e49752b31056f95c30053
Reviewed-on: https://gerrit.libreoffice.org/15628Reviewed-by: Andras Timar <andras.timar@collabora.com>
Tested-by: Andras Timar <andras.timar@collabora.com>

Change-Id: Iee9c1bd37664188e4dc10b929eb72b6370001417

since

Start slideshow with the next visible slide

commit 45e4a5ef
Author: Katarina Behrens <bubli@bubli.org>
Date:   Sat Mar 23 20:32:40 2013 +0100

Change-Id: Ib04aac2b223a38dbf46ee8cea3194e116e435eb6

Change-Id: I07aaf57291239e30b1600bd1061821b4759b8a0b

Change-Id: I236d7d951bae8c40d2caaa5ecd6dfd669f8ea982

Determine the effective data area before all operations, so obtaining
uniform row heights, the actual sort and setting dirty and setting row
heights after sort is done only on the trimmed area.

Also make sorting leading empty columns to the end work again if no row
headers were given.

Change-Id: If2b6a15ca69e0db2ca71a888a134f7441b04cc27

Change-Id: I7e9edae135bd8f3b958c2013fe6d944ab3de2100

Change-Id: Icb4273515082bffaeb9d8f3ebcec8d76fa4192ef

Change-Id: Ibdf762b0d397f798372d9bf882aa82a6e5fd0229

Change-Id: Icc4a3d3661c3d40f9e71215bdc875c6a25f40ac1

Change-Id: Iea5bb0504518b380c3fe7c238d376dc26be8a7ef

Zero or negative count is unhealthy..

Change-Id: I4cce6c896e73e8e964518cbe4a29eb03ed481251

... in case the header is the only row.

Change-Id: I5e6046007a8d668f9834e108aaf8af0072629fc8

This reverts commit 4c07124c.
It looks like these strings are used via an iteration in docundo.cxx
So, restore them.

Change-Id: I1b0428de3d56b04908284c94b47dc6021375cc8e
Reviewed-on: https://gerrit.libreoffice.org/15629Reviewed-by: Joren De Cuyper <jorendc@libreoffice.org>
Tested-by: Joren De Cuyper <jorendc@libreoffice.org>

Change-Id: I487aad5217cb685ece971717445a68b4fb3d2b3a

Change-Id: I3ec59f2a736d230eeb9f6e5db49af16f3e570afd

Change-Id: I99c6677bcb1a491ede67322b21f6ca051a2e924d

Change-Id: Ia605aeb2a9936f3c60db179420805a3e986fd64b

Change-Id: I4d60400f4cc3f34529d632e52b676007f87ff646

dbaccess::ODatabaseDocument::loadFromStorage() is still not implemented,
though.

Change-Id: I1be11f2a2274f67dd0a11f96a718394d5f72605f

Change-Id: Ib2cb804fa42ac3cebd309de4ba3cdbbce9a1763d

Change-Id: I6a8689408141035aaf5cc87838c65eb4b9bc71d3

Change-Id: I3e78e3104af97eed91c102e9ca5546e0785c5f8c

Change-Id: I674eadb84fc870031244a61f5c07d2cdf18a8dd1

With 64-bit MSVC, sizeof(long) is 4 but sizeof(void*) is 8, so this
would select sal_uInt64 but SAL_MAX_UINT32.

This should make sizeof(sal_Size) the same as sizeof(size_t) on all
supported platforms, but still sal_Size maps to different integer
type (long vs. int) than size_t on 32-bit.

Change-Id: I638aac6b502e624ed6b01f5921e20bc40f42480c

Change-Id: I6d41aecfb1b41bf3d9e8fac6d63ae9874283e3a3

follow-up to 8568c722 "tdf#66232: Cannot open
files by COM2"

Change-Id: I5a0858dd6f48136f9f07b0d04c5634a6173dd154

Change-Id: I41c7053296c634c0fed5b31f2e080c3eaf59bbba

Change-Id: I1f5272cdf05b36d1c6958159a29c04ed7374592d

Turns out that at least with the glibc version I have on openSUSE 12.3,
whether <stdint.h> defines SIZE_MAX and friends or not depends on whether
__STDC_LIMIT_MACROS is defined, and that gets defined (in our case) only when
<stdint.h> gets included through some boost include file, or through
<cstdint>. In the case of this compilation unit, apparently <stdint.h> gets
included directly somewhere before SIZE_MAX is used in sw/inc/docary.hxx.

Possibly a better fix would be to make sure we include <cstdint> everywhere
and not <stdint.h>. That wouldn't help in case it is through some 3rd-party C
library include file that the <stdint.h> gets included, though.

This reverts commit 8e6527a9

Change-Id: I200985ee70eaf90fc3fa9341fdf9d657750e49ee