If you build with --disable-mergelibs and then enable
--enable-mergelibs the existing library has its timestamp updated
but it retains its outdated contents.

Change-Id: I46899bcf3fe4f360fb12e8de8842c6ac12c6bf75

Change-Id: Ie85ba53c59eda5586b70527a518e3f80223fc49e

This has bitten me more than once when working on Library_merged.

Change-Id: I829f5d010b6df2935fd191bdec9bac04bb003f16

Change-Id: I69fbf910cf2eb579cc8bf7dc76c69432b810b194

Change-Id: Ic2a2cf04a691f628e862e81579d3d55d261fc492

Reported by: Du Jing
Patch by: zjchen
Review by: Andre Fischer

Reported by: Li Feng Wang
Patch by: Ma Bingbing
Reviewed by: Arming Le Grand and Andre Fischer

Reported by: Yan Ji
Patch by: Lei De Bin
Review by: Chen Zuo Jun

Reported by: liu ping tan
Patch by: Li Jian Yuan
Review by: Lei De Bin

Reported by: Du Jing
Patch by: Jianyuan Li
Review by: sunying

Patch by: JianYuan Li
Review by: alg

Reported by: Yan Ji
Patch by: Chen Zuo Jun
Review by: Lei De Bin

Patch by: Wang Zhe
Review by: Andre Fischer

* Corrected spelling errors in comments
* In accpara.cxx, SwAccessibleParagraph::_getRunAttributesImpl had the
  if statement commented out, but the scope block remained... minor in
  the grand scheme of things, but messy code
* Some unused commented out code removed

Change-Id: Id8b29c4051c7c09a1cc555996f9e56fd4f0f3611
Reviewed-on: https://gerrit.libreoffice.org/3135Tested-by: Thomas Arnhold <thomas@arnhold.org>
Reviewed-by: Thomas Arnhold <thomas@arnhold.org>

Change-Id: I39cea3e90f0ac57168e1cd2d4a9af9199e1416fb

Change-Id: I247a261bb96b0c437a43922070777892cdcbc1f6

Project: dictionaries  52f4aa133ba3d7703c4982a21e567fd92abcbfd0

Serbian dictionary updated to 2013.03.30

Change-Id: Ia3b99deff58bd5f4198f1fd57ea3a540f15d2708
Reviewed-on: https://gerrit.libreoffice.org/3120Reviewed-by: Andras Timar <atimar@suse.com>
Tested-by: Andras Timar <atimar@suse.com>

Project: dictionaries  b128def79cc0520515ea99db4fb9f8cc3d95f54d

Lightproof and Hunspell updates for pt-BR

VERO 3.1 beta 2

Change-Id: Iaf871e4ad7de5c51ed55674b742b8f2b20070001
Reviewed-on: https://gerrit.libreoffice.org/3130Reviewed-by: Andras Timar <atimar@suse.com>
Tested-by: Andras Timar <atimar@suse.com>

Change-Id: I61323685b33817a2641713455bac1a079687cb4d
Reviewed-on: https://gerrit.libreoffice.org/3131Reviewed-by: Andras Timar <atimar@suse.com>
Tested-by: Andras Timar <atimar@suse.com>

Change-Id: I54881476ae296bc95df16e233ff4ca5800477254

On the iOS Simulator it contains spaces, as in
/Users/tml/Library/Application Support/iPhone Simulator/6.1/Applications/9A6DFE86-77AF-4B78-8FFB-93FCA6C38EE1/LibreOffice.app

Change-Id: I90c76b909901c881aa51482880b1120fea19b99b

Change-Id: I01c9c26c6d282729101386266db898d759450a41

Change-Id: I551cb2422c8857bb847bd4b0c2224eaa1c024084

Change-Id: I8e27f0cb4a9605a97d8120f9f4697e7f939135c1

Change-Id: I176088d3d33f02a1721e4027571a1f22dc6bd986
Reviewed-on: https://gerrit.libreoffice.org/3116Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Eike Rathke <erack@redhat.com>

Change-Id: I66b85365d1c59f802253b8abdb1e04e25950a09b
Reviewed-on: https://gerrit.libreoffice.org/3098Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Eike Rathke <erack@redhat.com>

Change-Id: Ie2159d5eb8caf3d8f13d9c1ec49cc2bb6a0d94eb

Change-Id: I4824395653052486956572cedf3b56b94f9df39e

Change-Id: I23fc6bf62e74f192f77c74ed48d5276737375f3c

Change-Id: I2fb3bd758d7b903cd148f847c21c0eefd6f091e5

Change-Id: Ic6771c4ae12686580301e114fa5aec7f7d4dc9a8

Change-Id: I39c301a98854d19e849a9e06ffac0657408456fe

_PC_NAME_MAX is a symbolic constant meant to be used as a pathconf()
argument, not a buffer length.

See also 2cd4e55b

Change-Id: I94d19bd6f9f0ae501329d6a1edecad630f44b356

_PC_NAME_MAX is a symbolic constant meant to be used as a pathconf()
argument, not a buffer length.

This piece of code was even "honored" in
http://womble.decadent.org.uk/readdir_r-advisory.html
"The code that enumerates fonts and plugins in the appropriate
directories uses a stack buffer of type long[sizeof(struct dirent) +
_PC_NAME_MAX + 1]. I can only assume this is the result of a programmer
cutting his crack with aluminium filings."

Thanks to Heiner Rechtien for pointing this out.

Note: declaring the buffer of type long instead of char looks weird but
probably has the reason that it is casted to struct dirent* for the
readdir_r() call that might not work on architectures that forbid
arbitrary alignment on non-even boundaries.

Change-Id: Ie959b8905d1328c6fcaefbf6f6c915cc430e85b6

Change-Id: I067d52236b189e09955d7aa3ce2fbd25b0d330bd

Change-Id: I2ff581a72aa66bff8943d31499013b31e360333b

Change-Id: I4196e763e8374d1f50700c20362245224acb9e13

Change-Id: I1871e5e3a319b2d5177ae4bd470046c022c14eb7
Reviewed-on: https://gerrit.libreoffice.org/3121Reviewed-by: David Tardon <dtardon@redhat.com>
Tested-by: David Tardon <dtardon@redhat.com>