I /think/ a set-but-null LD_LIBRARY_PATH (Linux), DYLD_LIBRARY_PATH (macOS) are
each treated like an unset one.

Change-Id: If60698d6f21a16456dac9c3d4f4e1aa3561e1a65

harfbuzz needs freetype, freetype wants harfbuzz but can live
without it

Change-Id: I4b5dbd937ccfc9af46c86be5ae1d3f8eeb14e86b

Change-Id: I904424251a7d6a84a97e61d18c1a731c66866fae
Reviewed-on: https://gerrit.libreoffice.org/33917Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Change-Id: I9c23dcdc62973de657bf5a3a9a6dad0f0c89ee73
Reviewed-on: https://gerrit.libreoffice.org/33896Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Zdenek Crhonek <zcrhonek@gmail.com>

Change-Id: Ifb6609a4315851d4eae2c7017c25f0ad7b833123
Reviewed-on: https://gerrit.libreoffice.org/33916Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Akshay Deep <akshaydeepiitr@gmail.com>

Change-Id: I408c48ce0b28534b232b5d401ed13f7af76ed2ed
Reviewed-on: https://gerrit.libreoffice.org/33915Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Change-Id: I8c827db4c16a4a169dd710787bf9beb123ddc7b6

its there since initial checkin and I rather feel things
have moved on since then

Change-Id: Iea3aa1c5a1637cecdd18539987ef8b84db095e9a

Change-Id: Icef2a5ae31973c09620112dba062525d42bdd227

only one of 164 calls used the ctor with a restricting range.
handle this special case.
remove nFrom and nTo members.
use operator[] on range pointer to highlight we access lower [0]
and upper [1] range bound.
remove dtor.

Change-Id: I87b28502515e45bb190425eb88aa187ac0b5b2d2
Reviewed-on: https://gerrit.libreoffice.org/33911Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Jochen Nitschke <j.nitschke+logerrit@ok.de>

reverts parts of e6a9dc22
we want globbing here

Change-Id: Ibe4672f41b0fe6a75d1aca898cc55414e138f6ef
Reviewed-on: https://gerrit.libreoffice.org/33913Reviewed-by: Jochen Nitschke <j.nitschke+logerrit@ok.de>
Tested-by: Jochen Nitschke <j.nitschke+logerrit@ok.de>

Change-Id: I91ab4b1fc347a81d8cbd7914a578a922592f923c
Reviewed-on: https://gerrit.libreoffice.org/33901Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Zdenek Crhonek <zcrhonek@gmail.com>

double ".ccx" for each source files in the .filters in the
testVS2013Ide.
this patch fix that.

Change-Id: Ic06800c260800f72254038e30119db40fe048037
Reviewed-on: https://gerrit.libreoffice.org/33892Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: jan iversen <jani@documentfoundation.org>

this function generates a vcxproj, for the reverse engineering
i generate a vcxproj from VS2013 that:
-the compiler works good

i do the code that generates it
the code miss that the link to lib that i left commented

Change-Id: If98eb5645c9e1060a55701abf5b96db85eca6ce5
Reviewed-on: https://gerrit.libreoffice.org/33893Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: jan iversen <jani@documentfoundation.org>

Following up on the commit a3a87826

Change-Id: I4f525ecb030044fc9a330ad8964dbeca6ef35ee3
Reviewed-on: https://gerrit.libreoffice.org/33908Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: jan iversen <jani@documentfoundation.org>

This prevents the crash. It might be related to
gerrit #/c/33654

Change-Id: I92907c7874964af722b69ed8afd9249e110b3401
Reviewed-on: https://gerrit.libreoffice.org/33907Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: jan iversen <jani@documentfoundation.org>

Use $(..) instead of `..`,
use && and || instead of -a and -o,
double quote to prevent word splitting,
handle cd failure.

Change-Id: I860891323a81ac4a46c868ab028df1e1f837c115
Reviewed-on: https://gerrit.libreoffice.org/33897Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: jan iversen <jani@documentfoundation.org>

Change-Id: I74456399bde6075c9b3813d6c0ffe98fe1afbe3a
Reviewed-on: https://gerrit.libreoffice.org/33905Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>

Change-Id: If4d02161bd8a37100d7403c05ea0840a3d2d0cf8

... so let's finally do it right, and count in double precision directly.

Change-Id: Ic49cc42d9e61a1b9fa9853bfe93d2681a6be88b2

Change-Id: I1f2a4e249274a2dfd74da11553d7bb9fb35b5d53

Change-Id: Ic54695e86b4b462419fa7d5ded7b1ddb19ee8ed5
Reviewed-on: https://gerrit.libreoffice.org/33904Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
Tested-by: Julien Nabet <serval2412@yahoo.fr>

Change-Id: I3efb3630fbb8a605774796fd83514404c1fa35ee
Reviewed-on: https://gerrit.libreoffice.org/33903Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
Tested-by: Julien Nabet <serval2412@yahoo.fr>

Change-Id: If9d73319988b9d6ac5fcdee2ec815738320a62a0

Change-Id: Ic13529f4bc9ac49059a59067c50d5f2c0f569822
Reviewed-on: https://gerrit.libreoffice.org/33902Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
Tested-by: Julien Nabet <serval2412@yahoo.fr>

Change-Id: I36f443c266670bd26aa6b813075e18653f372f52
Reviewed-on: https://gerrit.libreoffice.org/33900Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
Tested-by: Julien Nabet <serval2412@yahoo.fr>

Change-Id: Icd5a2d4156ace852f522df162c4a8da1cf2af5c2
Reviewed-on: https://gerrit.libreoffice.org/33899Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
Tested-by: Julien Nabet <serval2412@yahoo.fr>

Change-Id: I17bd9ce74d95beaa230276bec40b739b05b81cf2
Reviewed-on: https://gerrit.libreoffice.org/33898Reviewed-by: Aleksandar Stefanović <theonewithideas@gmail.com>
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
Tested-by: Julien Nabet <serval2412@yahoo.fr>

This asserts because MarkManager::m_vAllMarks is not sorted.

info:sw.core:32273:1:sw/source/core/doc/docbm.cxx:286: 1493,39 1493,39 N2sw4mark8BookmarkE _Source_Device_Configuration
info:sw.core:32273:1:sw/source/core/doc/docbm.cxx:286: 1493,39 1493,39 N2sw4mark8BookmarkE _Deploying_the_Connector
info:sw.core:32273:1:sw/source/core/doc/docbm.cxx:286: 1493,39 1493,39 N2sw4mark8BookmarkE _Setting_up_Connection
info:sw.core:32273:1:sw/source/core/doc/docbm.cxx:286: 1493,39 1493,39 N2sw4mark8BookmarkE _Testing_the_Audit
info:sw.core:32273:1:sw/source/core/doc/docbm.cxx:286: 1493,0 1493,0 N2sw4mark23CrossRefHeadingBookmarkE __RefHeading___Toc270687978

This happens while called from SwRangeRedline::MoveFromSection; the
target paragraph has some normal bookmarks and a CrossRefBookmark
all at index 0 before, but the move messes up the sorting.

The reason is that SwIndexReg::Update() has a special case HACK to avoid
adjusting the content index of CrossRefBookmark away from 0.

Prevent the problem by tweaking the bookmark sort function to always
sort CrossRefBookmark before other marks in the paragraph.

Change-Id: I1373337a9c7e6760fdbe5b634a14c63428a2e73c

Change-Id: If9c151faefa3d109af2df7933a275c667880fb75

This reverts commit 27ca6387, at least together
with f7962f59 "sc rowlimit: Make the cursor
better fit the rendering" it makes CppunitTest_desktop_lib fail.

This reverts commit f7962f59; at least together
with 27ca6387 "sc tiled rendering: We keep
increasing the precision here..." it makes CppunitTest_desktop_lib fail.

Change-Id: Ieb3f114feea3b9e9c597e0bd16ed80734dcd5ec2
Reviewed-on: https://gerrit.libreoffice.org/33865Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Samuel Mehrbrodt <Samuel.Mehrbrodt@cib.de>

commit eca5ea9f
"make the AbstractDialog stuff extend from VclReferenceBase"
introduced this regression.

Fix up the asynchronous delete to work again - necessary because
the delete is triggered from deep inside the object it wants to
delete.

Change-Id: I18d5a899fbb2da772552ebba7fe6a8e810314010
Reviewed-on: https://gerrit.libreoffice.org/33881Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>

Change-Id: If4d02161bd8a37100d7403c05ea0840a3d2d0cf8

... so let's finally do it right, and count in double precision directly.

Change-Id: Ic49cc42d9e61a1b9fa9853bfe93d2681a6be88b2

in particular, libgpg-error and libassuan
This only downloads and unpacks the tarball. Building them needs
some work still

Change-Id: I562fd01571929ddfb47a319038f88ea8dbfb4bdd
Reviewed-on: https://gerrit.libreoffice.org/33712Reviewed-by: Siegmund Gorr <siegmund.gorr@cib.de>
Tested-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>
Reviewed-by: Thorsten Behrens <Thorsten.Behrens@CIB.de>

Change-Id: Ie9e6d4aa9dba97d45bf7b8afa2347d34e1651a60
Reviewed-on: https://gerrit.libreoffice.org/33871Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>

Change-Id: Iaedabfa4c4242f5a37f818572fddfbcca88d5b9a
Reviewed-on: https://gerrit.libreoffice.org/33849Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Muhammet Kara <muhammet.kara@pardus.org.tr>
Tested-by: Muhammet Kara <muhammet.kara@pardus.org.tr>

This forces writing a comparision of two variables in the
CPPUNIT_ASSERT_EQUALS(expected, actual) form, rather than
CPPUNIT_ASSERT(actual == expected). The benefit is when the test fail,
it's more clear what are these two values, rather than just seeing that
they are not equal.

In the relatively rare case when the types are not streamable, the
plugin can be silenced using CPPUNIT_ASSERT(bool(it == foo.end())). But
please always first consider making the type streamable instead. See
commit 0e1b831b
(CppunitTest_vcl_complextext: fix loplugin:cppunitassertequals warnings,
2016-12-16) for an example.

Change-Id: I564fb0724836b6dbbea9262b53ad646e285a339c