Change-Id: Iaf3d6a0423c5f11dda1e623dd730af01dbd6551c
Reviewed-on: https://gerrit.libreoffice.org/57284
Tested-by: Jenkins
Reviewed-by: Armin Le Grand <Armin.Le.Grand@cib.de>

Instead of failing with "is not an absolute URL" error.

For example in the HYPERLINK() spreadsheet function where Excel
supports it, but also anywhere else that calls OpenURL(). A
relative reference is relative to the current document, or for yet
unsaved documents or if no object shell is available relative to
the work path, as usual.

Additional benefit is that "\\" UNC path names are accepted as
smb:// Samba shares and DOS \ path name separators resolved for a
proper file:// URI.

Users are asking for this, found no related tdf# bug/RFE. See
https://ask.libreoffice.org/en/question/160280/relative-and-absolute-hyperlinks/
and other linked q&a therein.

Change-Id: Ib314b71d68fbe1793ec614cbf7c0c058fad14a10
Reviewed-on: https://gerrit.libreoffice.org/57277Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Jenkins

The attribute name of the numbering format code  11 is "japaneseCounting"
[1]. Although it seemd like a Japanese counting guessing from its name,
it's actually use the same characters used by Chinese lower numbers [2].

Ref:
[1]https://msdn.microsoft.com/en-us/library/dd923798(v=office.12).aspx
[2]https://msdn.microsoft.com/en-us/library/dd772856(v=office.12).aspx

Change-Id: I241be1e5e118197207cbc51334272b532e23c25f
Reviewed-on: https://gerrit.libreoffice.org/57138
Tested-by: Jenkins
Reviewed-by: Mark Hung <marklh9@gmail.com>

Move CjkNumberedListTestHelper to swmodeltestbase.hxx to get
rid of copy-pasted code, and rename it to getNumberingTypeOfParagraph.

Change-Id: I36f14be30bcd9a9d22d1d5071220a13c02602a22
Reviewed-on: https://gerrit.libreoffice.org/57137
Tested-by: Jenkins
Reviewed-by: Mark Hung <marklh9@gmail.com>

Change-Id: Ic5a78d3f650bff82ce8d86a62ecfc025ca2062e0
Reviewed-on: https://gerrit.libreoffice.org/57276
Tested-by: Jenkins
Reviewed-by: Justin Luth <justin_luth@sil.org>

Change-Id: Iddb48d2cde206824f7721c5f4374166cc4979ad9
Reviewed-on: https://gerrit.libreoffice.org/57272
Tested-by: Jenkins
Reviewed-by: andreas_kainz <kainz.a@gmail.com>

TabOverMargin in general is about allowing the cursor to jump over a
margin if there is an explicit tab stop there.

A corner-case is what to do when there is enough content so a line break
is necessary for the characters after the tab portion. Allow using the
area up to the edge of the whole text frame (i.e. over the tab
position), this matches what Word does.

Change-Id: Ie86edf030d54fba556eee26e7ea563fb8d4fbee4
Reviewed-on: https://gerrit.libreoffice.org/57275Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Jenkins

7c6ca00e broke the Mac build, because
the files from infoplist Package have double "/" because of the empty
path in the call to gb_Package_add_files.

Change-Id: I3a72e8de0a8f2256b068a491231aaaa3d3b00b6e

Change-Id: I204250a02ac88cc36267b79ef1d70cd361230752
Reviewed-on: https://gerrit.libreoffice.org/57245
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>

No point in creating funny MSIs that don't work.

(Thanks to bubli for the help with perl)

Change-Id: I54aa39ed282e31270de69566656d33b803a73802
Reviewed-on: https://gerrit.libreoffice.org/57266
Tested-by: Jenkins
Reviewed-by: Michael Stahl <Michael.Stahl@cib.de>

this gives another 2% perf

Change-Id: Ia2983339f3f11daef37c48044904c8037a7a0bf6
Reviewed-on: https://gerrit.libreoffice.org/57265Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
Tested-by: Noel Grandin <noel.grandin@collabora.co.uk>

There seems to be some confusion here. available() is actually the
number of bytes that can be read without blocking, but most
implementations seems to be just returning the number of bytes remaining
in the stream. Since we're doing that, let's do it properly.

(*) some of them were just casting, instead of clamping, which will
return wrong values sometimes.
(*) FileStreamWrapper_Impl/OInputStreamWrapper/OTempFileService were
doing unnecessary work, instead of just asking the underlying SvStream
for it's remaining size

Change-Id: I3ef26e0363e989ed3e00be0fdb993e1cdeb7819f
Reviewed-on: https://gerrit.libreoffice.org/57264
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>

* Update helpcontent2 from branch 'master'
  - Fix more typos in GIMMICKS name here and there
    
    Change-Id: Iae95e2aaa42f273effa3b7576da7e6c45495c5c6
    Reviewed-on: https://gerrit.libreoffice.org/57273
    Tested-by: Jenkins
Reviewed-by: Olivier Hallot <olivier.hallot@libreoffice.org>
    

(*) in GetTokenWithPrefix, use rtl_str_reverseCompare_WithLength
    faster than strncmp
(*) No need for NamespaceDefine to be stored via std::shared_tr

Change-Id: Ibd262a3f4f5a0f518ec6abe1fb19e7803f78fe8b
Reviewed-on: https://gerrit.libreoffice.org/57261Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
Tested-by: Noel Grandin <noel.grandin@collabora.co.uk>

Change-Id: I9e8ea29893b372daa27262b2224481c67b4e4482
Reviewed-on: https://gerrit.libreoffice.org/57258
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>

our nice threaded parser was not kicking in because it checks the size
of the stream with available(), and this available() implementation was
just returning 0.
This makes for a 5-10% performance improvement

Change-Id: I6516862c9c63136b54ca825dd98e35e6d87d4227
Reviewed-on: https://gerrit.libreoffice.org/57263
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>

* Update helpcontent2 from branch 'master'
  - Typo in GIMMICKS Basic library name
    
    Change-Id: I15a491ae867f16c13faf1a102ec3b11a6dcfc638
    Reviewed-on: https://gerrit.libreoffice.org/57252
    Tested-by: Jenkins
Reviewed-by: Olivier Hallot <olivier.hallot@libreoffice.org>
    

Member function was defined, but not declared.

Change-Id: Ia7aeae264be9e833f5ba1714df0bd7f976b21437

If a style contains tab definitions, then the paragraph inherits
these. They are added to any tabs defined at the paragraph
level. Unwanted inherited tabs must be explicitly removed.

(TODO: LO ought to be inheriting from ALL parents, so the same
logic ought to apply to a style's parent tabs, but currently
LO does not import that way. So the proof unit test looks
different in MSO compared to LO.)

Change-Id: Ida8ed2792482655d512c753fdff8d02062d895a8
Reviewed-on: https://gerrit.libreoffice.org/57255
Tested-by: Jenkins
Reviewed-by: Justin Luth <justin_luth@sil.org>

as the queue is modified only by either appending or clear().

Change-Id: I6866c047eaf7149221c77ff4f1f3759dbf10492e
Reviewed-on: https://gerrit.libreoffice.org/57182
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>

Regression from 13596e70 (Don't pass ODF
version, just retrieve it directly, 2018-05-17).

Change-Id: Ia031e597c610000db2820920c3dc1b58430dcc12
Reviewed-on: https://gerrit.libreoffice.org/57246
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>

Change-Id: I1bf2a55bc9fe7f70c5cc8ff51062b9c51cf70f80
Reviewed-on: https://gerrit.libreoffice.org/57248
Tested-by: Jenkins
Reviewed-by: andreas_kainz <kainz.a@gmail.com>

Change-Id: I97fdb9ffb506a005dac08f7c78a3fa4383098b0b
Reviewed-on: https://gerrit.libreoffice.org/57169
Tested-by: Jenkins
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>

When collecting font feature description do prioritize Graphite
collected descriptions over OT. This is important for descriptions
for ssxx and cvxx only contain general description and not what
exactly they do.

When collecting for Graphite use the UI language for descriptions.

Fix gathering the descriptions with graphite to include all of
the descriptions available.

Change-Id: I6b3b92eb160fc3fd70214edf0d37095865f9c563
Reviewed-on: https://gerrit.libreoffice.org/57250
Tested-by: Jenkins
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>

Change-Id: I35061f62244ea40ce66bbb0ba09edf9b45d62742
Reviewed-on: https://gerrit.libreoffice.org/57244
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>

Change-Id: I135c0a333633bbdc362fb80699c6d009a80b23a5
Reviewed-on: https://gerrit.libreoffice.org/57243
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>

Change-Id: I12635d45555bc59d14134b8eef3e093f3fd4e6a0
Reviewed-on: https://gerrit.libreoffice.org/57242
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>

Preparing for additional URL handling.

Change-Id: I7897a2c0771b42c02911d8a979c0064410ebd792

Unnecessary to init a bunch of stuff if not.

Change-Id: Ica80185370ecc341bda5b1615ff6d765b579f45e

Change-Id: I5bdb2876a871e2b1f4b8294e262705554cb3968c
Reviewed-on: https://gerrit.libreoffice.org/57237
Tested-by: Jenkins
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>

... instead of only top left that is referenced for other
elements.

Change-Id: I718946d7e4327b152e2d9f80712395fd7ab67dee
Reviewed-on: https://gerrit.libreoffice.org/57235
Tested-by: Jenkins
Reviewed-by: Eike Rathke <erack@redhat.com>

Change-Id: I8d5cf48fb37db8e6e5c8058f92b69659838f3471
Reviewed-on: https://gerrit.libreoffice.org/57229
Tested-by: Jenkins
Reviewed-by: Jan-Marek Glogowski <glogow@fbihome.de>

This reverts commit c290f692.

Reverting because

a) makes builds fail with errors from the schema validator in saving unit tests

b) the loext:max-blank-integer-digits is now saved for some number formats where it is not necessary
For example, load sc/qa/unit/data/ods/tdf102370_ExponentWithoutSign.ods
save as .xlsx
reload the .xlsx
save as .ods

(which effectively is what the ScExportTest::testExponentWithoutSignFormatXLSX() unit test does and makes the schema validator stumble)

Change-Id: Iaac7e42b23317e7977876c9b5d545c76a1e2a4d2
Reviewed-on: https://gerrit.libreoffice.org/57240Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Eike Rathke <erack@redhat.com>

Also don't use storeAsURL to store embedded datasource when the document
is being saved using storeToURL.

Change-Id: I69a7ee5ae066e591be5e45c87bcf57dff370decc
Reviewed-on: https://gerrit.libreoffice.org/57178
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>

* Update helpcontent2 from branch 'master'
  - tdf#118571 avoid MSI problems with double / in Package_html*.mk
    
    MSI helpfully converts "help//fuse.js" into "help/help/fuse.js",
    which isn't obvious.  Or helpful, for that matter.
    
    Actually it might be the Perl installer code that messes it up,
    can't quickly check whose fault it is...
    
    Change-Id: I69e9849b533a7fbe54310edb99bda3a879f4f031
    

Missing XML namespace, wrong attribute name, missing escaping for the
bookmark name.

Change-Id: I639c147574e059083646b4c33ede9d478d6c0c38
Reviewed-on: https://gerrit.libreoffice.org/57230Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Jenkins

Change-Id: Ie68bf48b44ab583f0ea8a4a02d330ce95d739c00
Reviewed-on: https://gerrit.libreoffice.org/57213
Tested-by: Jenkins
Reviewed-by: Jens Carl <j.carl43@gmx.de>

commit 0307a627 fixed the
top margin import of the paragraphs of document sections, and
import of first paragraph of the text frames with beforeAutospacing,
but nullified all other paragraph top margins in frames.

Note: there is no visible margin difference in the unit test (extended
by this commit) before and after the fix, because the first paragraph
uses also afterAutospacing, resulting still 14pt paragraph space.
But the tested beforeAutospacing value of the next paragraph checks
the fix correctly.

Change-Id: I0ab3b8bbff33c5488f4b4af1ea4dabf7105103f2
Reviewed-on: https://gerrit.libreoffice.org/57231Reviewed-by: László Németh <nemeth@numbertext.org>
Tested-by: László Németh <nemeth@numbertext.org>

Additionally adapted EscherPropertyContainer's AddOpt
methods to be more safe and flexible. Added support for
SvMemoryStream. Changed Data handling in EscherPropSortStruct
to use std::vector<sal_uInt8>, adapted all usages and
converions throgh all modules. This makes memorty handling
for these parts much safer (no more local new-ops, no
longer delete[] there).
Currently there are quite some usages of nProp.size() and
&nProp[0] to access the data. but the base for further
changes to work more on std::vector is done.

Change-Id: I982225c5bfc06fdd9e195d18cd3c550d15f1d48d
Reviewed-on: https://gerrit.libreoffice.org/57061
Tested-by: Jenkins
Reviewed-by: Armin Le Grand <Armin.Le.Grand@cib.de>

When embedding, the fonts are gathered and added to the ODF
document. This change in addition adds the functionallity to
embed used fonts only and filtering out of font scripts. When
embedding used fonts only, they are filtered depending on if
the style (referencing the font) is in use and in addition if
the font is used in direct formatting.

Change-Id: I1ba0066b1eb34bdde7247eeb63c1be5a91db984d
Reviewed-on: https://gerrit.libreoffice.org/57168
Tested-by: Jenkins
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>