we look for any kind of scalar variable now that deserves to be inlined,
and we check for variables that cannot be inlined because they are being
passed by reference, or modified, or have their address taken

Change-Id: Ia744a180e91d1516140a1555d4514f6fa4de1c0b
Reviewed-on: https://gerrit.libreoffice.org/38966Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>

When the field in question is read from inside a constructor
initializer.

In the process, create some needed infrastructure in the plugin classes.

Change-Id: I2f440efa6912801a236727c9fe3180404616958c
Reviewed-on: https://gerrit.libreoffice.org/38960Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>

Change-Id: I2f74dee5c05420275aae57485b8d81f60fff2d5d
Reviewed-on: https://gerrit.libreoffice.org/38823Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>

Change-Id: If8e0810c4998b2d56609a0c9fe10ce061d1a6347

Change-Id: Iddbe19638796564f5d443f6376b95ad7d69ecf0e
Reviewed-on: https://gerrit.libreoffice.org/38989Reviewed-by: Yousuf Philips <philipz85@hotmail.com>
Tested-by: Yousuf Philips <philipz85@hotmail.com>

Change-Id: I747562462accc49e7f3dc851e1ebb274e36b93d7
Reviewed-on: https://gerrit.libreoffice.org/38988Reviewed-by: Yousuf Philips <philipz85@hotmail.com>
Tested-by: Yousuf Philips <philipz85@hotmail.com>

Change-Id: If8ed724ccec136653db5fcf43f9e74a82826add7

Change-Id: Ibf6379425d312b2cc2c9d663a65d0e02209f74e4
Reviewed-on: https://gerrit.libreoffice.org/38981Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>

Tables-in-sections were already split across multiple pages, but not
secions-in-tables. To be safe still don't allow
sections-in-tables-in-sections, so you can combine these in both orders
now, but not recursively.

To achieve this, relax two "not in table" conditions to just require
"not in a table that is already in a section", and define that in case a
section-in-table is to be split, the follow section frame should be
inserted under the follow of its cell.

With this, finally the section frame in the bugdoc is split into two,
and the second section frame is moved to the second page as expected.

Change-Id: I16ebb2d30870b145a2378d46603324ab267b0dd3
Reviewed-on: https://gerrit.libreoffice.org/38965Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Jenkins <ci@libreoffice.org>

Change-Id: I9ba7a9d104dd6a5e47aa62058d0602101053e9af

ScXMLExternalRefTabSourceContext
ScXMLExternalRefRowsContext
ScXMLExternalRefRowContext
ScXMLExternalRefCellContext
ScXMLExternalRefCellTextContext

Change-Id: Ia626ed62186e53a352a5c7831b41be99ebc55018
Reviewed-on: https://gerrit.libreoffice.org/38881Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Mohammed Abdul Azeem <azeemmysore@gmail.com>

Change-Id: I48b302780db420346280d782c1804fe6e93c90be

Change-Id: I511a13f7785a0de6efaa8439d3f0bff20a1644ed

Add tests:
- test_meta
- test_meta_empty
- test_meta_field
- test_meta_field_empty
- test_bookmark1
- test_bookmark2
- test_refmark2
- test_refmark3
- test_toxmark2
- test_toxmark3
Change-Id part 1: I0920517b5e8806a5fc46708822fed33006f03fd5

Change-Id: If8e50aa036f10caf1b460c041dd7bce068f9cd7b
Reviewed-on: https://gerrit.libreoffice.org/38842Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>

Change-Id: I5c2ead67465c79fa30fc7d7db9d48748bdaf5491
Reviewed-on: https://gerrit.libreoffice.org/38961Reviewed-by: Yousuf Philips <philipz85@hotmail.com>
Tested-by: Yousuf Philips <philipz85@hotmail.com>

commit 6149da20
    replace misc double checked locking patterns
    ... with thread safe local statics

reintroduced bug #i77768# despite big comment

Change-Id: Ic6e204a277c8aa4f43c5fd15a42066c6f8a69dad
Reviewed-on: https://gerrit.libreoffice.org/38959Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
Tested-by: Jenkins <ci@libreoffice.org>

Change-Id: If8fb8f5e51bba8e631a6fc5db54487b418c19e05

The code snippet suggested has no quotes in atttributes values.

Change-Id: I604fb34827da3bebf3f6759b491d719a96854754
Reviewed-on: https://gerrit.libreoffice.org/38893Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Katarina Behrens <Katarina.Behrens@cib.de>

for processingInstruction method in XFastDocumentHandler
and for XFastNamespaceHandler API.

Change-Id: Ie0fd8ceb42e7ffbce041796b98edaa7f232dbb62
Reviewed-on: https://gerrit.libreoffice.org/38964Reviewed-by: Mohammed Abdul Azeem <azeemmysore@gmail.com>
Tested-by: Mohammed Abdul Azeem <azeemmysore@gmail.com>

Change-Id: Ia55f3fefe7c14327cff2e996ab0038dc52f9b017
Reviewed-on: https://gerrit.libreoffice.org/37496Reviewed-by: Samuel Mehrbrodt <Samuel.Mehrbrodt@cib.de>
Tested-by: Samuel Mehrbrodt <Samuel.Mehrbrodt@cib.de>

allow it to pass up to superclasses for parsing, from discussion:

<noel_grandin> caolan, is it intended that the VclAlignment x/y
align/scale members are effectively unused?
<caolan> noel_grandin: its based off the GtkAlignment so that's why it
had them in the first place. Never got around to doing anything about
scale so no surprise there and might as well remove those
<noel_grandin> caolan, will do, thanks
<caolan> I would have expected the xalign/yalign to do something, I
think looking closer that we should be using those and probably that
stub impl is blocking the underlying Window::set_property from using the
supported left/right top/bottom values, so you can try remove those too,
but I'd do it in a separate commit from the definitely useless scale
ones
<noel_grandin> caolan, at the moment, the change will remove them, but
not the parsing of the values, so it should not change the functionality
<noel_grandin> caolan, I'll do the removal of the parsing in a separate
commit, which will pass them up to Window::set_property

Change-Id: I5bb043a5c38d494355bf4766af8521838b7fc3f2
Reviewed-on: https://gerrit.libreoffice.org/38957Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>

Change-Id: I7c776d3b02bd8d35f160e82916213e402e9c88ab

Change-Id: Ic8a9691a8ddf32eaf58227a9533113f9bac596eb

Because that class isn't an iterator, but a class that produces
iterators (with its begin() and end() member functions).

Change-Id: I13125ced5e82406d049a61078e31690312a82620

Change-Id: I2f7bfbd62c0a5477f0cbb38e86684db4fdb9c37f
Reviewed-on: https://gerrit.libreoffice.org/38952Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Change-Id: I93d1ae29bc414c5a60e793acd9e92d35608552ee
Reviewed-on: https://gerrit.libreoffice.org/38940Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>

Both osl_add/removeSignalHandler apparently handle !handler gracefully (the
former even including explicit code for that case, which doesn't fit with having
an assert), and before bca760c2 "tdf#43157 -
osl: convert OSL_ASSERTs to assert in signalshared.cxx" those asserts were
merely OSL_ASSERTs.  So keep this published API accepting !handler (even if the
documentation is silent on this).

Change-Id: I63512916a66a2a58b932174ec5d92ce612c236a2

tell the plugin code when we are unit-testing it, so we can suppress all
the warnings except for the plugin we are currently testing

Change-Id: I240c8e37eba90c219e53c29531a3a43bc841a1c8

...assuming these values are indeed intended to be negative (and <https:
//developer.apple.com/fonts/TrueType-Reference-Manual/RM06/Chap6.html#Types>)
mentions a "Fixed: 16.16-bit sgined fixed-point number" type, apparently
represented here by F16Dot16, so the assumption looks plausible).

During CppunitTest_vcl_pdfexport, encountered negative values -1184 and -22 with
instdir/share/fonts/truetype/Carlito-Regular.ttf and -8 with
instdir/share/fonts/truetype/LiberationSans-Bold.ttf.

Change-Id: Ia07c61aebc2c9c67d0ed6173ecac94bab2abb2a8

This one has no extra functionality at all, and its only purpose is to
be used in range-based for loops. If there is a cleaner way to do
this, feel free. Not sure if this functionality could or should be
combined with either of the two existing iterator classes related to
FormulaTokenArray (FormulaTokenIterator and
FormulaTokenArrayPlainIterator). Probably not.

Change-Id: I32599b0800fd2585624d3742a46ad4896ce7e47a

Change-Id: I06dd3639e4110700e070f1224112a9fa6597a832

Change-Id: I05ee0c0db2d54625413c9b89a12e24b9de7c45da

Change-Id: I9a947078d9f2c1c06cb8524be137ba5e36e97a0b

Change-Id: Ib9769706e4f66a1ecae49869565e84091cd2279a
Reviewed-on: https://gerrit.libreoffice.org/38945Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Jenkins <ci@libreoffice.org>

Change-Id: I704919e0a834e1431e153cd57f48de08c0f09c11
Reviewed-on: https://gerrit.libreoffice.org/38948Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>

Change-Id: If038daf4350bfcd2e7146809f63ee4f062a0dc81
Reviewed-on: https://gerrit.libreoffice.org/38670Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Katarina Behrens <Katarina.Behrens@cib.de>

Change-Id: I85c0baa32b37201fc2caaa4e5bd390e060241265
Reviewed-on: https://gerrit.libreoffice.org/38951Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>

Change-Id: I8daa6c2c5941a894259a5af74a16f7ef32f8a867
Reviewed-on: https://gerrit.libreoffice.org/38577Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Katarina Behrens <Katarina.Behrens@cib.de>

in commit 42922c40
"RTFDocumentImpl::popState: use switch here"

found while running the unusedfields plugin

Change-Id: I4be41dba976a13dc6d39155fb2a04d039dc0443a
Reviewed-on: https://gerrit.libreoffice.org/38950Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>

Change-Id: I2bcbf84dff8a5f6a2311cd0b1f19e90c03e6175c