As discussed in the mail thread starting at <http://mail-archives.apache.org/
mod_mbox/openoffice-dev/201806.mbox/%3c651c8fee-b467-421c-eae1-a8710f41692c
@apache.org%3e> "Just a little side note on the scripting framework ...",
external code that uses the Java class
com.sun.star.script.framework.provider.ClassLoaderFactory stopped working
because LO changed that class in binary (and compile-time) incompatible ways
over time.

The class is not listed at
<https://api.libreoffice.org/docs/java/ref/index.html> (and neither at
<http://www.openoffice.org/api/docs/java/ref/overview-summary.html>), so it was
not considered part of the stable URE interface.  But it is apparently used by
external code, and it indeed seems to make sense that it is used by external
code that implements scripting providers.  (A follow-up commit should therefore
mark the class as part of the stable URE interface.  I keep that separate so
that it is easier to backport this functional fix.)

With ScriptProviderForooRexx.oxt from
https://svn.code.sf.net/p/bsf4oorexx/code@r589 installed in LO, "Tools - Macros
- Organize Macros - ooRexx... - My Macros - Create... - Library1 - OK -
Create... - Macro1 - OK - Edit" failed due to

> warn:cui.dialogs:21768:21768:cui/source/dialogs/scriptdlg.cxx:740: Caught exception trying to invoke N3com3sun4star3uno9ExceptionE msg: [jni_uno bridge error] UNO calling Java method invoke: non-UNO exception occurred: java.lang.NoSuchMethodError: com.sun.star.script.framework.provider.ClassLoaderFactory.getURLClassLoader(Lcom/sun/star/script/framework/container/ScriptMetaData;)Ljava/lang/ClassLoader;
> java stack trace:
> java.lang.NoSuchMethodError: com.sun.star.script.framework.provider.ClassLoaderFactory.getURLClassLoader(Lcom/sun/star/script/framework/container/ScriptMetaData;)Ljava/lang/ClassLoader;
> 	at com.sun.star.script.framework.provider.oorexx.ScriptEditorForooRexx.edit(ScriptEditorForooRexx.java:305)
> 	at com.sun.star.script.framework.browse.ScriptBrowseNode.invoke(ScriptBrowseNode.java:200)

cae57d2e "ClassLoader->URLClassLoader" (which
this commit reverts) had changed the return type of the two getURLClassLoader
overloads from ClassLoader to derived URLClassLoader (and ultimately only for
cosmetic effect; it was leftover from a previous attempt at fixing a Coverity
issue by using URLClassLoader.close(), but which is only available in Java 1.7,
so the attempt had been reverted).  That caused the above failure.

And 68cd011c "java: reduce scope, make some
methods private" (which this commit also reverts) had changed the second
getURLClassLoader overload (which is not called in the above scenario) from
public to private, which is also a binary-incompatible change.

Other commits removed throws clauses, which is only a compile-time issue but not
a binary-incompatible change.  I left those changes in for now, but if need be
they could also be reverted.

Change-Id: I98f533d88c7c1580956c3c281e72a1c78fa3f56f
Reviewed-on: https://gerrit.libreoffice.org/55871
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>

If a token is not suitable for the generic threading code, it's presumably
not suitable for threading with SwInterpreter either. This is possibly
already handled by checks in CheckToken(), but this can't hurt, just
in case a token is in CheckForThreading() but not in CheckToken().

Change-Id: Ib969a21e9021fc6debc2131dc718419c6b25d363
Reviewed-on: https://gerrit.libreoffice.org/55886Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
Tested-by: Luboš Luňák <l.lunak@collabora.com>

Now there's just one place to decide what is used first,
in InterpretFormulaGroup(). Place in other places now checks both threading
and OpenCL, which should be cheap, but it also allows e.g. falling
back from OpenCL to threading if it doesn't work out for a specific
formula group.

Change-Id: I0cac55197c5278174d303691c20f77b842995c84
Reviewed-on: https://gerrit.libreoffice.org/55885Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
Tested-by: Jenkins

Project: help  2bb9f040c74d9e609a53f048447abe815f99fc39

Helponline: Favicon refresh

Change-Id: I53e871ae17019c8eadc762ad72de75b2fb7f7177

Change-Id: I4eab47fd3358d26587e61c9f030f9ed10b067677
Reviewed-on: https://gerrit.libreoffice.org/55783
Tested-by: Jenkins
Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>

Change-Id: I06128ed7d28d17fa9a395878efaef4b890f5dbe0
Reviewed-on: https://gerrit.libreoffice.org/55760
Tested-by: Jenkins
Reviewed-by: Jan Holesovsky <kendy@collabora.com>

Project: help  79ed6797fcdaa70f6993d34070a3d033440761c2

Add help pages for internal Basic libraries (WIP)

LibreOffice ships Basic libraries with many nice Functions and
Subs that are not documented at all. These pages shed some
light to these libraries and is a work in progress.

Change-Id: Icb237bf552b785e576e984e47fd0c5cf4f7e29d7
Reviewed-on: https://gerrit.libreoffice.org/55876
Tested-by: Jenkins
Reviewed-by: Olivier Hallot <olivier.hallot@libreoffice.org>

Project: help  2fdf0556202b43940ee003e7d1e6d452c3d11ab7

Fix <bascode> rendering for paragraphs

Change-Id: I4c76bac48f3be30dc3bcd358df427843a5f97a7a
Reviewed-on: https://gerrit.libreoffice.org/55837
Tested-by: Jenkins
Reviewed-by: Olivier Hallot <olivier.hallot@libreoffice.org>

Change-Id: I6d95b5235d43dabc17515f3ac31063093c68d959
Reviewed-on: https://gerrit.libreoffice.org/55860
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Change-Id: Iced10ed59c475dff4d33ff06151b2015a27a860b
Reviewed-on: https://gerrit.libreoffice.org/55715
Tested-by: Jenkins
Reviewed-by: Tor Lillqvist <tml@collabora.com>

Change-Id: I23e21b5d108fcc4bfc4422bebbc50389565e36fb
Reviewed-on: https://gerrit.libreoffice.org/55864Reviewed-by: Tor Lillqvist <tml@collabora.com>
Tested-by: Tor Lillqvist <tml@collabora.com>

Change-Id: I6c58c0f26279d33c199109350da5e84a5764e950
Reviewed-on: https://gerrit.libreoffice.org/55832
Tested-by: Jenkins
Reviewed-by: andreas_kainz <kainz.a@gmail.com>

As it modifies ScAttrArray, which is not thread-safe (has std::vector
per each column, so multiple threads may try resize it etc.). So if
threaded calculation is done, delay the calls to the main thread.

Change-Id: I3d87665c0dd0d40f0c2efbcf8958240ee5580233
Reviewed-on: https://gerrit.libreoffice.org/55602Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
Tested-by: Jenkins

Change-Id: I08a6e29aeac8c16bbb987c5513cbf3d0777d4b3a
Reviewed-on: https://gerrit.libreoffice.org/55835
Tested-by: Jenkins
Reviewed-by: andreas_kainz <kainz.a@gmail.com>

Change-Id: I6c30cf858f72ef01839c82ac6962ed0bf40d6355
Reviewed-on: https://gerrit.libreoffice.org/55834Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>
Tested-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>

Change-Id: Ic3438ff0862f74273fe0aec443aaeacee331abdd
Reviewed-on: https://gerrit.libreoffice.org/55843
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

See https://bugs.documentfoundation.org/attachment.cgi?id=142760
Found thanks on gdb with command
catch throw com::sun::star::sdbc::SQLException

First SQLException was dealt but not the second one

Change-Id: I7d5d3ad612db132a3bda97cff14367912ae885f9
Reviewed-on: https://gerrit.libreoffice.org/55836
Tested-by: Jenkins
Reviewed-by: Lionel Elie Mamane <lionel@mamane.lu>

Change-Id: If7c4bcfc0c03bdab3209d27f0e8fa75a498feae1
Reviewed-on: https://gerrit.libreoffice.org/55842
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Change-Id: Ibd4f858abe825652c3df68ce7a21cbf16feea735
Reviewed-on: https://gerrit.libreoffice.org/55824
Tested-by: Jenkins
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Eike Rathke <erack@redhat.com>

Some opcodes, such as ocName, have no textual representation.

Change-Id: Ibca80a8ae0f05d8649095d111aadcf3ba9b77ebb

Change-Id: I2c8102ebd50d8df055c6c30a493a718b8ac8b528

Change-Id: I70a6f44ecd25155faf64adaf795d330b12c247c7
Reviewed-on: https://gerrit.libreoffice.org/55817
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

...between c3707757 "loplugin:unusedfields" and
8b0dae14 "weld SvxStdParagraphTabPage"

Change-Id: I1a76690a73759a8439df99848c35fa08bc18a253

Change-Id: Ib5098cc38ca9fecfcf2d57e59573253b714f2df4
Reviewed-on: https://gerrit.libreoffice.org/55785
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>

this speeds things up by 30% for me

Change-Id: I7fa99e91b0b4f354329803b9c8fab827bd367dac
Reviewed-on: https://gerrit.libreoffice.org/55812
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>

Change-Id: I4043efc204c8e3af8463ecd325313c99a9d61128
Reviewed-on: https://gerrit.libreoffice.org/55839Reviewed-by: Björn Michaelsen <bjoern.michaelsen@libreoffice.org>
Tested-by: Björn Michaelsen <bjoern.michaelsen@libreoffice.org>

SwUserFieldType::GetValue calls SwCalc::Calculate on this object, which
resets own m_sCommand to the variable name being verified. Restore it
after calculations, like other values.

Change-Id: I1582fa2d356064e6e951063fa8479cd615a6591a
Reviewed-on: https://gerrit.libreoffice.org/55789
Tested-by: Jenkins
Reviewed-by: Mike Kaganski <mike.kaganski@collabora.com>

Change-Id: I69b4f139be269ff58acd9d74ce0c7897e84a9419
Reviewed-on: https://gerrit.libreoffice.org/55774
Tested-by: Jenkins
Reviewed-by: Adolfo Jayme Barrientos <fitojb@ubuntu.com>

Change-Id: Iadc905916c0b6437eec8cc19ce838719c16a44d0
Reviewed-on: https://gerrit.libreoffice.org/55544Reviewed-by: Heiko Tietze <tietze.heiko@gmail.com>
Tested-by: Jenkins

Change-Id: I9b7e31f2335f01e20ee71137fdb2fd0315f3bd91
Reviewed-on: https://gerrit.libreoffice.org/55542Reviewed-by: Heiko Tietze <tietze.heiko@gmail.com>
Tested-by: Jenkins

Change-Id: I1a01ec978063a897b71abf7254ec17d3ed77f156
Reviewed-on: https://gerrit.libreoffice.org/55540Reviewed-by: Heiko Tietze <tietze.heiko@gmail.com>
Tested-by: Jenkins

Change-Id: Id175bff77492fe78c393ceb36e58f427c6d532c2
Reviewed-on: https://gerrit.libreoffice.org/55505Reviewed-by: Heiko Tietze <tietze.heiko@gmail.com>
Tested-by: Jenkins

Change-Id: I02b0f2e4f049b407a3ac0c7f304cbcd0cc652749
Reviewed-on: https://gerrit.libreoffice.org/55818
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

so when

commit dc5a7f14
Date:   Wed Jun 13 13:11:50 2018 +0200

    tdf#118110 - Mouse wheel should work with the shapes desk

removed the scrollwindow, the viewport can go too

Change-Id: I1f5617da272782a51c3dc38fa1c318a935aac59d
Reviewed-on: https://gerrit.libreoffice.org/55819
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Change-Id: I1b6fe3ea2ef17bd35de1e8d4d124bc7adf9861de
Reviewed-on: https://gerrit.libreoffice.org/55799
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Change-Id: Ieba56962ae1986686ae649ac15ba721a3fbfae5e

Change-Id: I7d9966d11f8272034523d9804ecf76e119552780
Reviewed-on: https://gerrit.libreoffice.org/55796
Tested-by: Jenkins
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Which effectively would be going to happen with
https://gerrit.libreoffice.org/49033 for 6-0 and
https://gerrit.libreoffice.org/55795 for 6-0-5
so have the same on master in case it would have unknown side
effects.

Change-Id: I312d7c5e723fb9417ac512641acb843ecd7a9867
Reviewed-on: https://gerrit.libreoffice.org/55811Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Jenkins

When extensions bring along help content (in the form of .xhp files), that was
completely unavailable in --without-help (i.e., defering to online HTML help)
and --with-help=html (i.e., displaying the new offline HTML help in a browser)
builds.  (Merely extended tooltips brought along by extensions would work in
those builds.)

With this change, at least context-sensitive help (i.e., pressing F1 in an
extension's dialog) will be shown.  Although not integrated with the rest of the
(browser-based HTML) help, but using LO's own old help viewer.  (Which will then
of course only offer the combined help content of all the installed extensions,
not any help content provided by LO itself.  Thus, links from extensions' help
content to LO help content will not work.)

This is only a cheap stopgap.  If an extension brings along help content other
than context-sensitive help (which is reachable by pressing F1 in the respective
context), there is no easy way for the user to reach it.  (That includes help
pages listed in an extension's help.tree, and index entries.  See the commit
message of 4a0ca73c "Add some help content to
test-passive.oxt" for an overview of what help conent an extension can provide.)

Regarding the details of this code change, I left the
comphelper::LibreOfficeKit::isActive() case alone, assuming that that case never
wants to display LO's own old help viewer, for one reason or another.  (But
please correct me/the code if that assumption is wrong.)

Change-Id: I2209b1aa48888c5f25674fbe4ce6643eff1268c6
Reviewed-on: https://gerrit.libreoffice.org/55816
Tested-by: Jenkins
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>

Change-Id: I40726ace2cb8c1c5e02d4c68154d920d079bdf93
Reviewed-on: https://gerrit.libreoffice.org/55546Reviewed-by: Heiko Tietze <tietze.heiko@gmail.com>
Tested-by: Jenkins