Related with commit c7c2b4b5.

There is nothing wrong with having several activities; in fact Android
development encourages using one activity per use case. After this
patch we have:

* One menu item that opens LibreOffice document browser. It covers the
  use case "open LibreOffice without documents".
* The viewer activity does not have a menu item but it responds to
  open intents triggered by any applications like file explorers,
  including the document browser.

Change-Id: Ie08b96e52d3a6a79aed3354620302705d7689381
Reviewed-on: https://gerrit.libreoffice.org/12108Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
Tested-by: Miklos Vajna <vmiklos@collabora.co.uk>

Writer documents currently always have one part, so it's not possible to
switch to an other part anyway. Also hide the matching menu item
accordingly.

Change-Id: Ic634d46c4ff307416b2878e5bc8a590b22a827a0

Not that we're currently using that, but..

Change-Id: I67b8fc324779875ba14e2d69204c40fe27cc180e

Change-Id: Ia90a4c371e66f4a5b9a37b4c4ac7d3118ad518a5

fix assert on loading ooo123653-1, we don't care about ranges that end before
the starting point we are interested in

Change-Id: I46ddf882ce4099dbcd77c62569c5c51f1ce5aa76

css::frame::XModel::getArgs() may or may not return a Sequence that
contains a DocumentService key, while
css::lang::XServiceInfo::supportsService() can always determine the
document type.

This fixes the problem that doc_getDocumentType() returned
LOK_DOCTYPE_OTHER for Writer documents on Android.

Change-Id: I380d59a963553fb30a3eb20fbe84dcfc6a1bbd61

Change-Id: I1b37d8332e9a5b779c27373b2ec038ccdf2b90df

...try

GetProcessServiceManager().createInstance("stardiv.UnoControls.ProgressMonitor").addText("foo", "bar", false)

in Basic...

Change-Id: I30318c3e8e671a97b6a3fe2dd9ec03add21794ab

...try

GetProcessServiceManager().createInstance("stardiv.UnoControls.StatusIndicator")

in Basic...

Change-Id: Iafc22188feb8a1d3f1b19ac4f6e209be62a44d17

Change-Id: Iebf1c42c384909f6226c25eb151985f8bc244c93

Change-Id: I6efaac9bd21ab986af77fed1a949d3d2b380e922

and just use OOoRunner, there is no point in having a stripped down
jar, the cost of firing up the Java VM completely dwarfs any benefit
of having a smaller jar.

Change-Id: Ibcc3c5bd6e9b9c918041142dd32db0ea5dddc25b
Signed-off-by: Stephan Bergmann <sbergman@redhat.com>

Change-Id: If5297d641b4a596f20bdde8203f199d8efcb92fe

Change-Id: I2b703312ddff29772b31cf74ab6408dbbe90a5db

Change-Id: I1dc3b0db6295c43a9d29576b81870b4a293a61aa

Change-Id: I8283ddd868639e6535e27798d91a7273d77f048c

Change-Id: I64690ce1d6960dea52bae6a5bc1d0f1de742cb63

Change-Id: I25bc6231f4efd76b33db028f6ffea19799519083

Change-Id: I214ee188793e7ef189417460ea32eeaf0a8b6a29

Change-Id: I07ee4f03ba74415c11188f7b63879ccad542537c

It looks cleaner this way.

Change-Id: Id46cb52863fb9366d83ac17d3f8521e5a13a59c9

Change-Id: I8fe6aeb2ee1cd67df3e0e657b271b255516dbc18

While leave the rest outside it.

Change-Id: I17f5b935dba8535f2bbfdd86b1037a95f2324174

Change-Id: Ie4cc8329398063b82b34cde5568686f48363ee1a

Noticed when loading
https://bugs.freedesktop.org/attachment.cgi?id=99844
with fdo#79249 fix where oox::xls::applyCellFormulas() groups the
consecutive =#N/A formulas. Only A1 result was displayed, other cells
were displayed empty.

Change-Id: I3a541f74d3d25e1515a1c6d47f02ec6a8e817c93

... instead of attempting to stringize a NaN coded error value.

Regression introduced with 30a20743

Change-Id: Ia7a8ca39938820ac75db169404446fa696c6ee1b

To use it, enable "Print layout" in Options->Load/Save->HTML Compatibility,
and select the HTML Document file type in the Writer Save As dialog.

Change-Id: I763ab8340a59050fd5c68677715679f41fd91fb3

is that mapping used at all?

Change-Id: I15019072b6e812b9ffe29d3ee6afacd9ab526948

The SwInputFld is copied because it doesn't have HasDummyChar() set
although it has 2 dummy characters; TODO why is that...

Change-Id: Iee91c1d0cf7a7a928e7383c1839f8192e8d4d5b0

... should fix field dupes"

This reverts commit f384598d.

The attributes have to be copied in case splitting a paragraph creates
an empty paragraph.

Change-Id: Icd5730dd9ab8a68f737492645988a4d2f07af7a0

Change-Id: Idaea12aae68ad21ee06f785167cdf1867101ec4b

Change-Id: I8571aa0ccc9f691d7fd9290a647117161755d3a7

Change-Id: I9965f58b8f06f1cec2c419dcf16d8aebf9cd97b8

Change-Id: Ifdb896108cf0db0a9be7ba419df785da536c0fab
Reviewed-on: https://gerrit.libreoffice.org/13465Reviewed-by: Björn Michaelsen <bjoern.michaelsen@canonical.com>
Tested-by: Björn Michaelsen <bjoern.michaelsen@canonical.com>

Change-Id: Ia6e166a4158d28f1a4145e51040f2f5637827197
Reviewed-on: https://gerrit.libreoffice.org/13466Reviewed-by: Björn Michaelsen <bjoern.michaelsen@canonical.com>
Tested-by: Björn Michaelsen <bjoern.michaelsen@canonical.com>

With f9218383, export no longer crashes,
and there is no reason to leave this piece disabled any more.  Let's
re-enable this.

Change-Id: Ibeca8869f152cbcd80f1dcb55f8199110125741d

I.e. make it possible for the opengl backend to decide to use a different
scaling algorithm.

Change-Id: I36e76de071797129c7636b3048c93bc6e5d93aca

Change-Id: I66dcf6284bd1b7afb1c60c838bbbb2052e06072b

i.e., after recent "fdo#83753: consider JAVA_HOME and PATH when selecting JRE"
fix, if jfw_findAndSelectJRE found no suitable JRE in
jfw_plugin_getJavaInfoFromJavaHome or jfw_plugin_getJavaInfosFromPath, do not
re-check those locations in jfw_plugin_getAllJavaInfos.

Change-Id: If4e085b4fceff5b2494c7b7b84ac51691dbc78cc

Change-Id: Ib6eac731a4b8193a8d9b9132da78e7961aec22ab