Change-Id: Ib5c5758eab44532b383ca49542288a65cb10d548

Change-Id: Iddb5936c088dfba0245a0d921c7122a0587bdc9b

Change-Id: I187c0b2bd2f9e4ddeeac80b6046c25d5cf674d71

SwRangeRedline::Hide() will copy and delete the redlined text range,
and during the delete it happens that FLY_AS_CHAR frames are not deleted
before their anchor SwTxtNode is destroyed, so SwFmtAnchor::m_pCntntAnchor
will have its NodeIndex adjusted by SwNodes::RemoveNode() to a wrong
node and then it cannot call RemoveAnchoredFly() on the proper node.

Avoid the problem by manually deleting all FLY_AS_CHAR from nodes
before they are deleted and before the SwNodeIndex adjustment.

(regression from 738fb2ad, which had
 the equivalent problem with differing SwNodeIndex keys)

Change-Id: I13c9577c01e3b8976b74fb9c58dc2ef2c2fc0666

If the for loop has a break after the first frame anchored on that node,
it makes no sense to assert that all frames have been visited.

(regression from bb95f7e6)

Change-Id: Ibd8d65f286b441127be7735c63f4b84ba94dcb6f

SwRangeRedline::Show() will move nodes around in the nodes-array, which
means that using SwNodeIndex as a key in a map that has a lifetime
not limited by the stack is a bad idea, as the map will become unsorted.

Remove SwFrmFmtAnchorMap from SwDoc and replace it with new
SwNode::m_pAnchoredFlys to do the same mapping.

(regression from 738fb2ad)

Change-Id: I396d92b9d0b2045e98bad6d0b374303cd4e62b59

Change-Id: Ie5376106adb3df901b9762097e59d7627790443f

Upgrade to the latest patch from http://bugs.python.org/issue17797
which appears to work even if you invoke from cmd.exe

Change-Id: I85f1cc5ad7d8c059d972ae2a6fd2be1bb5604c2c

Subtle failures in nested template instantiations there, for
sal_{u}Int8 and sal_{u}Int16 - use wide-int fallback to do bitmask
ops. Type is not used to store any dynamic values.

This partially reverts a1ecce3a

Change-Id: Ifa7070c1c3830fd70d7ced082d651411ff523444

This reverts commit 7ad4e562, which
was causing ICEs on precisely the compiler versions it was targetted
at.

Change-Id: I996792d723a8bd0d7edd5f248d74217322b5ce3f

Change-Id: Ib46e3d484f931cbdebf21ea38488313748942b3b

Change-Id: Icc026ebb422c299caca6c24d6a56416076589fc8

Change-Id: I186c81923857d06ac7a87caf9d5961f44b89b559

Change-Id: I9fa42eeeabb4a6c7ee18b52d24c480a3f7a6a753

Change-Id: Iafb48b20496924c64f9aaa3ec0965b023aef6a07

...unclear to me why 507efabe
"Resolves: #i124922# allow late setting of the font color..." left it in

Change-Id: I741ca3d7eca6fa5dff81af50d91bd6c5d1cba989

Change-Id: Ic57551d9a3590769f97742d41f089dbb3ce248a9

Change-Id: I1aeec1e8b00ab0f9e7ec130b6d72b69f35a304cf

Change-Id: Ie99d1dd241dc2bfa5c5ac2bff9f2330c491327bc

Change-Id: Ia2286661a8f62e8a46c44c390ab1a8a5dffe81d0

This should eventually enable more certainty that they are still
live by the time we emit the event later.

Change-Id: I7c9eb4c000753d4efe8c59a9e13ef3e11c93b2fa

Change-Id: Ice91c9e698d99d9674a7c7b3c688faa07603bf95

Change-Id: I50154e2deb66a77fc4354b1db0e4cd8e1b1ffd2e

Change-Id: I2931de2cc2a9916dae785d0cc130ce2aa73cf54e

These functions are only being used inside this file.

Change-Id: I473b33f794cfbaff590d00f22d2832456482d15a
Reviewed-on: https://gerrit.libreoffice.org/15680Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>

Gets rid of superfluous sequential callbacks with the same percentage when
loading large documents.

This reverts commit cec72eff.

Change-Id: I70f43f7e3a650c76cbcbbc60ebb2d47efaca06a5

Situation before/after the patch is the same, just silence lots of
harmless warnings for now.

Change-Id: I00428bdfd939d8cdd3bfd447339e0417e70f4689

Change-Id: Ib27854a8470f3ff5b208cb949a7bd02f2a86c969

Change-Id: I6079f387a697c5fe794655e455a19d0cb44b96bc

Change-Id: Ibb0f717ae21cd80386b26e078ee7eb2a873b4092

This helps avoid things dying during emission in a robust manner.
Bit of an unpleasant 3rd optional parameter; better names appreciated.

Change-Id: I27571823f9d96caef1d07602785a02390d3a3591

This way the data source's URL will be the vnd.sun.star.pkg: URL, that
gets stored in the user profile, and the data source will be usable when
we connect to the data source next time, too.

Change-Id: Ie2f45af453bfad4f813a1ea492edb633c148d08b

hit it with a large and simple hammer

Change-Id: Ib0e55ad64780efc7664ccd9a2684fbe7dba87498

Change-Id: If93609d032c9864e0251230bf9ed76e6b6741a62
Reviewed-on: https://gerrit.libreoffice.org/15672Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>

Change-Id: I69eaf2e4dd051d62f56caef962d40889cc3a3876

If not stopped, key-press events are interpreted by both LOKDocView and
GtkToolbar leading to poor editing experience.

Change-Id: Ie878d168e202a0f3e1d52191aa3fd7cc00a7cbfd
Reviewed-on: https://gerrit.libreoffice.org/15677Tested-by: Miklos Vajna <vmiklos@collabora.co.uk>
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>

Change-Id: Ic0062eddebf9225d298569bb4900047a0ee0b112

Change-Id: I0041724d340bf75a184910805f7f927efb3b2131

Change-Id: I33d722c27d85732d23510ebaaaba0b8792d0d61d

and we don't want to see the mis-spelt underlines

Change-Id: Ib3ae29297a1697e0f7b0caa816504da7edf721d1