Commit by vtokarev :: r1029735 khtml/branches/KDE/4.3/kdelibs/khtml/xml/dom_elementimpl.cpp:

backport r1029732

Commit by vtokarev :: r1029732 khtml/trunk/KDE/kdelibs/khtml/xml/dom_elementimpl.cpp:

don't handle event if it was before
BUG: 189173

Commit by vtokarev :: r1020539 khtml/trunk/KDE/kdelibs/khtml/editing/ (3 files):

remove obsolete files

Commit by vtokarev :: r1020538 khtml/trunk/KDE/kdelibs/khtml/ (41 files in 8 dirs):

Merge the GSoC editing branch into trunk

Commit by vtokarev :: r988705 khtml/trunk/KDE/kdelibs/khtml/rendering/render_table.cpp:

Fix behaviour of hit test check when only child considered
we could've incorrectly returned true as result though with no result node
BUG: 181769

Commit by vtokarev :: r975454 kdecore/trunk/KDE/kdelibs/kdecore/config/kconfigini.cpp:

optimization: use QByteArray data directly to avoid overhead of append() method call

Commit by vtokarev :: r971542 khtml/trunk/KDE/kdelibs/khtml/rendering/render_container.cpp:

Invalidate and delete line box subtree of the flow when taking out
inline flow objectf rom the rendering tree. We may insert it somewhere afterwards
but still would need to recalculate inline boxes for it.
In theory, we should now better support rendering tree modifications
which was triggered lately by better continuation merge.

Thanks to Maks and Germain for the assistance!

BUG:192380
BUG:190350
BUG:191027
BUG:192105

Commit by vtokarev :: r961195 khtml/trunk/KDE/kdelibs/khtml/xml/ClassNodeList.cpp:

We already have preparsed class attribute for every DOM element (for
faster css selectors) no reason to do it here all over again
less code and better performance - nice :)

Commit by vtokarev :: r949638 regression/trunk/tests/khtmltests/regression/ (4 files in 2 dirs):

Add test for bug 187938

Commit by vtokarev :: r949425 khtml/trunk/KDE/kdelibs/khtml/rendering/render_flow.cpp:

Set continuation to 0 when we actually removed it
BUG: 187938

Commit by vtokarev :: r949424 khtml/trunk/KDE/kdelibs/khtml/xml/ (dom_restyler.cpp dom_restyler.h):

Optimize DynamicDomRestyler with ElementImpl* specific multimap implementation
it makes mapping to element itself or it's parent cheaper (which is rather a common case)

Commit by vtokarev :: r949422 khtml/trunk/KDE/kdelibs/khtml/css/ (cssstyleselector.cpp cssstyleselector.h):

this header is not really needed here

Commit by vtokarev :: r948248 khtml/trunk/KDE/kdelibs/khtml/rendering/ (render_table.cpp render_table.h):

recalculate dirty table grid if needed before using it in adjacent cell accessors(change from WC)
thanks to spart for noticing it
BUG: 182907

Commit by vtokarev :: r945551 khtml/trunk/KDE/kdelibs/khtml/xml/ (3 files):

Use remaind bit in NodeImpl for storing :hover dependency of element on itself
which appeared to be very popular

Commit by vtokarev :: r941935 khtml/trunk/KDE/kdelibs/khtml/ (4 files in 2 dirs):

Nice performance improvement by Maksim Orlovich in IDString
and my utilization of that in prospective tokenizer, which makes
processAttribute/emitTag timing sane

Commit by vtokarev :: r941880 khtml/trunk/KDE/kdelibs/khtml/css/cssstyleselector.cpp:

misunderstood lang selector code and made it wrong with previous commit, fix it now

Commit by vtokarev :: r941605 khtml/trunk/KDE/kdelibs/khtml/ (3 files in 2 dirs):

Add caseSensitivity parameter, for startsWith/endsWith methods in
DOMStrngImpl, which allows to avoid some more conversions in checkSimpleSelector

Commit by vtokarev :: r941604 khtml/trunk/KDE/kdelibs/khtml/ (2 files in 2 dirs):

Use optimized version of getAttribute, store DOMString for contentLanguage
in the document, to avoid conversions

Commit by vtokarev :: r940589 kdecore/trunk/KDE/kdelibs/kdecore/config/kconfiggroup.cpp:

More QChar/QString conversion madness, seems to be used enough to care, at least in amarok

Commit by vtokarev :: r940588 kdecore/trunk/KDE/kdelibs/kdecore/io/kurl.cpp:

More QChar/QString converstion madness, make khtml a bit happier with it