reworked reflow to more accurately calculate its size
r=kmcclusk,dcone bug=12653
git-svn-id: svn://10.0.0.236/trunk@56847 18797224-902f-48f8-a5cc-f745e15eee43
Also, commented out the adding of border-padding to the maxelementsize
r=kmcclusk bug=12653
git-svn-id: svn://10.0.0.236/trunk@56845 18797224-902f-48f8-a5cc-f745e15eee43
and the the proper check against the style mVisible attr (bug 21701)
r=kmcclusk bug=12653
git-svn-id: svn://10.0.0.236/trunk@56843 18797224-902f-48f8-a5cc-f745e15eee43
r=dcone.
problem was we were failing to null out the editor member of the controller on frame destruction.
git-svn-id: svn://10.0.0.236/trunk@56505 18797224-902f-48f8-a5cc-f745e15eee43
options or one optionwith no lable and no content and it sizes out to be zero
high also fixed twhere visibility was being check incorrectly
r=kmcclusk
git-svn-id: svn://10.0.0.236/trunk@56404 18797224-902f-48f8-a5cc-f745e15eee43
with no lable and no content and it sizes out to be zero high
also fixed twhere visibility wasbeing check incorrectly
r=kmcclusk
git-svn-id: svn://10.0.0.236/trunk@56396 18797224-902f-48f8-a5cc-f745e15eee43
basically we ignore them and pass back our most recently reflowed size
r=karnaze bug=22063
git-svn-id: svn://10.0.0.236/trunk@56394 18797224-902f-48f8-a5cc-f745e15eee43
a=i'm leaving now for vacation, and i can't find anyone around, but i'm assuming that you
actually want them before January 3rd, so I'm checking them in.
git-svn-id: svn://10.0.0.236/trunk@56179 18797224-902f-48f8-a5cc-f745e15eee43
and 18500 (Can't drag-select in text widget unless focussed first)
This fix does NOT address other focus-related problems having to do with text controls.
r=rickg
a=chofmann
platform testing support=kin,dcone,akkana (THANKS!)
git-svn-id: svn://10.0.0.236/trunk@56143 18797224-902f-48f8-a5cc-f745e15eee43
r=troy, a=chofmann
the editor's internal mDisplayFrame was not being removed from the frame manager's map of content->frames, so after enough scrolling in the addressing pane, reflow would try to reference a deleted frame. Fix is to remove the frame from the frame manager every time it's destroyed.
git-svn-id: svn://10.0.0.236/trunk@56134 18797224-902f-48f8-a5cc-f745e15eee43
r=pollmann,a=chofmann
Basically, this just calls a method in the security manager to determine if a page loaded from the document's URI
can be allowed to open a page from the ACTION URI. This is the same check that's currently used on window.open,
for example.
git-svn-id: svn://10.0.0.236/trunk@56044 18797224-902f-48f8-a5cc-f745e15eee43
changed the way a temp nsSize was being created.
removed all the dom listener interfaces and code (they weren't being used)
r=beard,self a=chofmann bug 19416, 20957, 19300, 20250
git-svn-id: svn://10.0.0.236/trunk@55991 18797224-902f-48f8-a5cc-f745e15eee43
reworked again. But this will able some XUL selects to work that otherwise would be
broken and I haven't found it to cause any problems with a lot of testing. I have
regressed to this situation in the past and it worked OK.
r=kmcclusk, bug=none but hangas (or self) would have filed one by end of day.
git-svn-id: svn://10.0.0.236/trunk@55654 18797224-902f-48f8-a5cc-f745e15eee43
also fixed a a small problem with border padding being added in for fixed size comboboxes
r=self, bug 19382
git-svn-id: svn://10.0.0.236/trunk@55379 18797224-902f-48f8-a5cc-f745e15eee43
2.) WebShell now implements the new nsIScriptGlobalObjectOwner.
3.) WebShell supports GetInterface to nsIScriptGlobalObject.
4.) Documents no longer carry around a reference to nsIScriptContextOwner. Instead they hold on to a nsIScriptGlobalObject. nsIDocument::GetScriptContextOwner has now become nsIDocument::GetScriptGlobalObject(). Same change to the set methods.
git-svn-id: svn://10.0.0.236/trunk@55175 18797224-902f-48f8-a5cc-f745e15eee43
