7d9dc8d482
bug 628561 DNS caching forcing one to restart Firefox to eliminate temporary DNS redirection; r=bzbarsky
...
A force-reload now clears persistent connections to the server related
to the force-reloaded resource. This will allow renogitation of DNS or
server load balancing.
--HG--
extra : rebase_source : 5c55e90ea64039b9cdc0a2d85a51086d2b1d40df
2011-03-24 12:01:57 -04:00
748b83231a
Bug 542401 - Look into whether Firefox should set TCP_NODELAY for non-SSL HTTP socket r=jduell
...
indeed without nagle disabled (which is what nodelay does) a pipeline
can be interrupted for a full rtt waiting for an ack, which defeats
the purpose of pipelining. Other parts of FF can just coalesce the
writes internally, but sometimes that is not possible for pipelines
which may not have an opportunity for a pipelined request when the
"runt" packet goes out, but discovers such an opportunity a long time
before the rtt is expired. In other places (such as long posts) we do
a fine job of coalescing already, so this should not result in any
extra packets on the wire, just improved latency.
--HG--
extra : rebase_source : f83f5b080a3da5875abaee001e8af18503fcff38
2011-03-23 11:03:09 -04:00
2c0ac326b8
643352 keep-alive header syntax invalid r=bz
...
don't send the keep-alive request header. It is redundant to
connection: keep-alive and we don't send the right syntax anyhow.
--HG--
extra : rebase_source : c6d9cb95d2d1cac30bc718884eb3b909db0d6a43
2011-03-23 11:02:33 -04:00
0269c0b807
Bug 643291 - chnage nshttpcompressconv NS_ERROR to NS_WARNING for data input error r=honzab
...
--HG--
extra : rebase_source : 14dbf0bc5cd5acbc1c525f7c8f19282c494e24f7
2011-03-23 11:01:34 -04:00
6985fe57fb
Bug 632419 - HTTP 100 level parser race condition r=bz
...
--HG--
extra : rebase_source : 0cd0a74bfe850d0c16d7f74b77c8ca0c286e1144
2011-03-23 10:59:58 -04:00
710a053423
Bug 640946 - Add a method to nsIFormControl to know if a form control can have draggable children. r=smaug
2011-03-24 15:55:32 +01:00
0a0ba7aa70
Bug 563669 - Move nsGenericHTMLFormElement::AcceptAutofocus to nsIFormControl. r=smaug
2011-03-24 16:10:42 +01:00
f272e1571a
Bug 641409 - Move methods using only using GetType from nsGenericHTMLFormElement to nsIFormControl. r=bz
2011-03-24 17:06:58 +01:00
5cebe2c042
Merge layout work from projects/birch, part 3. a=smaug
2011-03-24 08:38:19 -07:00
32b2099d0d
bug 635639 - part 2: reftests for zero-width diacritics; r=jdaggett
2011-03-24 15:07:30 +00:00
645d9c645b
bug 635639 - vowels are not rendered correctly in some Persian/Arabic/Hebrew fonts. r=jdaggett
2011-03-24 15:07:11 +00:00
f1f417485f
Bug 620931 part 5 - Enable omni.jar by default on xulrunner. r=bsmedberg
2011-03-24 15:28:43 +01:00
d982a02746
Bug 620931 part 4 - Fix resource://app/ to always point to the same as resource:///. r=bsmedberg
2011-03-24 15:28:31 +01:00
0cfcd44b08
Bug 620931 part 3 - Allow GRE and XUL application to use omni.jar independently. r=bsmedberg
...
We now store two independent locations for an omni.jar, allowing GRE/XRE and
XUL application to each have their own omni.jar. And since xulrunner setups
are very independent from the XUL applications, we implement support for both
omni.jar and non omni.jar cases in the same runtime, with the side effect of
allowing to switch from one to the other manually without rebuilding the
binaries.
We let the mozilla::Omnijar API handle both cases, so that callers don't need
too much work to support them.
We also make the preferences service load the same set of preferences in all
the various cases (unified vs. separate, omni.jar vs. no omni.jar).
The child process launcher for IPC is modified to pass the base directories
needed for the mozilla::Omnijar API initialization in the child process.
Finally, the startupcache file name canonicalization is modified to separate
APP and GRE resources.
2011-03-24 15:28:14 +01:00
60f6dbb70c
Bug 620931 part 2 - When building --with-libxul-sdk, use the right preferences directory. r=bsmedberg
2011-03-24 15:27:55 +01:00
2fbedf06e8
Bug 620931 part 1 - Use chrome manifest to register resource://gre-resources/. r=bsmedberg
2011-03-24 15:27:37 +01:00
d96dcb149e
Bug 644297 - Avoid _MD5*/_SHA* symbols not found during Shark universal build. r=ted
2011-03-24 15:27:00 +01:00
f1f22c4fac
Bug 643695 - Don't depend on libdbus-glib when not necessary. r=khuey
2011-03-24 15:26:58 +01:00
05d6cf9bb9
Bug 639175 - Do not throw an exception when select.size is set to 0. r=smaug
2011-03-24 12:49:19 +01:00
9c1ffb2319
Bug 630889 - textarea.rows and textarea.cols should be unsigned long and limited to only non-negative numbers greater than zero. r=smaug
2011-03-24 13:04:23 +01:00
ae216c0d07
Check for the |var| element rather than the |variable| element in nsHTMLCopyEncoder::IncludeInContext. (Bug 642716) r=bzbarsky
2011-03-18 14:02:44 -07:00
710e58f66c
Fix Windows optimized bustage from bug 636039 patch 15: use placement new in a loop instead of using placement new[].
2011-03-18 14:02:44 -07:00
b455d1fb83
Remove changes that should have been removed from bug 636039 patch 14 when I revised by 636039 patch 13.
2011-03-18 14:02:44 -07:00
4f8c7dcfa9
Bug 641366 - Remove Tiger's entry in CellRenderSettings. r=mstange
2011-03-18 15:10:01 +01:00
5aed3e0c4a
Fix bustage from bug 636039, patch 13, by not trying to access stub entries in kOffsetTable for the CSS_PROP_STUB_NOT_CSS entries. (Bug 636039)
2011-03-17 21:58:26 -07:00
a089e0afb6
Fix Windows bustage by using the alloca-inclusion incantation from jstracer.cpp. (Bug 636039)
2011-03-17 20:43:06 -07:00
cde7bf42ca
Remove support for the variable attribute on the HTML pre element. (Bug 642227) r=bzbarsky
2011-03-17 20:14:32 -07:00
8ff298d446
Make nsCachedStyleData store arrays of nsStyleStruct*, and use style struct IDs for faster access to those arrays. (Bug 639231, patch 3) r=bzbarsky
2011-03-17 20:14:32 -07:00
52152cb738
Fix indentation. (Bug 639231, patch 2) r=bzbarsky
2011-03-17 20:14:32 -07:00
c50f6e81a4
Keep reset and inherited style struct IDs separate. (Bug 639231, patch 1) r=bzbarsky
2011-03-17 20:14:31 -07:00
401e3dc9b6
Add poisoning for nsRuleData::mValueOffsets. (Bug 636039, patch 19) r=bzbarsky
...
I tested manually that after:
(a) removing the |ruleData.mValueOffsets[aSID] = 0;| in
nsRuleNode::WalkRuleTree
(b) removing the NS_ABORT_IF_FALSE(aRuleData->mValueOffsets[aSID] == 0,
...) from nsRuleNode::CheckSpecifiedProperties and
UnsetPropertiesWithoutFlags
that we crash dereferencing the poison address in a SetCoord call inside
nsRuleNode::ComputeTextResetData
2011-03-17 20:14:31 -07:00
c12115ab74
Make test_property_database.html give all the errors rather than throwing an exception for the first missing longhand property. (Bug 636039, patch 18) r=bzbarsky
...
The main fix is in the first of the three for loops modified, but the
new structure of that one seems like an improvement worth applying to
the other two.
2011-03-17 20:14:31 -07:00
a71cf7e9a9
Remove now unused nsRuleData* structs. (Bug 636039, patch 17) r=bzbarsky
2011-03-17 20:14:31 -07:00
71d08383cb
Remove Moz prefixes from names of subproperty tables. (Bug 636039, patch 16) r=bzbarsky
...
This fixes the FIXME introduced in patch 4.
2011-03-17 20:14:31 -07:00
ec0055aa6b
Instead of stack-allocating nsRuleData* structs in separate methods for each style struct, allocate an array of nsCSSValue using alloca. (Bug 636039, patch 15) r=bzbarsky
2011-03-17 20:14:31 -07:00
fce790e6bb
Add index and count data for the properties that need to be computed for each style struct. (Bug 636039, patch 14) r=bzbarsky
...
This adds a second occurrence of the ugliest pattern of nsCSSPropList.h
inclusion.
2011-03-17 20:14:31 -07:00
032ddb90c8
Make the three CSS_PROP_INCLUDE_NOT_CSS properties much more like normal CSS properties, and (importantly) give them property IDs in the longhand range. Replace CSS_PROP_INCLUDE_NOT_CSS with CSS_PROP_STUB_NOT_CSS for callers that need stubs. (Bug 636039, patch 13) r=bzbarsky
2011-03-17 20:14:31 -07:00
ed2a690b21
Remove now-unneeded parameters to COMPUTE_START_{INHERITED,RESET}. (Bug 636039, patch 12) r=bzbarsky
2011-03-17 20:14:31 -07:00
eaaf37ca08
Convert nsRuleNode::Compute*Data to property getters instead of accessing struct members. (Bug 636039, patch 11) r=bzbarsky
2011-03-17 20:14:31 -07:00
ec99680d49
Correct checks that were for the wrong pseudo-property (script-level vs. script-size-multiplier). (Bug 636039, patch 10) r=bzbarsky
2011-03-17 20:14:31 -07:00
612e7dd7b0
Remove unused *AtOffset methods. (Bug 636039, patch 9) r=bzbarsky
2011-03-17 20:14:31 -07:00
ad7b7cc2ed
Convert custom style rules to property getters instead of accessing struct members. (Bug 636039, patch 8) r=bzbarsky
2011-03-17 20:14:31 -07:00
07ad99a5d0
Convert attribute mapping functions to property getters instead of accessing struct members. (Bug 636039, patch 7) r=bzbarsky
2011-03-17 20:14:31 -07:00
84ca9844bf
Fix error in CSS vs. HTML precedence for the variable atttribute on the pre element. (Bug 636039, patch 6) r=bzbarsky
2011-03-17 20:14:30 -07:00
3a63faa598
Add nsRuleData::ValueForBackgroundColor, etc., methods for each CSS property. (Bug 636039, patch 5) r=bzbarsky
2011-03-17 20:14:30 -07:00
49daed21e9
Allow the method field in nsCSSPropList.h to be used with or without prefixes. (Bug 636039, patch 4) r=bzbarsky
2011-03-17 20:14:30 -07:00
f9e49f5a38
Don't bloat nsCSSTable for members that only need to exist in nsRuleDataTable. (Bug 636039, patch 3) r=bzbarsky
2011-03-17 20:14:30 -07:00
982559da8c
Correct comment describing nsRuleData::ValueFor. (Bug 636039, patch 2) r=bzbarsky
...
nsRuleData::ValueFor does not return null; it checks conditions that
would cause it to do so with NS_ABORT_IF_FALSE. Callers are required to
check mSIDs.
2011-03-17 20:14:30 -07:00
e11dc52475
Bug 641856. column-rule needs to work even when overflow is not visible. r=roc
2011-03-17 13:41:52 -04:00
cc19a1ceb0
Bug 577002 Part 3: Make various ImportRule functions non-virtual and cleanup some return types. r=bzbarsky
2011-03-17 13:41:52 -04:00
Patrick McManus
Mounir Lamouri
L. David Baron
Jonathan Kew
Mike Hommey
Boris Zbarsky
Craig Topper