For posterity, this issue was superceded by
http://gwt-code-reviews.appspot.com/826802/show
On 2010/08/23 20:29:44, conroy wrote:
Discussed in person with jat and scottb:
Tests are failing for this change because the ThreadLocal assumptions
don't
actually hold under JUnit. As a result, I'm
Discussed in person with jat and scottb:
Tests are failing for this change because the ThreadLocal assumptions
don't actually hold under JUnit. As a result, I'm going to be dropping
this change in favor of more explicitly caching based on the generator
context lifecycle.
On 2010/08/20 20:46:05,
LGTM, though your format settings still seem to be off.
http://gwt-code-reviews.appspot.com/766803/diff/29001/30006
File
user/src/com/google/gwt/resources/rebind/context/AbstractResourceContext.java
(right):
http://gwt-code-reviews.appspot.com/766803/diff/29001/30006#newcode87
LGTM, though your format settings still seem to be off.
I noticed that M3 was branched and this hasn't made it into svn yet--is
there any chance it will get merged into M3 before its released?
Thanks,
Stephen
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
We're really trying to reduce the number of moving parts in order to get M3
out the door, so this might not get cherry picked in. Is there a specific
reason you wouldn't want to build from the revision (or subsequent
revisions) that includes this fix?
On Fri, Aug 20, 2010 at 9:39 PM, Stephen
Is there a specific reason you wouldn't want to build from the
revision (or subsequent revisions) that includes this fix?
Oh, not really--just convenience. I have a few projects/libraries with
GWT Mx versions and it would be nice to have them all on M3 and just
work. Especially since with
These caches contain types which have a circular reference pattern
with the
TypeOracle. If we make a hard reference to the cache here, we end up
pinning the
TypeOracle.
You could break hard reference on TypeOracle for example by using weak
reference on method from
http://gwt-code-reviews.appspot.com/766803/show
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
http://gwt-code-reviews.appspot.com/766803/show
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
http://gwt-code-reviews.appspot.com/766803/diff/1/2
File dev/core/src/com/google/gwt/core/ext/typeinfo/TypeOracle.java
(right):
http://gwt-code-reviews.appspot.com/766803/diff/1/2#newcode499
dev/core/src/com/google/gwt/core/ext/typeinfo/TypeOracle.java:499: *
manage static state.
On
Still need answer on JJS - i18n clear cache.
http://gwt-code-reviews.appspot.com/766803/diff/1/3
File dev/core/src/com/google/gwt/dev/javac/CompilationStateBuilder.java
(right):
http://gwt-code-reviews.appspot.com/766803/diff/1/3#newcode223
On 2010/08/19 16:50:01, scottb wrote:
Still need answer on JJS - i18n clear cache.
http://gwt-code-reviews.appspot.com/766803/diff/1/3
File
dev/core/src/com/google/gwt/dev/javac/CompilationStateBuilder.java
(right):
http://gwt-code-reviews.appspot.com/766803/diff/1/3#newcode223
http://gwt-code-reviews.appspot.com/766803/show
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
On 2010/08/19 20:05:46, conroy wrote:
Rietveld's interdiff handling is absurdly bugged.
ClearStaticData is no longer deleted, contrary to what it shows on the
summary. The only change to CompilationStateBuilder vs. HEAD is a new
comment clarifying the resourceContentCache/keepAliveLatest
LGTM if it LGTjat.
http://gwt-code-reviews.appspot.com/766803/show
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
LGTM
Does this fix the pinning of CCL as well?
http://gwt-code-reviews.appspot.com/766803/diff/1/7
File user/src/com/google/gwt/i18n/rebind/ResourceFactory.java (right):
http://gwt-code-reviews.appspot.com/766803/diff/1/7#newcode64
user/src/com/google/gwt/i18n/rebind/ResourceFactory.java:64:
No. Still tracking down the CCL JNI reference problem, but these leaks
are affecting basically all apps whereas that one is harder to repro.
On 2010/08/18 21:31:40, jat wrote:
LGTM
Does this fix the pinning of CCL as well?
http://gwt-code-reviews.appspot.com/766803/diff/1/7
File
http://gwt-code-reviews.appspot.com/766803/diff/1/8
File
user/src/com/google/gwt/resources/rebind/context/AbstractResourceContext.java
(right):
http://gwt-code-reviews.appspot.com/766803/diff/1/8#newcode50
user/src/com/google/gwt/resources/rebind/context/AbstractResourceContext.java:50:
new
http://gwt-code-reviews.appspot.com/766803/diff/1/8
File
user/src/com/google/gwt/resources/rebind/context/AbstractResourceContext.java
(right):
http://gwt-code-reviews.appspot.com/766803/diff/1/8#newcode50
user/src/com/google/gwt/resources/rebind/context/AbstractResourceContext.java:50:
new
http://gwt-code-reviews.appspot.com/766803/diff/1/2
File dev/core/src/com/google/gwt/core/ext/typeinfo/TypeOracle.java
(right):
http://gwt-code-reviews.appspot.com/766803/diff/1/2#newcode3
dev/core/src/com/google/gwt/core/ext/typeinfo/TypeOracle.java:3: *
Please tweak your Eclipse auto-format
20 matches
Mail list logo
