http://gwt-code-reviews.appspot.com/1518803/diff/1/dev/core/src/com/google/gwt/dev/CompileModule.java
File dev/core/src/com/google/gwt/dev/CompileModule.java (right):
http://gwt-code-reviews.appspot.com/1518803/diff/1/dev/core/src/com/google/gwt/dev/CompileModule.java#newcode194
dev/core/src/com/google/gwt/dev/CompileModule.java:194: Map<String,
Set<String>> unitsInArchives = new HashMap<String, Set<String>>();
On 2011/08/11 15:29:46, jbrosenberg wrote:
I'm a little confused by this comment.
When you say "session", is it referring specifically to a dev mode
session? Or
could it also apply to a web-mode compile? In which case what's a
session?
What I mean is an invocation of CompileModule, which is a standalone
program. Updated to make that more clear.
What's the significance of units being written to the classpath?
http://gwt-code-reviews.appspot.com/1518803/diff/1/dev/core/src/com/google/gwt/dev/CompileModule.java#newcode197
dev/core/src/com/google/gwt/dev/CompileModule.java:197: // modules
compiled in the same session.
On 2011/08/11 15:29:46, jbrosenberg wrote:
maybe a more specific name ("newlyCompiledModules")?
Done.
http://gwt-code-reviews.appspot.com/1518803/diff/1/dev/core/src/com/google/gwt/dev/CompileModule.java#newcode231
dev/core/src/com/google/gwt/dev/CompileModule.java:231:
On 2011/08/11 15:29:46, jbrosenberg wrote:
session?
Done.
http://gwt-code-reviews.appspot.com/1518803/diff/1/dev/core/src/com/google/gwt/dev/CompileModule.java#newcode238
dev/core/src/com/google/gwt/dev/CompileModule.java:238:
On 2011/08/11 15:29:46, jbrosenberg wrote:
Is this really the right comment for this entire for loop?
Done.
http://gwt-code-reviews.appspot.com/1518803/diff/1/dev/core/src/com/google/gwt/dev/cfg/ModuleDef.java
File dev/core/src/com/google/gwt/dev/cfg/ModuleDef.java (right):
http://gwt-code-reviews.appspot.com/1518803/diff/1/dev/core/src/com/google/gwt/dev/cfg/ModuleDef.java#newcode137
dev/core/src/com/google/gwt/dev/cfg/ModuleDef.java:137: new
LinkedHashMap<String, Class<? extends Linker>>();
On 2011/08/11 15:29:46, jbrosenberg wrote:
whitespace? Here and below?
Done.
http://gwt-code-reviews.appspot.com/1518803/diff/1/dev/core/src/com/google/gwt/dev/cfg/ModuleDefLoader.java
File dev/core/src/com/google/gwt/dev/cfg/ModuleDefLoader.java (right):
http://gwt-code-reviews.appspot.com/1518803/diff/1/dev/core/src/com/google/gwt/dev/cfg/ModuleDefLoader.java#newcode232
dev/core/src/com/google/gwt/dev/cfg/ModuleDefLoader.java:232: }
On 2011/08/11 15:29:46, jbrosenberg wrote:
whitespace?
Done.
http://gwt-code-reviews.appspot.com/1518803/diff/1/dev/core/src/com/google/gwt/dev/cfg/ModuleDefLoader.java#newcode241
dev/core/src/com/google/gwt/dev/cfg/ModuleDefLoader.java:241: }
On 2011/08/11 15:29:46, jbrosenberg wrote:
will this add inherited modules from multiple levels in the hierarchy?
Or only
first level inheritance?
nestedLoad is invoked recursively, if that's what you mean.
It seems like it would be relevant to detecting
whether a module's compilation units are already loaded, etc.
Essentially, I just moved a Set<> out of ModuleDefLoader we were keeping
into ModuleDef because I wanted to be able to query the set after the
module had been loaded to fix the problem in CompileModule.
Compile Module wants to exclude units from an archive if they are a part
of an inherited module, but the inheritance information was lost after
load time.
http://gwt-code-reviews.appspot.com/1518803/
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
