04c5a5f199
Clean up hints
...
R=scheglov@google.com
Review URL: https://codereview.chromium.org/1382403002 .
2015-10-05 09:20:44 -07:00
0ff03b00b8
Add the 'docRange' property and use it to compute documentation comment.
...
R=brianwilkerson@google.com
BUG=
Review URL: https://codereview.chromium.org/1376473007 .
2015-10-02 15:13:30 -07:00
66a122a9db
Stop defensively copying AST structures before resolution
...
R=scheglov@google.com
Review URL: https://codereview.chromium.org/1383043002 .
2015-10-02 12:38:18 -07:00
28e7dbf876
Don't make defensive copies of 'definedNames'.
...
This change reduces total analysis time by 13%.
For analyzer + analysis_server.
For the third full analysis, i.e. not cold VM.
R=brianwilkerson@google.com
BUG=
Review URL: https://codereview.chromium.org/1370993009 .
2015-10-02 10:15:24 -07:00
a8570d9e7c
fixes strong mode assignability
...
this fixes https://github.com/dart-lang/dev_compiler/issues/356
The issue here is that StrongTypeSystemImpl has toType and fromType backwards from how they are defined on the base class.
The way it works is to read it as "is the first type assignable to the second type"?
(I can see how this may have happened: the base class uses "leftType" and "rightType", which refers to the parameter order, not the order they would appear in an assignment expression.)
R=leafp@google.com
Review URL: https://codereview.chromium.org//1380963004 .
2015-10-01 15:40:50 -07:00
0f30af628a
Fix computation of errors to not run extra tasks
...
R=scheglov@google.com
Review URL: https://codereview.chromium.org//1370323003 .
2015-09-30 11:55:17 -07:00
1d28f99524
Fix for 'implemented' notification after an incremental change.
...
R=brianwilkerson@google.com , paulberry@google.com
BUG=
Review URL: https://codereview.chromium.org//1374943005 .
2015-09-30 11:08:41 -07:00
20e1f91fbb
fix inference of hashCode with library prefix
...
fixes https://github.com/dart-lang/dev_compiler/issues/349
R=leafp@google.com
Review URL: https://codereview.chromium.org//1374813002 .
2015-09-28 16:02:10 -07:00
253fc292cb
fix inference of object members on library prefix
...
See https://github.com/dart-lang/dev_compiler/issues/349 for context.
The prefix fix handled `toString()` but not `library_prefix.toString()`
R=brianwilkerson@google.com , leafp@google.com
Review URL: https://codereview.chromium.org//1374773003 .
2015-09-28 14:40:20 -07:00
ac282f8a1d
Fix for the libraries to parts map.
...
R=brianwilkerson@google.com
BUG=
Review URL: https://codereview.chromium.org//1377613002 .
2015-09-28 13:21:41 -07:00
9be2107f39
Improve the handling of inference for instance fields in the task model
...
strong mode inference. This CL adds instance fields to the set of variables
for which dependencies are computed, and re-resolves each instance field
initializer using type information inferred for the static variables upon
which it depends.
This CL also stops static variables with existing type information from having
their types overwritten.
This CL ports over a number of the DDC inference tests, more remain to be ported.
There is still at least one outstanding issue with instance variable inference,
see #354 (and the two tests marked fail added in this CL).
BUG=
R=brianwilkerson@google.com
Review URL: https://codereview.chromium.org//1370793002 .
2015-09-28 10:57:04 -07:00
9ac032d545
infer field formal parameter types
...
this fixes https://github.com/dart-lang/dev_compiler/issues/346
R=brianwilkerson@google.com
Review URL: https://codereview.chromium.org//1371813002 .
2015-09-28 09:20:20 -07:00
7bd3db6d74
analyzer strong mode: infer final field from initializer if super getters are dynamic
...
fixes https://github.com/dart-lang/dev_compiler/issues/348
R=brianwilkerson@google.com , leafp@google.com
Review URL: https://codereview.chromium.org//1367183004 .
2015-09-25 17:49:55 -07:00
c09b338e25
fix inference of object members when a method has no target
...
we incorrectly thought `toString()` was a call to `Object.toString()`
See https://github.com/dart-lang/dev_compiler/issues/349 for more context
R=brianwilkerson@google.com
Review URL: https://codereview.chromium.org//1368793004 .
2015-09-25 17:22:56 -07:00
6bab7d9cbc
Use Element.nameLength where possible.
...
...or don't use it when there is a better alternative.
R=brianwilkerson@google.com
BUG=
Review URL: https://codereview.chromium.org//1372673003 .
2015-09-25 15:03:46 -07:00
3924b15d85
Remove ElementLocator.locateWithOffset altogether.
...
It was needed only for the old Angular implementation support, which
overlayed its own Element subclasses into normal Dart AST, for example
for string literals - names of components.
As we don't support this anymore, there is no need for the method.
R=brianwilkerson@google.com
BUG=
Review URL: https://codereview.chromium.org//1370833002 .
2015-09-25 14:57:15 -07:00
6c029f9cb1
Issue 24440. Fix for renaming libraries using 'part of' directives.
...
R=brianwilkerson@google.com
BUG= https://github.com/dart-lang/sdk/issues/24440
Review URL: https://codereview.chromium.org//1364853007 .
2015-09-25 14:24:05 -07:00
4f8649f1b3
Issue 24442. Fix for renaming library names with spaces.
...
R=brianwilkerson@google.com
BUG= https://github.com/dart-lang/sdk/issues/24442
Review URL: https://codereview.chromium.org//1370793003 .
2015-09-25 13:55:13 -07:00
327412cbad
Add Element.nameLength getter and specialize it for LibraryElement.
...
R=brianwilkerson@google.com
BUG= https://github.com/dart-lang/sdk/issues/24442
Review URL: https://codereview.chromium.org//1364583006 .
2015-09-25 13:35:49 -07:00
59eb287389
[analyzer] copy generic type args when inferring a new function type for an ExecutableElement
...
This matches what resolver.dart visitMethodDeclaration does when it creats the FunctionTypeImpl for the method. So we want to preserve these generic type args when we're building the new inferred function type, otherwise we'll fail to substitute them later on.
Context: https://github.com/dart-lang/dev_compiler/issues/342#issuecomment-142748853
R=brianwilkerson@google.com
Review URL: https://codereview.chromium.org//1364353003 .
2015-09-25 10:03:10 -07:00
2b84d9ada6
Merge pull request #24341 from srawlins/patch-1
...
Fix ConstantEvaluator's doc comment
2015-09-25 09:54:51 -07:00
cac248765b
Remove CompilationUnit name requirement
...
Some of our language tests have '.lib' extensions for dart files that are imported.
R=brianwilkerson@google.com
Review URL: https://codereview.chromium.org//1372433003 .
2015-09-24 15:51:08 -07:00
5dd6a543e7
Plugin manifest parsing.
...
R=brianwilkerson@google.com
Review URL: https://codereview.chromium.org//1366023002 .
2015-09-24 14:53:15 -07:00
83a2449062
Infer setter parameter types in strong mode
...
This change also decouples setters and getters, which matches what DDC currently does. The benefit is they're inferred as independent methods, which allows code reuse between method/getter/setters. It also prevents ordering issues that might occur otherwise (i.e. if getter inference considers setter's type, and setter inference considers the getter's type).
More context: https://github.com/dart-lang/dev_compiler/issues/342#issuecomment-142713453
R=brianwilkerson@google.com
Review URL: https://codereview.chromium.org//1359243003 .
2015-09-24 13:21:04 -07:00
9535715f92
Fix handling of empty .analysis_options files and test this case
...
Fixes #24431
R=brianwilkerson@google.com
Review URL: https://codereview.chromium.org//1369573003 .
2015-09-24 12:38:53 -07:00
72ee80467a
ErrorHandler typedef fix.
...
Lo and behold this harkens back to lint request #1 (https://github.com/dart-lang/linter/issues/1 ). Could have saved me a headache!
R=brianwilkerson@google.com
Review URL: https://codereview.chromium.org//1367723003 .
2015-09-23 14:39:07 -07:00
7ff04709e7
Plugin config cleanup.
...
R=scheglov@google.com
Review URL: https://codereview.chromium.org//1366663002 .
2015-09-23 10:44:02 -07:00
e2ec118554
Plugin config format validation (and tests).
...
R=brianwilkerson@google.com
Review URL: https://codereview.chromium.org//1362033002 .
2015-09-23 09:51:29 -07:00
ff86501c85
This adds support for a DDC special case of inline JS. It also addresses an issue from an earlier CL, makes some readability improvements, and eliminates some now-redundant TODOs.
...
BUG=
R=scheglov@google.com , vsm@google.com
Review URL: https://codereview.chromium.org//1355393003 .
2015-09-23 08:58:04 -07:00
27a8bbd415
Reset Source modification stamp to -1 if it does not exist.
...
R=brianwilkerson@google.com
BUG=
Review URL: https://codereview.chromium.org//1361003002 .
2015-09-23 08:20:12 -07:00
45a81cbab4
Fix a repeated spelling error
...
R=scheglov@google.com
Review URL: https://codereview.chromium.org//1361833003 .
2015-09-23 07:44:05 -07:00
f4c3b30d8a
Basic plugin config parsing.
...
R=brianwilkerson@google.com
Review URL: https://codereview.chromium.org//1364493004 .
2015-09-22 17:06:14 -07:00
4fe661995d
Add additional strong mode inference. This adds in all the remaining inference cases from the strong mode RestrictedStaticTypeAnalyzer.
...
BUG=
R=brianwilkerson@google.com
Review URL: https://codereview.chromium.org//1352763004 .
2015-09-22 14:03:08 -07:00
163ad75127
Fix status pages when running new task model
...
R=scheglov@google.com
Review URL: https://codereview.chromium.org//1351243003 .
2015-09-22 11:03:56 -07:00
3aabaa96e6
Fix for a stupid copy/paste bug in the incremental resolver.
...
R=brianwilkerson@google.com
BUG=
Review URL: https://codereview.chromium.org//1360583003 .
2015-09-21 09:17:35 -07:00
a25cc8129a
Limit cache invalidation by the updated source bounds.
...
R=brianwilkerson@google.com
BUG=
Review URL: https://codereview.chromium.org//1355853005 .
2015-09-18 18:11:54 -07:00
fcb0c319da
Fix for updating HINTS during incremental resolution with TM.
...
The key change is not updating LIBRARY_UNIT_ERRORS in IncrementalResolver.
Otherwise it becomes VALID and we don't recompute it, so don't recompute HINTS.
R=brianwilkerson@google.com
BUG=
Review URL: https://codereview.chromium.org//1355503004 .
2015-09-18 13:41:09 -07:00
14b726f3e6
Partially fix the cache invalidation problem during incremental resolution.
...
It still takes 90% of total incremental resolution time.
But at least it's not 10 seconds anymore, just 300ms in Analyzer + Server, or 80ms in pure Analyzer.
R=brianwilkerson@google.com
BUG=
Review URL: https://codereview.chromium.org//1350423002 .
2015-09-18 12:02:30 -07:00
f912eabcbf
Make type system changes backwards compatible
...
This CL eliminates the additional typeSystem arguments added to the API by https://codereview.chromium.org/1329743005/ wherever possible, instead projecting out the typeSystem from the context as available. Where not possible, the additional argument is made optional, defaulting to the standard TypeSystemImpl. I believe that the result of this should be backwards compatible for clients (but not implementors) of the API.
BUG=
R=brianwilkerson@google.com , paulberry@google.com
Review URL: https://codereview.chromium.org//1355773002 .
2015-09-18 09:59:12 -07:00
d7cd11f86a
allow dart:* libs to implement otherwise disallowed types (e.g. int)
...
all current and experimental Dart implementations (dart2js, vm, fletch, ddc) get an error in their SDK from this. The CL fixes it by allowing dart:* libraries to implement the disallowed types, like int/double/String/bool.
R=brianwilkerson@google.com
Review URL: https://codereview.chromium.org//1346923007 .
2015-09-18 07:11:44 -07:00
594fe93422
Support for NAVIGATION/OCCURRENCES notifications for any source (such as HTML).
...
R=brianwilkerson@google.com
BUG=
Review URL: https://codereview.chromium.org//1343283004 .
2015-09-17 12:37:36 -07:00
f4426b109b
Do not emit semicolons for await expressions in ToSourceVisitor
...
Closes #24364
BUG=
R=brianwilkerson@google.com , scheglov@google.com
Review URL: https://codereview.chromium.org//1345403002 .
2015-09-17 11:20:45 -07:00
ff95a65900
Reset AnalysisDriver on priority sources changes.
...
It turned out that our WorkManager(s) don't have the problem we discussed.
We remove targeted results from queues only when they are VALID or ERROR.
So, reset is safe to do.
R=brianwilkerson@google.com
BUG=
Review URL: https://codereview.chromium.org//1350993002 .
2015-09-17 07:57:51 -07:00
a19bd39da8
First cut at abstracting over the type system. This factors out uses of isAssignableTo and isSubtypeOf into calls through a type system object. Currently there is only one kind of type system, which just falls over to the existing implementations.
...
BUG=
R=brianwilkerson@google.com , paulberry@google.com
Review URL: https://codereview.chromium.org//1329743005 .
2015-09-16 14:55:59 -07:00
aab6cf0b35
Produce LINE_INFO for HTML files.
...
R=brianwilkerson@google.com
BUG=
Review URL: https://codereview.chromium.org//1347223002 .
2015-09-16 13:28:16 -07:00
4e18780ded
Tweak for incremental result updates.
...
The result itself is not invalidated, but everything dependent on it is.
R=brianwilkerson@google.com
BUG=
Review URL: https://codereview.chromium.org//1339603004 .
2015-09-14 13:21:16 -07:00
004839fc86
Fixes for the new task model
...
R=scheglov@google.com
Review URL: https://codereview.chromium.org//1342903002 .
2015-09-14 13:16:08 -07:00
e878562d35
Add an extension point for contributing to HTML_ERRORS.
...
R=brianwilkerson@google.com
BUG=
Review URL: https://codereview.chromium.org//1341723003 .
2015-09-14 12:00:39 -07:00
3ac37fd7ea
Add ReentrantSynchronousStream and use it for cache invalidation events.
...
Also include the (just invalidated) value into InvalidatedResult.
R=brianwilkerson@google.com
BUG=
Review URL: https://codereview.chromium.org//1342543007 .
2015-09-14 11:23:16 -07:00
e3aca49123
Fix ConstantEvaluator's doc comment
...
ConstantEvalulator's doc comment, when parsed as Markdown and rendered at [dartdocs.org](http://www.dartdocs.org/documentation/analyzer/0.26.1%2B3/index.html#analyzer/analyzer.ConstantEvaluator ), looks terrible.
I think there was a valiant effort to make it render ok, what with the <blockquote>, but in any case, it was just a blob. This fixes a few things:
* One summary sentence first paragraph.
* Make the blockquote and list into an actual blockquote and an actual list.
* Fix some un-subscripted subscripts.
* Render the HTML that was attempted by making certain list items _blocks_, by starting out with the span tags.
2015-09-13 15:05:20 -07:00
Brian Wilkerson