4b47b10a77
Move buildMethodInvocation to BodyBuilder.
...
R=paulberry@google.com
Review-Url: https://codereview.chromium.org/2906333002 .
2017-05-31 15:30:54 +02:00
cdfd07b9d3
Don't read platform files during testing, and avoid trimming broken programs.
...
R=johnniwinther@google.com
Review-Url: https://codereview.chromium.org/2912353002 .
2017-05-31 15:03:30 +02:00
8ea78cbc3f
Avoid using Invalid{Statement,Expression,Initializer}.
...
R=paulberry@google.com
Review-Url: https://codereview.chromium.org/2906253004 .
2017-05-31 14:53:44 +02:00
29dac16eba
Clean up shared names.
...
R=johnniwinther@google.com , paulberry@google.com
Review-Url: https://codereview.chromium.org/2909013002 .
2017-05-31 11:08:30 +02:00
b89e8cd1a1
Fix inferInvocation for top level inference.
...
When inferring the type of an invocation for top level inference, if
type parameters are not specified, then it is an error. So we should
not recurse into subexpressions (even if inferenceNeeded is `true`).
This avoids confusion by making it clear to the user that we don't
take subexpressions into account when doing top level type inference,
even when recovering from type inference errors.
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2911323002 .
2017-05-30 14:43:06 -07:00
f3a79c4e4e
update fasta StringToken.toString to make formatter happy
...
R=paulberry@google.com
Review-Url: https://codereview.chromium.org/2915593004 .
2017-05-30 16:55:08 -04:00
0017ffd872
Change the approach of front_end top level inference to match the spec.
...
Specifically, we can't walk all the subexpressions when doing top
level inference, or we will pick up bogus dependencies. This means
that we can't determine the dependencies during BodyBuilder execution;
we have to do it in a follow-up pass once we have the AST.
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2913743003 .
2017-05-30 13:42:12 -07:00
0dd81884bf
Perform MixinFullResolution only for source libraries.
...
R=ahe@google.com , kmillikin@google.com , paulberry@google.com , sigmund@google.com
BUG=
Review-Url: https://codereview.chromium.org/2912503003 .
2017-05-30 09:33:26 -07:00
9497f5d3f8
Remove accessor properties not used by Fasta.
...
R=johnniwinther@google.com , paulberry@google.com
Review-Url: https://codereview.chromium.org/2908123002 .
2017-05-30 13:08:31 +02:00
4cd8e437f6
Various constant checks.
...
R=johnniwinther@google.com , paulberry@google.com
Review-Url: https://codereview.chromium.org/2908113002 .
2017-05-30 12:44:57 +02:00
6a809aa432
Add type inference logic for "for-in" loops.
...
R=ahe@google.com
Review-Url: https://codereview.chromium.org/2908973002 .
2017-05-29 07:35:46 -07:00
0eae9218f4
Complain about duplicated names.
...
R=paulberry@google.com
Review-Url: https://codereview.chromium.org/2902113005 .
2017-05-29 09:45:05 +02:00
767d5ecbfd
Revert "improve fasta unterminated string recovery"
...
This reverts commit 6907345fb9https://codereview.chromium.org/2907093002 .
2017-05-28 23:40:08 -04:00
6907345fb9
improve fasta unterminated string recovery
...
* enhance handling of unterminated strings
* update ast builder and string validator to handle unterminated strings
Review-Url: https://codereview.chromium.org/2906193004 .
2017-05-28 23:35:32 -04:00
9a64339646
Change dependency walker's Node.dependencies to a static getter.
...
This prevents it from being accidentally overridden in a derived class.
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2907023003 .
2017-05-27 21:27:21 -07:00
e4042e5de2
Implement type inference for await expressions in front_end.
...
This required implementing FutureOr rules in
type_constraint_gatherer.dart. It also required fixing an incorrect
type in analyzer's mock SDK, which had some follow on effects on
analyzer unit tests.
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2904263003 .
2017-05-27 07:30:46 -07:00
5d69e056d3
Add type inference support for "call" invocations of function expressions.
...
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2911683003 .
2017-05-26 13:34:42 -07:00
d859cc6e4e
Add type inference logic for map literals
...
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2905353002 .
2017-05-26 13:09:53 -07:00
3c730528fc
Treat type inference context dynamic as equivalent to an empty context.
...
This replicates analyzer behavior.
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2903153006 .
2017-05-26 11:26:35 -07:00
8928329339
improve fasta interpolation recovery
...
* improve fasta interpolation recovery by adding closing brace
for unterminated string
* update analyzer scanner test because fasta does not support
does not support \r line endings
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2910583002 .
2017-05-26 08:43:48 -04:00
f60bad7fbc
Fix type inference of redirecting factory constructor invocations.
...
Their types should be inferred based on the initial target of the
invocation, not the redirected target.
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2904133002 .
2017-05-25 17:19:20 -07:00
ee6fd5e1a3
Perform type inference on redirecting initializers (C() : this(...);).
...
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2905163002 .
2017-05-25 14:47:53 -07:00
069e1bda1c
Correctly infer the return context while compiling a factory constructor.
...
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2907643003 .
2017-05-25 14:17:59 -07:00
30b51f7403
Files are added with absolute URIs, but invalidated with file URIs.
...
R=ahe@google.com , paulberry@google.com , sigmund@google.com
BUG=
Review-Url: https://codereview.chromium.org/2907673002 .
2017-05-25 13:42:18 -07:00
cefc4cc768
Perform type inference on explicitly typed fields.
...
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2905123002 .
2017-05-25 13:22:26 -07:00
560cf61b3b
Add type inference logic for cascade expressions.
...
R=ahe@google.com , scheglov@google.com
Review-Url: https://codereview.chromium.org/2903243002 .
2017-05-25 09:30:26 -07:00
9b4bc8b0ff
Read file names from libraries.json, not URIs.
...
R=danrubel@google.com , scheglov@google.com
Review-Url: https://codereview.chromium.org/2906673002 .
2017-05-25 16:20:05 +02:00
bcb2ea962b
Complain about function literals not being constant expressions.
...
R=paulberry@google.com
Review-Url: https://codereview.chromium.org/2899343002 .
2017-05-25 16:02:30 +02:00
27761fd369
Read SDK and patches from a JSON file.
...
R=sigmund@google.com
Review-Url: https://codereview.chromium.org/2895983002 .
2017-05-25 15:57:57 +02:00
8a92a2a79c
Infer the types of initializing formals from the corresponding fields.
...
This inference is enabled for both strong and non-strong mode, since
it is mandated by the Dart 1.0 spec.
R=ahe@google.com , scheglov@google.com
Review-Url: https://codereview.chromium.org/2908453002 .
2017-05-25 04:47:44 -07:00
54d6e3a1c5
cleanup fasta token classes
...
* replace fasta.SymbolToken --> analyzer.Token
* replace fasta.BeginGroupToken --> analyzer.BeginToken
* replace fasta.SymbolToken.eof --> analyzer.Token.eof
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2899043006 .
2017-05-25 07:26:50 -04:00
97d587520c
Add LimitedBinaryPrinter, tests and switch incremental generator to it.
...
R=ahe@google.com , kmillikin@google.com , paulberry@google.com , sigmund@google.com
BUG=
Review-Url: https://codereview.chromium.org/2896493002 .
2017-05-24 16:17:28 -07:00
2b8fd9dee8
Transform async code only for source libraries.
...
R=ahe@google.com , kmillikin@google.com , paulberry@google.com , sigmund@google.com
BUG=
Review-Url: https://codereview.chromium.org/2905463002 .
2017-05-24 15:51:18 -07:00
dff793d05f
update lazy assignment operators
...
and remove scanLazyAssignmentOperators flag
R=ahe@google.com
Review-Url: https://codereview.chromium.org/2903063003 .
2017-05-24 17:05:02 -04:00
a21b9e5638
Remove AstFactory from the front end.
...
R=ahe@google.com , scheglov@google.com
Review-Url: https://codereview.chromium.org/2904673003 .
2017-05-24 11:28:12 -07:00
9fd47a08d1
Address follow up comments
...
R=ahe@google.com
Review-Url: https://codereview.chromium.org/2902643002 .
2017-05-24 11:20:40 -07:00
5f8bc4feb2
fasta: revert check for null fileUri in the error reporting of the verifier.
...
(as discussed in https://codereview.chromium.org/2894283002 , we are reverting
this because we don't know how to reproduce this issue anymore. When it comes up
again we can either fix the issue or re-revert and include an explanation).
R=ahe@google.com
Review-Url: https://codereview.chromium.org/2902153003 .
2017-05-24 11:17:15 -07:00
6e981d0cbf
enable analyzer scanner adapter generic method support
...
This turns on generic method support in the scanner adapter layer
and updates the token.offset when the analyzer scanner
is being used to parse generic method comments.
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2900963009 .
2017-05-24 13:37:52 -04:00
6b100c007b
Use declared return type for inference of return and yield statements.
...
R=ahe@google.com
Review-Url: https://codereview.chromium.org/2899073004 .
2017-05-24 09:28:26 -07:00
b9b949f436
Use FutureOr as the return context in async closures.
...
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2898373004 .
2017-05-24 08:24:13 -07:00
e642722f42
Implement subtyping rules for FutureOr in kernel.
...
These rules are only enabled when in strong mode, so existing Dart 1.0
functionality should be unaffected.
R=kmillikin@google.com , scheglov@google.com
Review-Url: https://codereview.chromium.org/2900113002 .
2017-05-24 06:14:22 -07:00
61d855bb47
Prepare for development branch by updating externally-used API.
...
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2898183002 .
2017-05-24 14:32:45 +02:00
9ed1ba2c12
Temporarily uncomment logEvent to make build bot green.
...
Review-Url: https://codereview.chromium.org/2901273002 .
2017-05-24 12:04:19 +02:00
f5012090f9
New parser event needed to check case expressions.
...
R=paulberry@google.com
Review-Url: https://codereview.chromium.org/2900993002 .
2017-05-24 11:48:42 +02:00
54bd8a79c4
Fix top level type inference in the event that there are circularities.
...
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2900193002 .
2017-05-23 14:01:46 -07:00
b16b0e8edb
Fix type inference of yield* statements
...
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2898313002 .
2017-05-23 13:10:40 -07:00
a3e11d2e6e
Always record inference of implicit function expression params, even if dynamic.
...
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2904653002 .
2017-05-23 12:32:49 -07:00
0d31c8c524
Populate TypeInferrerImpl.closureContext more frequently.
...
Previously, we only used a ClosureContext in the case where the
innermost function being inferred was a function expression. This
meant that we couldn't use the declared return type as the context for
`return` or `yield` statements in any other sort of function. As a
first step towards fixing that, this CL makes the following changes:
- The entry point to type inference is now a separate
TypeInferrer.inferFunctionBody method rather than re-using
TypeInferrer.inferStatement; this gives us an opportunity to set up
the context.
- Some of the logic from KernelFunctionExpression._inferExpression
(particularly to do with asynchronous and generator functions) is
moved to ClosureContext so that it can be used for other sorts of
functions.
- The only time TypeInferrerImpl.closureContext is ever `null` is now
when inferring a top level or field initializer expression, so we
can eliminate some null checks from
KernelReturnStatement._inferStatement and
KernelYieldStatement._inferStatement.
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2902683004 .
2017-05-23 11:21:45 -07:00
77d4eb609b
Use LibraryIndex in CoreTypes instead of extending it.
...
A smaller step this time - change only clients, but keep implementation
the same for now.
Not all patched SDKs have all classes and procedures, so I have to make
some them lazy. All of them could be done for consistency, but this
might slightly affect performance for often used dart:core classes.
Also it's a bit of mess now.
Maybe group these classes into logical structures, coreTypes.core.list,
coreTypes.async.future, coreTypes.async.futureMicrotask, etc. Something
like it was before this change, but fixed types and names instead of
more flexible (but in general slower) strings.
R=ahe@google.com , kmillikin@google.com , paulberry@google.com , sigmund@google.com
BUG=
Review-Url: https://codereview.chromium.org/2890933002 .
2017-05-23 09:30:37 -07:00
b14be9d1ce
Add a test for including all libraries from changed to the entry point.
...
R=ahe@google.com , paulberry@google.com , sigmund@google.com
BUG=
Review-Url: https://codereview.chromium.org/2899883002 .
2017-05-22 14:44:47 -07:00
Peter von der Ahé