* The script had stopped working due to a pair of changes in Dart.
First, building Dart no longer produces a packages directory to use
as a package-root.
Second, generated Dart files are now formatted and the formatting
code relies on invoking Dart from the Dart SDK in order to find
dartfmt. (Specifically: invoking Dart from the build directory does
not work because there is no dartfmt there, and invoking Dart via
sdk/bin/dart from the SDK repository does not work because it is a
shell script that invokes Dart from the build directory.)
* Enable the test that the generated summary code has not changed.
* Regenerate the generated summary code.
BUG=
R=paulberry@google.com, whesse@google.com
Review-Url: https://codereview.chromium.org/2997703002 .
The API for pkg/front_end is still in flux so we want to avoid having
any packages depend directly on it (other than analyzer and
back-ends).
This CL re-exports some of the critical pieces of front_end needed by
analyzer clients so that those clients can access them via analyzer,
without having to directly depend on front_end. It also updates
pkg/analyzer_plugin to make use of those re-exports.
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2993123002 .
This eliminates much of the duplication between the two token hierarchies
by merging much of the fasta.Token hierarchy into the analyzer.Token.
* merge token hierarchies
* cleanup fasta.Token.copy()
* set preceedingComments field via constructor
* replace isBuiltInIdentifier --> type.isBuiltIn
* change all fasta.Token references to analyzer.Token
* move fasta.Token.value() implementation into analyzer.SimpleToken
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2890523002 .
* move isPseudo getter out of Token into TokenType
* move special case charCount logic out of fasta.Token into fasta.ErrorToken
* move more accessors from fasta.Token into analyzer.Token
* add @overrides to denote which methods have been declared in analyzer.Token
* make fasta LazySubstring and its subclasses private
* rename fasta.previousToken to previous to match analyzer.Token
* flatten fasta.Token.preceedingCommentTokens into preceedingComments
* normalize fasta.Token.next and analyzer.Token.next
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2872433005 .
This CL hooks up enough infrastructure for doing simple top level type
inference of variables whose initializer is an int.
Additional type inference rules will be added in follow-up CLs.
R=karlklose@google.com
Review-Url: https://codereview.chromium.org/2828253003 .
* update the fasta scanner to generate comment and dartdoc tokens
* replace fasta.preceedingComments field
with preceedingCommentTokens field and preceedingComments getter/setter
* update token translator
R=paulberry@google.com
Review-Url: https://codereview.chromium.org/2763833002 .
These utilities are needed for generating the code that supports
summaries; in order to move summary logic into front_end, we will need
these code generation tools in front_end as well.
Also make a minor change to summary code generation logic so that it
talks to the file system directly rather than depending on analyzer's
file system abstraction.
R=brianwilkerson@google.com
Review-Url: https://codereview.chromium.org/2742333005 .
The end goal is to move summary logic to the front end, which means
that it won't be able to depend on analyzer. This removes a key
dependency on analyzer (the dependency of the code generator on the
analyzer parser).
R=scheglov@google.com
Review-Url: https://codereview.chromium.org/2748803003 .
Previously this was working for a few use cases by dumb luck, but it
wasn't well tested and became broken by
b29c6bf330. This CL fixes it for real
and adds tests to verify the fix.
Note that there is a change to how we represent an unresolved
import/export: instead of storing a URI_EXISTS modifier in the element
model, we simply mark the imported library element as synthetic.
(This library element existed anyway; we just weren't marking it as
synthetic). This makes it easy to tie in with summaries, since it is
easy for the resynthesizer to detect that the library doesn't exist at
the time the library handle is resolved.
R=brianwilkerson@google.com
Review URL: https://codereview.chromium.org/2235373003 .
Konstantin Shcheglov