This adjusts all UTF-8 tests to the new semantics in the breaking
change described here: https://github.com/dart-lang/sdk/issues/41100
This has three parts:
- Unpaired surrogates are encoded as replacement characters, and
encoded surrogates are considered malformed input when decoding.
- Decoding errors are generally reported on the position of the byte
that conclusively makes the input malformed.
- The number of replacement characters emitted by the decoder is
generally one per unfinished sequence or undecodable byte.
The code changes to implement the new semantics are placed in subsequent
commits.
Change-Id: I4cc8ce660e39287e734070764ab8e1f0ebb8b9e0
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/143815
Reviewed-by: Lasse R.H. Nielsen <lrn@google.com>
- remove unused variables
- use expanded test cases that were skipped with a mistyped variable
- Fix unintended string concat
Change-Id: I9bdfad6d31655261b378f878f36f4d07b2f64fe4
Reviewed-on: https://dart-review.googlesource.com/c/75789
Auto-Submit: Kevin Moore <kevmoo@google.com>
Reviewed-by: Lasse R.H. Nielsen <lrn@google.com>
Also, --limit-ints-to-64-bits is enabled when running tests in _2 test
suites.
This is the re-landing of 92ebd8aefa with
fixes:
* Revert changes in pkg/dev_compiler/test/browser/language_tests.js
and tests/language/language_dart2js.status as they describe
'language' test suite, not 'language_2'
* Correct tests/standalone_2/io/fuzz_support.dart as file_fuzz test was
timing out on Windows and failing on android/arm.
Closes https://github.com/dart-lang/sdk/issues/31396
Change-Id: If9ca77fca300ddc605f17a7be39d1707e9724e25
Reviewed-on: https://dart-review.googlesource.com/21700
Reviewed-by: Zach Anderson <zra@google.com>
Commit-Queue: Alexander Markov <alexmarkov@google.com>