Fix strict test suite #24431
Fix strict test suite #24431
Conversation
… 'test/runner strict' tests.
juj
force-pushed
the
fix_strict_test_suite
branch
from
9775c08 to
ff33a9e
Compare
| @@ -1336,17 +1336,17 @@ def in_dir(self, *pathelems): | |||
| def add_pre_run(self, code): | |||
| assert not self.get_setting('MINIMAL_RUNTIME') | |||
| create_file('prerun.js', 'Module.preRun = function() { %s }\n' % code) | |||
| self.emcc_args += ['--pre-js', 'prerun.js'] | |||
| self.emcc_args += ['--pre-js', 'prerun.js', '-sINCOMING_MODULE_JS_API=[preRun]'] | |||
There was a problem hiding this comment.
The problem with doing this here is that then the test doesn't have any way to add more to INCOMING_MODULE_JS_API.
We currently don't have any way to make settings like INCOMING_MODULE_JS_API additive.
I've tried to make them additive by default (which seems like the right behaviour in almost all cases I can think of): #19938. But that PR got hung up with disagreement about how / if to change the default behaviour.
There was a problem hiding this comment.
The problem with doing this here is that then the test doesn't have any way to add more to
INCOMING_MODULE_JS_API.
Yeah, I was thinking about the additivity as well while writing this. That should definitely be the best behavior to have.
The add_pre_run() is only for tests, and so far it seems that no test needed to concatenate (or that would have shown up in strict run). If/when such a test comes up, that directive can be refactored to occur at the call site.
There was a problem hiding this comment.
That is, either that specific test can then be written to duplicate the -sINCOMING_MODULE_JS_API=[preRun,theothersymbol], or -sINCOMING_MODULE_JS_API=[preRun] be hoisted from add_pre_run() to all the tests.
|
For errors like |
| @@ -9414,7 +9414,11 @@ def test_pthread_dylink_longjmp(self): | |||
| @needs_dylink | |||
| @node_pthreads | |||
| def test_pthread_dylink_main_module_1(self): | |||
| self.emcc_args += ['-Wno-experimental', '-pthread', '-lhtml5'] | |||
| # TODO: For some reason, -lhtml5 must be passed in -sSTRICT mode, but can NOT | |||
| # be passed when not compiling in -sSTRICT mode. That does not seem intentional? | |||
There was a problem hiding this comment.
This certainly sounds like a bug. Could you open one an link it here (and below).
Enable
test/runner strictto pass (tested on Windows)