test: eval a strict function #5250
Conversation
mscdex
added
vm
|
Test fails for me in 3.x and passes in 4.0.0 and up, which is consistent with it detecting the bug it is designed to test, so that's good. (EDIT: Have to comment out |
|
@Trott may be because template strings were not available before 4.x? |
|
@thefourtheye Yeah, and maybe other stuff too. I think there's at least one arrow functions in there now, for example. |
|
@Trott @thefourtheye should I rewrite this without template strings to make tests running this on old versions possible? |
|
@kthelgason Ya, since we claim that this |
kthelgason
force-pushed
the
regr-test-eval-strict
branch
from
f7ea0f9
to
cd5b3fc
Compare
|
@thefourtheye agreed. Fixed and pushed. |
| @@ -0,0 +1,28 @@ | |||
| /* eslint-disable strict */ | |||
There was a problem hiding this comment.
Why this is necessary?
There was a problem hiding this comment.
in strict mode running this will fail as bar is undefined until the code has been evaled. If this line is not present the build will fail linting
There was a problem hiding this comment.
it would be better to disable the no-undef rule inline
There was a problem hiding this comment.
Sorry, I could have been clearer above, the issue here is that this code is not valid in strict mode, since eval in strict mode does not introduce new variables into the surrounding scope. Therefore this file cannot have strict mode enabled, and hence this linter rule.
|
This test ran perfectly fine in 3.x (with The problem (which should not be "fixed") was in If you want to restore the template string for better code hygiene, feel free. |
|
@Trott Ok, I still feel that there might be some value to being able to run the test under 0.10 to see the failure. |
|
@kthelgason Fair enough. I'm not opposed to the current format or anything. |
|
@Trott willing to land this or looking for more eyes on it first? |
|
LGTM |
|
LGTM. CI Run: https://ci.nodejs.org/job/node-test-pull-request/1912/ |
PR-URL: #5250 Reviewed-By: James M Snell <[email protected]> Reviewed-By: Sakthipriyan Vairamani <[email protected]>
|
Landed in 25eedf0 |
PR-URL: #5250 Reviewed-By: James M Snell <[email protected]> Reviewed-By: Sakthipriyan Vairamani <[email protected]>
PR-URL: #5250 Reviewed-By: James M Snell <[email protected]> Reviewed-By: Sakthipriyan Vairamani <[email protected]>
PR-URL: #5250 Reviewed-By: James M Snell <[email protected]> Reviewed-By: Sakthipriyan Vairamani <[email protected]>
PR-URL: #5250 Reviewed-By: James M Snell <[email protected]> Reviewed-By: Sakthipriyan Vairamani <[email protected]>
This PR adds a regression test for the bug mentioned in #2245 where when a strict function is
evaled it would not capture it's context.