You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Commit 97067ca
Browse filesBrowse the repository at this point in the historyBrowse files
* Remove the call-site @inlined attribute
@inlined was translated into Lapply's ap_inlined, but no pass in
compiler/core ever read it: the only use in lam_pass_remove_alias was
commented out. So (@inlined f)(x) was accepted and ignored,
(@inlined(never) f)(x) still inlined, and the natural ReScript placement
@inlined f(x) already reported warning 53.
Delete the attribute's translation and the ap_inlined field. Any
remaining @inlined now reaches check_attribute and is reported as a
misplaced attribute (warning 53) instead of being silently ignored.
See #8733.
Signed-off-by: Christoph Knittel <christoph@knittel.cc>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Add CHANGELOG link for #8734
Signed-off-by: Christoph Knittel <christoph@knittel.cc>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Keep the @inlined warning fixture formattable
The formatter drops attributes on functor applications, so @inlined F({}) cannot appear in a formatted fixture. Cover the two expression placements only.
Signed-off-by: Christoph Knittel <christoph@knittel.cc>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Check attributes on primitive callees
A fully applied primitive call is translated without passing its callee
through transl_exp, so check_attribute never saw the callee's attributes
and `(@inlined id)(1)` with an external `id` got no warning. Check them
on that path too.
Signed-off-by: Christoph Knittel <christoph@knittel.cc>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Remove a stale commented-out inlining check
Signed-off-by: Christoph Knittel <christoph@knittel.cc>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Honor @warning on primitive callees when checking attributes
The attribute check added for fully applied primitives ran outside the
callee's warning scope, so `(@warning("-53") @inlined id)(1)` still
reported warning 53. Run it in the same scope transl_exp uses.
Signed-off-by: Christoph Knittel <christoph@knittel.cc>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Translate module expressions in their warning scope
With @inlined no longer removed from a functor application's callee, it
reaches check_attribute_on_module, which ran outside any warning scope:
`(@warning("-53") @inlined F)(X)` still reported warning 53. Translate a
module expression inside the warning scope of its attributes, like
transl_exp does for expressions.
Signed-off-by: Christoph Knittel <christoph@knittel.cc>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Copy file name to clipboardExpand all lines: CHANGELOG.md
+1Lines changed: 1 addition & 0 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -14,6 +14,7 @@
14
14
15
15
#### :boom: Breaking Change
16
16
17
+
- Remove the call-site `@inlined` attribute, which was parsed but never affected code generation. It is now reported as a misplaced attribute (warning 53). https://github.com/rescript-lang/rescript/pull/8734
17
18
- Remove `@deriving(abstract)` and `@deriving(jsConverter)`. Use record types (with optional fields, mutable fields and `@as` renaming) and polymorphic variants directly instead. https://github.com/rescript-lang/rescript/pull/8729
0 commit comments