Parse attributes after `= private` and a doc comment before a manifest by cknitt · Pull Request #9115 · rescript-lang/rescript

cknitt · 2026-10-28T15:34:23Z

type t = private @attr {x: int} and type z = /** d */ int were both rejected, although type t = @attr {x: int}, type t = private {x: int} and type z = @attr int parse.

After = private, an attribute counts as the start of a type expression, so @attr {x: int} was parsed as a manifest and the record failed. After =, a doc comment was always sent to the type representation parser, which then expected a constructor, so /** d */ int, {"a": int}, (int, int), [#A] and M.t failed.

The lookahead that already decided between representation and manifest for an attribute after = is now a shared helper, used for both attributes and doc comments, after = and after = private: past attributes and doc comments, a record, .., | or a constructor makes a representation, anything else a manifest. Attributes and doc comments before a private record or .. go on the type declaration, as for the non-private forms (#8919).

Tests: new cases in the parser fixture typedefinition/typeRepresentation.res (attributes and doc comments after private before a record, .., constructors, a type and an object; doc comments after = before int, an object, a tuple, a polymorphic variant, M.t and M.t = A | B) and in the printer fixture typeDef/attributes.res, which also checks that the newly accepted forms round-trip.

Not changed: type t = private {x: {y: int}} (a nested record after private) is still rejected, with or without an attribute; that is a separate limitation of the private record path. A spread keeps its old meaning: after a doc comment {...base, y: int} is a record declaration, after an attribute an object type, so @attr {...base, y: int} and /** d */ {...base, "y": int} are still rejected, as on master; tracked in #9122.

A differential run against master over 672 declarations (no prefix, one or two doc comments, an attribute, and both orders of doc comment and attribute, times 28 bodies such as constructors, variant and record spreads, records, objects, .., int, M.t = A, 'a and int => int; with and without private; in .res and .resi) gives the same parse tree for every declaration master accepts. The only changes are forms master rejected: a doc comment before a type expression or object, and an attribute after private before constructors, a record or ...

Fixes #9106

🤖 Generated with Claude Code

https://claude.ai/code/session_01Kbf3gGkPma9LAZimuQjnrP

github-actions · 2026-10-28T15:34:51Z

Changelog

🐛 Bug fix

Accept attributes and doc comments after = private, and a doc comment after = in front of a type expression such as int.

Semantic Diff

1 commit · Full page

✏️ Parse attributes after = private and a doc comment before a manifest

_{bea1cbc · cknitt + Claude · Oct 28, 2026, 15:34 UTC · edited}

Why

private @attr {x: int} took the type expression path and rejected the record, and a doc comment right after = always started a constructor list, so /** d */ int failed. Both now use the lookahead that attributes after = already had: past attributes and doc comments, a record, .., | or constructor is a representation, anything else a manifest.

AST modified (doc comment before a manifest)

type z = /** d */ int

diagnostic · Before

Did you mean `Int` instead of `int`?

parse · After

type_declaration "z"
  ptype_kind =
    Ptype_abstract
  ptype_private = Public
  ptype_manifest =
    Some
      core_type
        attribute  "res.doc"
        …
        Ptyp_constr "int"

Format modified (attribute after private)

type t = private @attr {x: int}
type u = private /** d */ int

diagnostic · Before

An inline record type declaration is only allowed in a variant
constructor's declaration or nested inside of a record type declaration

rescript · After

@attr type t = private {x: int}
type u = private /** d */ int

Fixes #9106

Review

Does a leading doc comment change any form that parsed before? No. After a doc comment the lookahead keeps the representation for every start the constructor and record parsers accept (|, a constructor, ...base, {..., a record, ..); only starts they rejected, such as int, M.t, 'a, (a, b), [#A], {.} or {"a": int}, become a manifest. A differential run over 672 declarations (prefixes x bodies, with and without private, .res and .resi) gives the same tree as before for every one that parsed.

codecov · 2026-10-28T15:39:51Z

Codecov Report

❌ Patch coverage is 96.29630% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 81.46%. Comparing base (85f31fb) to head (bea1cbc).

Files with missing lines	Patch %	Lines
compiler/syntax/src/res_core.ml	96.29%	1 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #9115   +/-   ##
=======================================
  Coverage   81.46%   81.46%           
=======================================
  Files         460      460           
  Lines       64675    64684    +9     
=======================================
+ Hits        52689    52698    +9     
  Misses      11986    11986

Files with missing lines	Coverage Δ
compiler/syntax/src/res_core.ml	`92.02% <96.29%> (+0.01%)`	⬆️

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

pkg-pr-new · 2026-10-28T15:48:10Z

Open in StackBlitz

rescript

npm i https://pkg.pr.new/rescript@9115

@rescript/belt

npm i https://pkg.pr.new/@rescript/belt@9115

@rescript/darwin-arm64

npm i https://pkg.pr.new/@rescript/darwin-arm64@9115

@rescript/linux-arm64

npm i https://pkg.pr.new/@rescript/linux-arm64@9115

@rescript/linux-x64

npm i https://pkg.pr.new/@rescript/linux-x64@9115

@rescript/runtime

npm i https://pkg.pr.new/@rescript/runtime@9115

@rescript/win32-x64

npm i https://pkg.pr.new/@rescript/win32-x64@9115

commit: bea1cbc

github-actions · 2026-10-28T15:51:32Z

Developer playground preview: https://rescript-lang.github.io/rescript/dev-playground/?version=pr-9115

cknitt · 2026-10-28T16:40:48Z

@codex review

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 5a8d70f94b

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2026-10-28T16:42:44Z

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-10-28T16:52:18.922384Z	`bea1cbc`	Manual request

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

cknitt · 2026-10-28T16:46:05Z

@codex review

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 6883c20c44

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

@attr

Why: `private @attr {x: int}` took the type expression path and rejected the record, and a doc comment right after `=` always started a constructor list, so `/** d */ int` failed. Both now use the lookahead that attributes after `=` already had: past attributes and doc comments, a record, `..`, `|` or constructor is a representation, anything else a manifest. AST modified (doc comment before a manifest): Example: type z = /** d */ int Before (diagnostic): Did you mean `Int` instead of `int`? After (parse): type_declaration "z" ptype_kind = Ptype_abstract ptype_private = Public ptype_manifest = Some core_type attribute "res.doc" … Ptyp_constr "int" Format modified (attribute after `private`): Example: type t = private @attr {x: int} type u = private /** d */ int Before (diagnostic): An inline record type declaration is only allowed in a variant constructor's declaration or nested inside of a record type declaration After (rescript): @attr type t = private {x: int} type u = private /** d */ int Review: Does a leading doc comment change any form that parsed before? No. After a doc comment the lookahead keeps the representation for every start the constructor and record parsers accept (`|`, a constructor, `...base`, `{...`, a record, `..`); only starts they rejected, such as `int`, `M.t`, `'a`, `(a, b)`, `[#A]`, `{.}` or `{"a": int}`, become a manifest. A differential run over 672 declarations (prefixes x bodies, with and without `private`, .res and .resi) gives the same tree as before for every one that parsed. Fixes #9106 Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_01Kbf3gGkPma9LAZimuQjnrP Signed-off-by: Christoph Knittel <[email protected]>

cknitt · 2026-10-28T16:50:43Z

@codex review

chatgpt-codex-connector · 2026-10-28T16:52:17Z

Codex Review: Didn't find any major issues. Chef's kiss.

Reviewed commit: bea1cbcda3

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector Bot reviewed Oct 28, 2026

View reviewed changes

Comment thread compiler/syntax/src/res_core.ml Outdated

cknitt force-pushed the fix-9106-private-attr-doc-comment branch from 5a8d70f to 6883c20 Compare October 28, 2026 16:45

chatgpt-codex-connector Bot reviewed Oct 28, 2026

View reviewed changes

Comment thread compiler/syntax/src/res_core.ml

cknitt force-pushed the fix-9106-private-attr-doc-comment branch from 6883c20 to bea1cbc Compare October 28, 2026 16:50

cknitt mentioned this pull request Oct 28, 2026

Parser: an attribute or doc comment in front of a {... spread type changes how it is read #9122

Open

Navigation Menu

Repository navigation

Parse attributes after `= private` and a doc comment before a manifest - #9115

Parse attributes after `= private` and a doc comment before a manifest#9115
cknitt wants to merge 1 commit into
masterfrom
fix-9106-private-attr-doc-comment

cknitt commented Oct 28, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented Oct 28, 2026 •

edited

Loading

Uh oh!

codecov Bot commented Oct 28, 2026 •

edited

Loading

Uh oh!

pkg-pr-new Bot commented Oct 28, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented Oct 28, 2026

Uh oh!

cknitt commented Oct 28, 2026

Uh oh!

chatgpt-codex-connector Bot left a comment

Uh oh!

Uh oh!

chatgpt-codex-connector Bot commented Oct 28, 2026 •

edited

Loading

Uh oh!

cknitt commented Oct 28, 2026

Uh oh!

chatgpt-codex-connector Bot left a comment

Uh oh!

Uh oh!

cknitt commented Oct 28, 2026

Uh oh!

chatgpt-codex-connector Bot commented Oct 28, 2026

Uh oh!

This branch has not been deployed

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

cknitt commented Oct 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions Bot commented Oct 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Changelog

🐛 Bug fix

Semantic Diff

Uh oh!

codecov Bot commented Oct 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

pkg-pr-new Bot commented Oct 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions Bot commented Oct 28, 2026

Uh oh!

cknitt commented Oct 28, 2026

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

chatgpt-codex-connector Bot commented Oct 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codex Review Summary

Uh oh!

cknitt commented Oct 28, 2026

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

cknitt commented Oct 28, 2026

Uh oh!

chatgpt-codex-connector Bot commented Oct 28, 2026

Uh oh!

This branch has not been deployed

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

cknitt commented Oct 28, 2026 •

edited

Loading

github-actions Bot commented Oct 28, 2026 •

edited

Loading

codecov Bot commented Oct 28, 2026 •

edited

Loading

pkg-pr-new Bot commented Oct 28, 2026 •

edited

Loading

chatgpt-codex-connector Bot commented Oct 28, 2026 •

edited

Loading