Show existential types as $a and name their constructor by cknitt · Pull Request #9080 · rescript-lang/rescript

cknitt · 2026-10-27T13:14:11Z

Existential types of GADT constructors are printed $a instead of their internal name $Any_'a, in errors and in hover. An escaping one is explained by naming the constructor that binds it:

This has type: $a
But it's expected to have type: 'a
$a is an existential type bound by the constructor Any.
It is only known in the scope of the pattern that matches Any.

Existentials are named $a before the compiler makes the name unique, as in OCaml 5, so two existentials binding the same variable print as $a and $a1. The constructor binding each name is recorded for the explanation.

New fixtures gadt_existential_escape and gadt_existential_same_name and analysis test HoverExistential.

🤖 Generated with Claude Code

https://claude.ai/code/session_01Kbf3gLMawVgtmiKy3TMBc3

github-actions · 2026-10-27T13:14:29Z

Changelog

💅 Polish

Existential types of GADT constructors are shown as $a instead of $Any_'a, and an escaping one is explained by naming the constructor that binds it.

Semantic Diff

3 commits · Full page

Show existential types as $a and name their constructor

_{bff79a6 · cknitt + Claude · Oct 27, 2026, 12:44 UTC}

Why

An existential type of a GADT constructor is an abstract type named $Any_'a internally, and that name was printed as is in errors and hover. It is now printed $a, as in OCaml 5, and the escape error says which constructor binds it instead of calling it a type constructor.

Diagnostic modified (existential escape)

type rec any = Any('a): any

let get = (Any(x)) => x

diagnostic

-This has type: $Any_'a
+This has type: $a
 But it's expected to have type: 'a
-The type constructor $Any_'a would escape its scope
+$a is an existential type bound by the constructor Any.
+It is only known in the scope of the pattern that matches Any.

Review

Why decode the name instead of recording the constructor? The name already holds both parts, so printing needs no table that would have to be reset between compilations in one process.

Give distinct existential types distinct names

_{2294674 · cknitt + Claude · Oct 27, 2026, 20:25 UTC}

Codex review on #9080: printing $C_'a as $a made existentials of different constructors that bind the same variable print alike, so a mismatch between them read $a vs $a. Existentials are now named $a before the compiler makes the name unique, as OCaml 5 does, so the second one is $a1; the constructor binding each name is recorded for the escape explanation instead of being decoded from the name.

Diagnostic modified (existentials of two constructors)

type rec any = First('a): any | Second('a): any

let same = (First(x), Second(y)) => x == y

diagnostic

-This has type: $a
+This has type: $a1
 But it's being compared to something of type: $a

Key existential constructors by identifier

_{f56b61c · cknitt + Claude · Oct 27, 2026, 20:38 UTC}

Codex review on #9080: the constructor of each existential was recorded by its name, but the counter that makes names unique is reset for each binding, so a nested module's $a could overwrite the constructor of an outer $a. It is recorded by the identifier's stamp now.

codecov · 2026-10-27T13:20:28Z

Codecov Report

❌ Patch coverage is 88.88889% with 2 lines in your changes missing coverage. Please review.
✅ Project coverage is 81.52%. Comparing base (beb423d) to head (f56b61c).

Files with missing lines	Patch %	Lines
compiler/ml/printtyp.ml	75.00%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #9080      +/-   ##
==========================================
+ Coverage   81.45%   81.52%   +0.06%     
==========================================
  Files         460      460              
  Lines       64645    64659      +14     
==========================================
+ Hits        52657    52713      +56     
+ Misses      11988    11946      -42

Files with missing lines	Coverage Δ
compiler/ml/ctype.ml	`67.80% <100.00%> (+0.43%)`	⬆️
compiler/ml/printtyp.ml	`72.06% <75.00%> (+0.07%)`	⬆️

... and 4 files with indirect coverage changes

🚀 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 · 2026-10-27T13:25:37Z

Open in StackBlitz

rescript

npm i https://pkg.pr.new/rescript@9080

@rescript/belt

npm i https://pkg.pr.new/@rescript/belt@9080

@rescript/darwin-arm64

npm i https://pkg.pr.new/@rescript/darwin-arm64@9080

@rescript/linux-arm64

npm i https://pkg.pr.new/@rescript/linux-arm64@9080

@rescript/linux-x64

npm i https://pkg.pr.new/@rescript/linux-x64@9080

@rescript/runtime

npm i https://pkg.pr.new/@rescript/runtime@9080

@rescript/win32-x64

npm i https://pkg.pr.new/@rescript/win32-x64@9080

commit: f56b61c

github-actions · 2026-10-27T13:36:40Z

Developer playground preview: https://rescript-lang.github.io/rescript/dev-playground/?version=pr-9080

cknitt · 2026-10-27T14:20:10Z

@codex review

chatgpt-codex-connector · 2026-10-27T14:20:34Z

Codex Review Summary

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

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-10-27T20:40:26.519315Z	`9569e25`	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 · 2026-10-27T17:51:10Z

@codex review

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 4bc31934fb

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".

Codex review on #9080: printing `$C_'a` as `$a` made existentials of different constructors that bind the same variable print alike, so a mismatch between them read `$a` vs `$a`. Existentials are now named `$a` before the compiler makes the name unique, as OCaml 5 does, so the second one is `$a1`; the constructor binding each name is recorded for the escape explanation instead of being decoded from the name. Diagnostic modified (existentials of two constructors): Example: type rec any = First('a): any | Second('a): any let same = (First(x), Second(y)) => x == y Before (diagnostic): This has type: $a But it's being compared to something of type: $a After (diagnostic): This has type: $a1 But it's being compared to something of type: $a Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Kbf3gLMawVgtmiKy3TMBc3 Signed-off-by: Christoph Knittel <ck@cca.io>

cknitt · 2026-10-27T20:25:39Z

@codex review

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: a757919c44

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".

Codex review on #9080: the constructor of each existential was recorded by its name, but the counter that makes names unique is reset for each binding, so a nested module's `$a` could overwrite the constructor of an outer `$a`. It is recorded by the identifier's stamp now. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Kbf3gLMawVgtmiKy3TMBc3 Signed-off-by: Christoph Knittel <ck@cca.io>

cknitt · 2026-10-27T20:38:46Z

@codex review

chatgpt-codex-connector · 2026-10-27T20:40:24Z

Codex Review: Didn't find any major issues. Another round soon, please!

Reviewed commit: 9569e25834

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".

Why: An existential type of a GADT constructor is an abstract type named `$Any_'a` internally, and that name was printed as is in errors and hover. It is now printed `$a`, as in OCaml 5, and the escape error says which constructor binds it instead of calling it a type constructor. Diagnostic modified (existential escape): Example: type rec any = Any('a): any let get = (Any(x)) => x Before (diagnostic): This has type: $Any_'a But it's expected to have type: 'a The type constructor $Any_'a would escape its scope After (diagnostic): This has type: $a But it's expected to have type: 'a $a is an existential type bound by the constructor Any. It is only known in the scope of the pattern that matches Any. Review: Why decode the name instead of recording the constructor? The name already holds both parts, so printing needs no table that would have to be reset between compilations in one process. Fixes: #9068 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Kbf3gLMawVgtmiKy3TMBc3 Signed-off-by: Christoph Knittel <ck@cca.io>

Codex review on #9080: printing `$C_'a` as `$a` made existentials of different constructors that bind the same variable print alike, so a mismatch between them read `$a` vs `$a`. Existentials are now named `$a` before the compiler makes the name unique, as OCaml 5 does, so the second one is `$a1`; the constructor binding each name is recorded for the escape explanation instead of being decoded from the name. Diagnostic modified (existentials of two constructors): Example: type rec any = First('a): any | Second('a): any let same = (First(x), Second(y)) => x == y Before (diagnostic): This has type: $a But it's being compared to something of type: $a After (diagnostic): This has type: $a1 But it's being compared to something of type: $a Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Kbf3gLMawVgtmiKy3TMBc3 Signed-off-by: Christoph Knittel <ck@cca.io>

Codex review on #9080: the constructor of each existential was recorded by its name, but the counter that makes names unique is reset for each binding, so a nested module's `$a` could overwrite the constructor of an outer `$a`. It is recorded by the identifier's stamp now. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Kbf3gLMawVgtmiKy3TMBc3 Signed-off-by: Christoph Knittel <ck@cca.io>

cknitt added this pull request to stack #9087 October 27, 2026 13:14

cknitt mentioned this pull request Oct 27, 2026

Type a let destructuring a GADT without switch diagnostics #9081

Open

cknitt marked this pull request as ready for review October 27, 2026 14:20

chatgpt-codex-connector Bot reviewed Oct 27, 2026

View reviewed changes

Comment thread compiler/ml/printtyp.ml Outdated

chatgpt-codex-connector Bot reviewed Oct 27, 2026

View reviewed changes

Comment thread compiler/ml/ctype.ml Outdated

cknitt and others added 3 commits October 28, 2026 07:48

cknitt force-pushed the fix-9068-existential-names branch from 9569e25 to f56b61c Compare October 28, 2026 08:10

Conversation

cknitt commented Oct 27, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions Bot commented Oct 27, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Changelog

💅 Polish

Semantic Diff

Uh oh!

codecov Bot commented Oct 27, 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 27, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions Bot commented Oct 27, 2026

Uh oh!

cknitt commented Oct 27, 2026

Uh oh!

chatgpt-codex-connector Bot commented Oct 27, 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 27, 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 27, 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 27, 2026

Uh oh!

chatgpt-codex-connector Bot commented Oct 27, 2026

Uh oh!

This branch has not been deployed

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

cknitt commented Oct 27, 2026 •

edited

Loading

github-actions Bot commented Oct 27, 2026 •

edited

Loading

codecov Bot commented Oct 27, 2026 •

edited

Loading

pkg-pr-new Bot commented Oct 27, 2026 •

edited

Loading

chatgpt-codex-connector Bot commented Oct 27, 2026 •

edited

Loading