Repository navigation
Version Packages - #868
Merged
Merged
Version Packages#868
Conversation
github-actions
Bot
force-pushed
the
changeset-release/main
branch
2 times, most recently
from
October 16, 2026 15:30
5f9206b to
0b11be0
Compare
github-actions
Bot
force-pushed
the
changeset-release/main
branch
from
October 16, 2026 15:50
0b11be0 to
54f8a01
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR was opened by the Changesets release GitHub action. When you're ready to do a release, you can merge this and the packages will be published to npm automatically. If you're not ready to do a release yet, that's fine, whenever you add more changesets to main, this PR will be updated.
Releases
@effect/tsgo@0.53.3
Patch Changes
1ab5264:
effect-tsgo diagnosticsnow reports a diagnostic's related locations. They used to be dropped in every format, somissingEffectContextnever showed which layer could provide the missing service andapiStabilityLeaknever showed where the unstable declaration lives.With
--format json, each diagnostic that has related locations gets arelatedInformationarray. Each entry has the same location fields as the diagnostic, plus its owncodeandmessage. Diagnostics without related locations are unchanged.{ "line": 23, "column": 3, "code": 377004, "name": "missingEffectContext", "message": "This Effect requires a service that is missing from the expected Effect context: `Db | Fs`.", "relatedInformation": [ { "file": ".../src/index.ts", "start": 662, "length": 14, "line": 22, "column": 3, "endLine": 22, "endColumn": 17, "code": 377059, "message": "Adjusting this layer composition could provide the missing service." } ] }With
--format pretty, the related locations are printed under the diagnostic, each with its own code snippet, astsc --prettydoes:The
~markers in pretty output are also aligned again on lines 10 and above. They used to sit one column to the left of the code.4672ba1: Make
globalDateandglobalDateInEffectdo less work per file. Their shared walker used to look up the Effect context flags of every node it visited, although only call expressions (Date.now()) andnewexpressions (new Date()) can produce a diagnostic. It now looks up the flags only on those two node kinds, likeglobalFetch,globalConsole,globalTimers,globalRandomandcryptoRandomUUIDalready do.Reported diagnostics are unchanged.
b5df4e9: Speed up
apiStabilityLeakby reusing one stability analysis per file.The rule used to allocate a fresh analysis, with its maps and work queues, for every export of a file. It now keeps one analysis per file and resets it before each export. The reset analysis behaves exactly like a fresh one, so the reported diagnostics are unchanged. A published surface that a later export reads from the shared cache is no longer queued for publication again, and a published surface without retained findings no longer allocates a findings map.
On
packages/effectof effect 4.0.5 with every rule enabled, the rule allocates 252 MB → 140 MB.d3b53a7: Make the Effect diagnostics write less to the checker's per-node caches. Rules run on the checker goroutine right after each file is checked, and on the Effect monorepo most of their cost was cache traffic (map misses, nil inserts, map growth) and the garbage collection it caused, not type computation.
Effect.gen?" no longer probe or fill the symbol cache for statements, blocks, clauses and tokens, which can never resolve to a symbol.Effect.gen,Effect.fnor an Effect thunk stores an entry only where the answer changes, about 2% of nodes, instead of two entries for every node of every file.Effect, and whether a file has been checked is asked once per file instead of on every query.Reported diagnostics are unchanged.
@effect/tsgo-darwin-arm64@0.53.3
@effect/tsgo-darwin-x64@0.53.3
@effect/tsgo-linux-arm@0.53.3
@effect/tsgo-linux-arm64@0.53.3
@effect/tsgo-linux-x64@0.53.3
@effect/tsgo-win32-arm64@0.53.3
@effect/tsgo-win32-x64@0.53.3