perf: cache conditionalMapping + per-plugin description-file lookups#556
Merged
alexander-akait merged 2 commits intomainfrom Apr 23, 2026
Merged
Conversation
On every resolve that descends through a description file, plugins were re-walking the same `package.json` content to answer questions whose inputs are stable: does this file have an exports/imports/alias field, what's the main module, does a conditional mapping resolve to the same target under the resolver's fixed condition set. Add narrow caches that key on stable object references (parsed JSON content, conditional mapping objects, conditionNames Sets) so repeat resolves against the same package skip the lookup work. * `util/entrypoints.js`: memoize `conditionalMapping(mapping, conditionNames)` in a `WeakMap<ConditionalMapping, WeakMap<Set<string>, ...>>`. Uses a sentinel for `null` results so the hit path is a single `get()`. * `ExportsFieldPlugin`/`ImportsFieldPlugin`: cache `null` in the existing per-content processor WeakMap for description files that have no exports/imports field, skipping the `DescriptionFileUtils.getField` walk on subsequent resolves. Builds the processor inside the existing try/catch so malformed fields still surface their error. * `MainFieldPlugin`: cache the fully-resolved `mainModule` string (with forceRelative already applied) per content, with a `NO_MAIN` sentinel for packages whose main field is absent / "." / "./". * `AliasFieldPlugin`: cache the resolved alias-map object per content (or a sentinel when the field isn't a valid alias config). * `SelfReferencePlugin`: cache the `"name"` string per content when self-reference applies (or a sentinel when it can't). Benchmarks (--no-opt, local): realistic-midsize warm: 723 ops/s -> ~763 ops/s (+5.5%) realistic-midsize cold: 34.0 -> ~34.7 (~+2%) exports-patterns-many: 507 -> 526 (+3.8%) mixed-conditions: 2718 -> 2730-2765 (~+1%) self-reference: 2233 -> 2265 (+1.4%) exports-field (warm): ~1240 -> ~1230 (within noise) https://claude.ai/code/session_01JM6aPUSLLfmEdtxeffMAiZ
|
| Name | Type |
|---|---|
| enhanced-resolve | Patch |
Click here to learn what changesets are, and how to add one.
Click here if you're a maintainer who wants to add a changeset to this PR
|
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #556 +/- ##
==========================================
- Coverage 96.64% 96.53% -0.11%
==========================================
Files 50 50
Lines 2830 2887 +57
Branches 881 901 +20
==========================================
+ Hits 2735 2787 +52
- Misses 79 84 +5
Partials 16 16
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Merging this PR will improve performance by 29.86%
Performance Changes
Comparing 
|
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
2 participants
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.
On every resolve that descends through a description file, plugins were
re-walking the same
package.jsoncontent to answer questions whoseinputs are stable: does this file have an exports/imports/alias field,
what's the main module, does a conditional mapping resolve to the same
target under the resolver's fixed condition set. Add narrow caches that
key on stable object references (parsed JSON content, conditional
mapping objects, conditionNames Sets) so repeat resolves against the
same package skip the lookup work.
util/entrypoints.js: memoizeconditionalMapping(mapping, conditionNames)in a
WeakMap<ConditionalMapping, WeakMap<Set<string>, ...>>. Uses asentinel for
nullresults so the hit path is a singleget().ExportsFieldPlugin/ImportsFieldPlugin: cachenullin the existingper-content processor WeakMap for description files that have no
exports/imports field, skipping the
DescriptionFileUtils.getFieldwalk on subsequent resolves. Builds the processor inside the existing
try/catch so malformed fields still surface their error.
MainFieldPlugin: cache the fully-resolvedmainModulestring (withforceRelative already applied) per content, with a
NO_MAINsentinelfor packages whose main field is absent / "." / "./".
AliasFieldPlugin: cache the resolved alias-map object per content(or a sentinel when the field isn't a valid alias config).
SelfReferencePlugin: cache the"name"string per content whenself-reference applies (or a sentinel when it can't).
Benchmarks (--no-opt, local):
realistic-midsize warm: 723 ops/s ->
763 ops/s (+5.5%)+1%)realistic-midsize cold: 34.0 ->
34.7 (+2%)exports-patterns-many: 507 -> 526 (+3.8%)
mixed-conditions: 2718 -> 2730-2765 (
self-reference: 2233 -> 2265 (+1.4%)
exports-field (warm): ~1240 -> ~1230 (within noise)
https://claude.ai/code/session_01JM6aPUSLLfmEdtxeffMAiZ