performance: Improve metro-file-map's Node crawler hot-path by kitten · Pull Request #1677 · react/metro

kitten · 2026-04-08T00:08:08Z

Summary

When a larger amount of paths are being crawled, the Node crawler becomes a hot-path and minor tweaks can have a larger impact on startup time, than previously anticipated (at least for Expo projects)

Remove repeated absoluteToNormal calls (medium impact)
Add a dictionary for extensions lookup (very low impact)
Strip forEach calls for for-of (formatting only; virtually no impact)
Refactor Promise code (no impact, just formatting)

Changelog:

 - **[Performance]**: 33% faster non-Watchman filesystem crawl on startup

Test plan

CI tests should pass unchanged
(Rob: See benchmarks and tests in PR comments)

…al` calls

A previous implementation kept a `visited` set in the Node crawler. Instead, we can also deduplicate and remove overlapping roots (directories within other roots) in the `FileMap` preemptively for all crawlers to prevent duplicate crawls.

See: react/metro#1677

…ceNodeFilesystemAPI` flag) (#1693) Summary: This is a proposal to drop the native find binary crawler code path entirely. We've recently dropped the native find binary crawler in Expo, which was accidentally activated in some cases. We discovered (which is particularly obvious on large projects/repos and `expo/expo` itself) that it's consistently much slower than the Node FS crawler. To get into the "correct" preference and use the Node crawler, we're currently forcing `useWatchman: null`, which then defaults to use watchman, if it's installed, but falls back to `forceNodeFilesystemAPI: true`. This is outside of what the types specify as valid, but is the cleanest way to disable the native find binary crawling. This crawler basically just adds overhead over the Node.js implementation and is consistently slower, and less efficient, while the Node crawler is pretty optimised already (and has very low overhead once #1677 is merged) The timings of the native find binary crawler in `expo/expo` was also observed to scale with the size of this repo, compared to a minimal reproduction, so the overhead isn't just limited to the child process call and result parsing. Changelog: [Internal] Drop native find binary crawler and remove `forceNodeFilesystemAPI` flag Pull Request resolved: #1693 Test Plan: - Existing tests should pass - Quick generated script to play around with and assess the crawlers' comparative performance: kitten@74dffd4 Reviewed By: huntie Differential Revision: D101985455 Pulled By: robhogan fbshipit-source-id: efe2ab02fc5c0edac4d68f037db57c635fef8e50

Resolves the conflict with react#1693, which removed the native `find` crawler: `findNative` is dropped, and `nodeCrawl` keeps this PR's flattened promise, calling `find` directly.

A crawl root that is the direct parent of `rootDir` normalizes to `..`, which the `'..' + path.sep` check treated as inside `rootDir`, so files under `rootDir` were recorded as `../<rootDir name>/…` instead of normal paths. This is the monorepo case of a `watchFolders` entry one level above the project. Treat `..` as outside, like any other up-fragment, so paths are derived with `absoluteToNormal` until the crawl re-enters `rootDir`. A filesystem root (`/`, `C:\`) already ends with a separator, so appending another produced `//name` for `lstat` and `ignore`, and `removeOverlappingRoots` compared nested roots against `//` and kept them. Only append a separator when there isn't one. Also imports `node:path` in `removeOverlappingRoots`, per lint.

Keep `main`'s `Array.from(new Set(options.roots))` for now, leaving this PR to the crawler hot path. Deduplicating nested roots can follow separately. Also removes the committed `types/` declaration, which `main` no longer uses.

Only a root can already end with a separator (a filesystem root), so check that once per root and pass each directory's prefix down to its children, instead of checking every directory as we recurse.

`exts` was a plain object, so a lookup also found properties inherited from `Object.prototype`: a file such as `foo.constructor` or `foo.toString` was treated as having a listed extension, `lstat`ed and added to the file map. A `Set` has no inherited keys, and is as fast here - within about 1ns a lookup of the object, and much faster than `extensions.includes`, measured over the extension of every entry in a 53k-file crawl with 35 to 100 extensions.

Measured over the Dirent arrays of a 61k-entry crawl, `for...of` is within ~0.25ns an entry of the indexed loop once optimised, and ~0.2ms a crawl slower on a single cold pass, which is within run-to-run noise and well below what's measurable end to end.

With a single crawler implementation, `find` no longer needs a callback and positional arguments shared with an alternative: it takes the crawler options and returns a promise of the file data, which `nodeCrawl` awaits. Completion is checked in one place, which also covers an empty `roots` without a special case. Also rewrites the comment on deriving normal paths to cover both the within-rootDir and outside-rootDir cases, and takes the extension from the entry name.

robhogan · 2026-09-24T13:25:14Z

I've made a few superficial changes here and split out the root-deduplication, since that's independent of the other wins and I'd like to make sure the "watchman root within a non-watchman root" is handled to avoid any possibility of regression. But otherwise, this looks good - the headline is ~33% faster crawl, which merits a Performance changelog.

Changes I've made:

Fixed a couple of edge cases (watch root is a filesystem/drive root, object prototype props like 'constructor' were matched as extensions)
Flattened nodeCrawl since there's now only one "find" implementation, which allowed tidying up some of the optimisations
Prefer readability where perf difference is negligible -for .. of over indexed loop, Set over {[string]?: true}

Claude-driven benchmarks

Benchmarked the current head (3c9e509) against main (a5d11f7), on an RN 0.87 app installed with pnpm (3,031 symlinks, 42k files). Both built from source, 30 rounds, each running both builds in separate processes in random order. Values are medians ± 95% bootstrap CI, and the change is the median per-round ratio. M5 Pro, Node 22.13.1, macOS only. The crawl output is identical to main's on this app.

Metric	main	PR	Change	95% CI of change
`FileMap.build()`	162.4 ± 3.0 ms	108.6 ± 2.2 ms	−33.3%	−34.2% to −31.2%
Warm start	19.7 ± 0.2 ms	19.4 ± 0.3 ms	−0.8%	−2.2% to +0.3% (unchanged)
Snapshot size	1,523 KB	1,523 KB	0%	snapshot is identical
Heap after first pass	24.16 MB	23.30 MB	−3.6%	same every run

Bold where the CI excludes zero.

Resolves conflicts with react#1677, which rewrote the Node crawler: its crawler already checks the extension before `lstat` and builds normal paths per directory, so this PR's crawler change reduces to skipping the `lstat` for a file with no previous mtime, reading `previousState` from the crawler options. The crawler tests follow: new files are now recorded without an `lstat`, the filesystem root test checks the path passed to `ignore()` rather than `lstat`, and `forceNodeFilesystemAPI` (removed by react#1693) is dropped from this PR's new tests. Also updates the API snapshot for `FileSystem.getMtimeByNormalPath`.

@robhogan

Summary: Publish Metro 0.87.2 from `main`. Changelog: [Internal] Test plan: ``` node scripts/updateVersion.js 0.87.2 sl diff --stat ``` Only the 17 `packages/*/package.json` manifests change, and every `0.87.1` reference is replaced. Draft release notes (from the 95 commits in v0.87.1...main - internal-only changes omitted): ``` - **[Feature]**: Add a built-in `metro:runtime/*` specifier, resolvable from any importer, eg for `transformer.asyncRequireModulePath` (#2083 by @robhogan) - **[Feature]**: When `transformer.enableBabelRuntime` is `true`, pass Metro's own `@babel/runtime` (`metro:babel-runtime`) and its installed version to Babel presets via caller data (#1957 by @robhogan) - **[Fix]**: Resolver: Fix stale resolutions after adding a nearer `package.json` or retargeting a symlinked directory, which previously needed a restart (#1948 by @robhogan) - **[Fix]**: Resolver: Packages and directories named with an asset extension (eg `lodash.zip`) resolve as modules rather than failing as assets (#2055 by @robhogan) - **[Fix]**: Resolver: Fall through to parent conditions when a nested `"exports"` or `"imports"` condition has no matching target (#2074 by @maximilliangrand) - **[Fix]**: Resolver: An exact `"exports"` or `"imports"` subpath with a `null` or unmatched target no longer resolves through a less specific subpath pattern (#1987 by @kwy404) - **[Fix]**: Use indexed watch folder paths (`[metro-watchFolders]/N/`) for assets outside `projectRoot` (#1892 by @robhogan) - **[Fix]**: Fix `/[metro-project]/` and `/[metro-watchFolders]/` bundle serving on Windows (#2015 by @motiz88) - **[Fix]**: Prevent HMR crash on `[metro-watchFolders]` or `[metro-project]` URLs (#2017 by @motiz88) - **[Fix]**: Remove erroneous `[metro-project]` and `[metro-watchFolders]` resolution of entry points from the API and CLI (#2030 by @motiz88) - **[Fix]**: Fix backslashes in async bundle paths on Windows (`includeAsyncPaths`) (#1938 by @robhogan) - **[Fix]**: Fix stale modules on the client when a file is deleted and recreated during a build, or changes during a build that fails (#2007 by @YevheniiKotyrlo) - **[Fix]**: Fix files created or modified while Metro starts up being missed until restart (#2054 by @cpruijsen) - **[Fix]**: Fix watch mode losing files when a directory replaces a file or symlink, eg when unlinking a package (#2009 by @robhogan) - **[Fix]**: Fix the fallback (Linux/Windows) watcher hanging on stop, and missing changes in a replaced directory, after a watch error (#1967 by @robhogan) - **[Fix]**: Fix the fallback watcher missing changes when a directory changes faster than Windows can report (#1967 by @robhogan) - **[Fix]**: Close the watch handles of a deleted directory and every directory under it in `FallbackWatcher`, which kept the process busy on Windows (#2004 by @YevheniiKotyrlo) - **[Fix]**: `metro-file-map`: Fix removing a non-active duplicate mock changing which mock is active (#1756 by @durvesh1992) - **[Fix]**: Propagate transformer initialisation errors from `Bundler` (#1854 by @Boulea7) - **[Fix]**: Release transform workers and file watchers when Metro fails to initialise (#1963 by @robhogan) - **[Fix]**: Ignore hyphenated and namespaced SVG attributes (eg `stroke-width`) when reading SVG asset dimensions (#1972 by @GabrielDumbrava) - **[Fix]**: Fix generated positions when iterating the mappings of indexed source maps, eg in `composeSourceMaps` (#1976 by @kwy404) - **[Fix]**: `composeSourceMaps` keeps every mapping at a generated column that the last map maps more than once (#1981 by @robhogan) - **[Fix]**: Source maps no longer lose lines after a gap of more than about two million unmapped generated lines (#1983 by @robhogan) - **[Fix]**: `metro-symbolicate` no longer fails on stack frames with no source position, such as `[native code]:0:0` in crash reporter output (#1891 by @robhogan) - **[Fix]**: Fix `metro-symbolicate` parsing of stack frames with Windows absolute paths (#1939 by @robhogan) - **[Performance]**: `metro-source-map`: Faster `Consumer` decoding and position lookups, and `composeSourceMaps` encodes its output directly - 3.2x faster composition of a bundle map with its Hermes map, using 15% less memory (#1978, #1979, #1980, #1983 by @robhogan) - **[Performance]**: Resolver: Ask for the package of a directory once for all extension candidates, and check `hierarchicalLookup` candidates from their own file map nodes. First-pass resolution is 35% faster (172ms -> 113ms on a mid-size RN app), net of a ~3% first-build cost from dropping `PackageCache`'s closest-package memo, which also cuts the resolver's retained heap by 37% (#1945, #1946, #1948 by @robhogan) - **[Performance]**: 33% faster non-Watchman file system crawl on startup (#1677 by @kitten) - **[Deprecated]**: `server.enhanceMiddleware` logs a deprecation warning when set to anything other than its default. Use `server.unstable_middleware` or `server.unstable_priorityMiddleware` (#2031 by @vzaidman) - **[Deprecated]**: `runServer`'s `unstable_extraMiddleware` option. Use `unstable_priorityMiddleware`, which runs the same middleware in the same place (#2031 by @vzaidman) - **[Deprecated]**: String values of `transformer.enableBabelRuntime` (use `true`) (#1957 by @robhogan) - **[Types]**: Add TypeScript definitions for `metro-runtime` modules, and generate (rather than hand-maintain) types for `metro-file-map` workers (#1996 by @robhogan) > NOTE: Experimental features are not covered by semver and can change at any time. - **[Experimental]**: Add `server.unstable_middleware` and `server.unstable_priorityMiddleware` config options, and `runServer` options of the same names, to add middleware after and before Metro's own. `mergeConfig` combines them from both configs: a later config's `unstable_middleware` runs last and its `unstable_priorityMiddleware` runs first (#2031 by @vzaidman) - **[Experimental]**: Add `server.unstable_onServerCreated` config option and `runServer` option of the same name, called with Metro's server instance before the HTTP server listens (#2031 by @vzaidman) - **[Experimental]**: Fix mismatched asset paths in (release) builds under `unstable_serverRoot`, make them always relative to `projectRoot` (#1917 by @robhogan) **Full Changelog**: v0.87.1...v0.87.2 ```

@robhogan

Summary: Publish Metro 0.87.2 from `main`. Changelog: [Internal] Test plan: ``` node scripts/updateVersion.js 0.87.2 sl diff --stat ``` Only the 17 `packages/*/package.json` manifests change, and every `0.87.1` reference is replaced. Draft release notes (from the 95 commits in v0.87.1...main - internal-only changes omitted): ``` - **[Feature]**: Add a built-in `metro:runtime/*` specifier, resolvable from any importer, eg for `transformer.asyncRequireModulePath` (#2083 by @robhogan) - **[Feature]**: When `transformer.enableBabelRuntime` is `true`, pass Metro's own `@babel/runtime` (`metro:babel-runtime`) and its installed version to Babel presets via caller data (#1957 by @robhogan) - **[Fix]**: Resolver: Fix stale resolutions after adding a nearer `package.json` or retargeting a symlinked directory, which previously needed a restart (#1948 by @robhogan) - **[Fix]**: Resolver: Packages and directories named with an asset extension (eg `lodash.zip`) resolve as modules rather than failing as assets (#2055 by @robhogan) - **[Fix]**: Resolver: Fall through to parent conditions when a nested `"exports"` or `"imports"` condition has no matching target (#2074 by @maximilliangrand) - **[Fix]**: Resolver: An exact `"exports"` or `"imports"` subpath with a `null` or unmatched target no longer resolves through a less specific subpath pattern (#1987 by @kwy404) - **[Fix]**: Use indexed watch folder paths (`[metro-watchFolders]/N/`) for assets outside `projectRoot` (#1892 by @robhogan) - **[Fix]**: Fix `/[metro-project]/` and `/[metro-watchFolders]/` bundle serving on Windows (#2015 by @motiz88) - **[Fix]**: Prevent HMR crash on `[metro-watchFolders]` or `[metro-project]` URLs (#2017 by @motiz88) - **[Fix]**: Remove erroneous `[metro-project]` and `[metro-watchFolders]` resolution of entry points from the API and CLI (#2030 by @motiz88) - **[Fix]**: Fix backslashes in async bundle paths on Windows (`includeAsyncPaths`) (#1938 by @robhogan) - **[Fix]**: Fix stale modules on the client when a file is deleted and recreated during a build, or changes during a build that fails (#2007 by @YevheniiKotyrlo) - **[Fix]**: Fix files created or modified while Metro starts up being missed until restart (#2054 by @cpruijsen) - **[Fix]**: Fix watch mode losing files when a directory replaces a file or symlink, eg when unlinking a package (#2009 by @robhogan) - **[Fix]**: Fix the fallback (Linux/Windows) watcher hanging on stop, and missing changes in a replaced directory, after a watch error (#1967 by @robhogan) - **[Fix]**: Fix the fallback watcher missing changes when a directory changes faster than Windows can report (#1967 by @robhogan) - **[Fix]**: Close the watch handles of a deleted directory and every directory under it in `FallbackWatcher`, which kept the process busy on Windows (#2004 by @YevheniiKotyrlo) - **[Fix]**: `metro-file-map`: Fix removing a non-active duplicate mock changing which mock is active (#1756 by @durvesh1992) - **[Fix]**: Propagate transformer initialisation errors from `Bundler` (#1854 by @Boulea7) - **[Fix]**: Release transform workers and file watchers when Metro fails to initialise (#1963 by @robhogan) - **[Fix]**: Ignore hyphenated and namespaced SVG attributes (eg `stroke-width`) when reading SVG asset dimensions (#1972 by @GabrielDumbrava) - **[Fix]**: Fix generated positions when iterating the mappings of indexed source maps, eg in `composeSourceMaps` (#1976 by @kwy404) - **[Fix]**: `composeSourceMaps` keeps every mapping at a generated column that the last map maps more than once (#1981 by @robhogan) - **[Fix]**: Source maps no longer lose lines after a gap of more than about two million unmapped generated lines (#1983 by @robhogan) - **[Fix]**: `metro-symbolicate` no longer fails on stack frames with no source position, such as `[native code]:0:0` in crash reporter output (#1891 by @robhogan) - **[Fix]**: Fix `metro-symbolicate` parsing of stack frames with Windows absolute paths (#1939 by @robhogan) - **[Performance]**: `metro-source-map`: Faster `Consumer` decoding and position lookups, and `composeSourceMaps` encodes its output directly - ~3.2x faster composition of a bundle map with its Hermes map, using ~15% less memory (#1978, #1979, #1980, #1983 by @robhogan) - **[Performance]**: Resolver: Ask for the package of a directory once for all extension candidates, and check `hierarchicalLookup` candidates from their own file map nodes. First-pass resolution is ~35% faster, net of a ~3% first-build cost from dropping `PackageCache`'s closest-package memo, which also cuts the resolver's retained heap by ~37% (#1945, #1946, #1948 by @robhogan) - **[Performance]**: ~33% faster non-Watchman file system crawl on startup (#1677 by @kitten) - **[Deprecated]**: `server.enhanceMiddleware` logs a deprecation warning when set to anything other than its default. Use `server.unstable_middleware` or `server.unstable_priorityMiddleware` (#2031 by @vzaidman) - **[Deprecated]**: `runServer`'s `unstable_extraMiddleware` option. Use `unstable_priorityMiddleware`, which runs the same middleware in the same place (#2031 by @vzaidman) - **[Deprecated]**: String values of `transformer.enableBabelRuntime` (use `true`) (#1957 by @robhogan) - **[Types]**: Add TypeScript definitions for `metro-runtime` modules, and generate (rather than hand-maintain) types for `metro-file-map` workers (#1996 by @robhogan) > NOTE: Experimental features are not covered by semver and can change at any time. - **[Experimental]**: Add `server.unstable_middleware` and `server.unstable_priorityMiddleware` config options, and `runServer` options of the same names, to add middleware after and before Metro's own. `mergeConfig` combines them from both configs: a later config's `unstable_middleware` runs last and its `unstable_priorityMiddleware` runs first (#2031 by @vzaidman) - **[Experimental]**: Add `server.unstable_onServerCreated` config option and `runServer` option of the same name, called with Metro's server instance before the HTTP server listens (#2031 by @vzaidman) - **[Experimental]**: Fix mismatched asset paths in (release) builds under `unstable_serverRoot`, make them always relative to `projectRoot` (#1917 by @robhogan) **Full Changelog**: v0.87.1...v0.87.2 ```

This fixes three regressions since 0.87.1, all in paths that reach a file or directory other than the way the file map stores it. ## Symlinked roots on a warm start Since #2064, `build()` drops cached files outside the configured roots on a cache hit, so that files from an earlier instance's `addRoot` don't survive. Watchman lists a root's files under its real path, though, so every file of a root given through a symlink (eg a watch folder like `node_modules/my-linked-lib`) is outside it. They're dropped on every warm start and not re-crawled, so imports into that package fail until `--reset-cache`. This also treats each root's real path as a root, at the cost of one `realpath` per root on a cache hit. A custom crawler's file system need not be the disk's, so its roots are taken as given. ## `hierarchicalLookup` through a symlink to an ancestor of the root Once the walk reaches an ancestor of the root, it took that ancestor's path by slicing `..` segments off the start path, which is wrong when the start path reaches it through a symlink instead (eg `app/monorepo -> ..`). That was a harmless miss until #1946 began resuming from the ancestor's node, since when the closest `package.json` of `app/monorepo/shared` is reported at `app/mo/package.json` and resolution fails with `ENOENT`. This uses the ancestor's canonical path, and the start path's real path relative to it. ## Unnormalised roots in the Node crawler Since #1677 the Node crawler builds paths by appending names to the root as given, so a root with a trailing separator (`path.join(__dirname, 'packages/')`) produced keys like `packages//a.js`, and one with `..` segments reached `ignore` unnormalised, defeating `blockList` patterns built from resolved paths. Roots are now resolved first. Changelog: Internal ## Test plan New unit tests for each, which fail without this change. With real Watchman on macOS, a watch folder linked from `node_modules` keeps all 3 of its files on a warm start, against 1 on `main` and 3 on 0.87.1. A differential fuzz of `hierarchicalLookup` (50k lookups in random trees with symlinks) against the code before #1946 and a simple real-path model: wherever this differs from the code before #1946, it matches the model. Those cases are the regression and lookups that previously threw `Failed to collapse`. The Node crawler on a real tree gives `sub/a.js` instead of `sub//a.js`, and now excludes a blocked `../shared/dist`.

Fixes three regressions since 0.87.1: - #2064 drops cached files outside the roots on a warm start. Watchman lists a symlinked root's files under its real path, so they're all dropped, and imports into it fail until `--reset-cache`: ```js watchFolders: [path.resolve(__dirname, 'node_modules/my-linked-lib')] ``` Real paths of roots now count too (built-in crawlers only). - #1946 broke `hierarchicalLookup` through a symlink in a root to its parent: ``` app/monorepo -> .. (watchFolders: ['/repo']) import './monorepo/shared' -> ENOENT: /repo/app/mo/package.json ``` - #1677 stopped the Node crawler normalising roots: ``` roots: ['/proj/sub/'] -> 'sub//a.js' roots: ['/proj/other/../shared'] -> blockList for '/shared/dist' ignored ``` Changelog: Internal Test plan: - New unit tests for each, failing without this change. - Real Watchman, watch folder linked from `node_modules`: 3/3 files on a warm start (`main`: 1/3). - `hierarchicalLookup` fuzz, 50k lookups in random symlinked trees: every difference from before #1946 matches a real-path model.

…2093) Fixes three regressions since 0.87.1: - #2064 drops cached files outside the roots on a warm start. Watchman lists a symlinked root's files under its real path, so they're all dropped, and imports into it fail until `--reset-cache`: ```js watchFolders: [path.resolve(__dirname, 'node_modules/my-linked-lib')] ``` Real paths of roots now count too (built-in crawlers only). - #1946 broke `hierarchicalLookup` through a symlink in a root to its parent: ``` app/monorepo -> .. (watchFolders: ['/repo']) import './monorepo/shared' -> ENOENT: /repo/app/mo/package.json ``` - #1677 stopped the Node crawler normalising roots: ``` roots: ['/proj/sub/'] -> 'sub//a.js' roots: ['/proj/other/../shared'] -> blockList for '/shared/dist' ignored ``` Changelog: Internal Test plan: - New unit tests for each, failing without this change. - Real Watchman, watch folder linked from `node_modules`: 3/3 files on a warm start (`main`: 1/3). - `hierarchicalLookup` fuzz, 50k lookups in random symlinked trees: every difference from before #1946 matches a real-path model.

kitten added 4 commits April 8, 2026 00:50

Create extensions dictionary

84922de

Append crawl child paths replacing repeated `pathUtils.absoluteToNorm…

b7550a0

…al` calls

Tweak formatting/prefer for-of

4523468

Flatten Promise in nodeCrawl (formatting-only)

76b6439

meta-cla Bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Apr 8, 2026

facebook-github-tools Bot added the Shared with Meta Applied via automation to indicate that an Issue or Pull Request has been shared with the team. label Apr 8, 2026

robhogan reviewed Apr 9, 2026

View reviewed changes

Comment thread packages/metro-file-map/src/crawlers/node/index.js Outdated

kitten force-pushed the @kitten/refactor/metro-file-map/tweak-node-crawler-hotpath branch from 59a90f5 to 416901a Compare April 9, 2026 11:37

This was referenced Apr 11, 2026

feat(metro-file-map): Lazily stat files and populate symlinks for Node crawled file trees #1686

Open

refactor(metro-file-map): Pre-resolve symlink targets and store normal posix paths #1687

Merged

kitten force-pushed the @kitten/refactor/metro-file-map/tweak-node-crawler-hotpath branch from ea59e3f to c46237d Compare April 11, 2026 12:59

kitten added 4 commits April 11, 2026 14:01

Remove overlapping roots in FileMap

5e32080

A previous implementation kept a `visited` set in the Node crawler. Instead, we can also deduplicate and remove overlapping roots (directories within other roots) in the `FileMap` preemptively for all crawlers to prevent duplicate crawls.

Fix crawling above root directory

b708850

Check file type and extname before stating

f85bea0

Rebuild ts-defs

c1cea5c

kitten force-pushed the @kitten/refactor/metro-file-map/tweak-node-crawler-hotpath branch from c46237d to c1cea5c Compare April 11, 2026 13:01

kitten mentioned this pull request Apr 22, 2026

refactor(metro-file-map): Drop "native find binary" crawler (and forceNodeFilesystemAPI flag) #1693

Closed

kitten added a commit to expo/expo that referenced this pull request May 5, 2026

refactor: Tweak Node crawler hot-path (react/metro#1677)

0fc1063

See: react/metro#1677

robhogan added 7 commits September 24, 2026 12:30

Merge react/metro main into tweak-node-crawler-hotpath

b7e0154

Resolves the conflict with react#1693, which removed the native `find` crawler: `findNative` is dropped, and `nodeCrawl` keeps this PR's flattened promise, calling `find` directly.

Drop removeOverlappingRoots

148015b

Keep `main`'s `Array.from(new Set(options.roots))` for now, leaving this PR to the crawler hot path. Deduplicating nested roots can follow separately. Also removes the committed `types/` declaration, which `main` no longer uses.

Compute the directory prefix once per root

5c0f7aa

Only a root can already end with a separator (a filesystem root), so check that once per root and pass each directory's prefix down to its children, instead of checking every directory as we recurse.

robhogan approved these changes Sep 24, 2026

View reviewed changes

robhogan changed the title ~~refactor: Tweak metro-file-map's Node crawler hot-path~~ performance: Improve metro-file-map's Node crawler hot-path Sep 24, 2026

robhogan merged commit d35efc4 into react:main Sep 24, 2026
15 checks passed

robhogan mentioned this pull request Oct 15, 2026

[0.84] Picks for 0.84.7 #2070

Draft

robhogan mentioned this pull request Oct 23, 2026

metro-file-map: Fix regressions with symlinks and unnormalised roots #2093

Merged

Navigation Menu

Repository navigation

performance: Improve metro-file-map's Node crawler hot-path - #1677

performance: Improve metro-file-map's Node crawler hot-path#1677
robhogan merged 15 commits into
react:mainfrom
kitten:@kitten/refactor/metro-file-map/tweak-node-crawler-hotpath

kitten commented Apr 8, 2026 •

edited by robhogan

Loading

Uh oh!

Uh oh!

robhogan commented Sep 24, 2026 •

edited

Loading

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

kitten commented Apr 8, 2026 • edited by robhogan Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Test plan

Uh oh!

Uh oh!

robhogan commented Sep 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Claude-driven benchmarks

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

kitten commented Apr 8, 2026 •

edited by robhogan

Loading

robhogan commented Sep 24, 2026 •

edited

Loading