summaryrefslogtreecommitdiff
path: root/ext/node
AgeCommit message (Collapse)Author
2024-07-31chore: forward v1.45.5 release commit to main (#24818)denobot
2024-07-31fix(ext/node): handle node child_process with --v8-options flag (#24804)Luca Casonato
Makes `v8flags` package from NPM work.
2024-07-31fix(node/fs/promises): watch should be async iterable (#24805)Marvin Hagemeister
The way `fs.watch` works is different in `node:fs/promises` than `node:fs`. It has a different function signature and it returns an async iterable instead, see https://nodejs.org/api/fs.html#fspromiseswatchfilename-options Fixes https://github.com/denoland/deno/issues/24661
2024-07-31fix(node/timers/promises): add scheduler APIs (#24802)Marvin Hagemeister
This PR adds the experimental `scheduler` APIs in Node's `timers/promises` module. See https://nodejs.org/api/timers.html#timerspromisesschedulerwaitdelay-options Fixes https://github.com/denoland/deno/issues/24800
2024-07-30fix(node): Rework node:child_process IPC (#24763)Nathan Whitaker
Fixes https://github.com/denoland/deno/issues/24756. Fixes https://github.com/denoland/deno/issues/24796. This also gets vitest working when using [`--pool=forks`](https://vitest.dev/guide/improving-performance#pool) (which is the default as of vitest 2.0). Ref https://github.com/denoland/deno/issues/23882. --- This PR resolves a handful of issues with child_process IPC. In particular: - We didn't support sending typed array views over IPC - Opening an IPC channel resulted in the event loop never exiting - Sending a `null` over IPC would terminate the channel - There was some UB in the read implementation (transmuting an `&[u8]` to `&mut [u8]`) - The `send` method wasn't returning anything, so there was no way to signal backpressure (this also resulted in the benchmark `child_process_ipc.mjs` being misleading, as it tried to respect backpressure. That gave node much worse results at larger message sizes, and gave us much worse results at smaller message sizes). - We weren't setting up the `channel` property on the `process` global (or on the `ChildProcess` object), and also didn't have a way to ref/unref the channel - Calling `kill` multiple times (or disconnecting the channel, then calling kill) would throw an error - Node couldn't spawn a deno subprocess and communicate with it over IPC
2024-07-30perf(ext/node): improve `Buffer` from string performance (#24567)Divy Srivastava
Fixes https://github.com/denoland/deno/issues/24323 - Use a Buffer pool for `fromString` - Implement fast call base64 writes - Direct from string `create` method for each encoding op ``` $ deno bench -A bench.mjs # 1.45.1+fee4d3a cpu: Apple M1 Pro runtime: deno 1.45.1+fee4d3a (aarch64-apple-darwin) benchmark time (avg) (min … max) p75 p99 p999 ----------------------------------------------------------- ----------------------------- Buffer.from base64 550 ns/iter (490 ns … 1'265 ns) 572 ns 606 ns 1'265 ns Buffer#write base64 285 ns/iter (259 ns … 371 ns) 307 ns 347 ns 360 ns $ ~/gh/deno/target/release/deno bench -A bench.mjs # this PR cpu: Apple M1 Pro runtime: deno dev (aarch64-apple-darwin) benchmark time (avg) (min … max) p75 p99 p999 ----------------------------------------------------------- ----------------------------- Buffer.from base64 151 ns/iter (145 ns … 770 ns) 148 ns 184 ns 648 ns Buffer#write base64 62.58 ns/iter (60.79 ns … 157 ns) 61.65 ns 75.79 ns 141 ns $ node bench.mjs # v22.4.0 cpu: Apple M1 Pro runtime: node v22.4.0 (arm64-darwin) benchmark time (avg) (min … max) p75 p99 p999 ----------------------------------------------------------- ----------------------------- Buffer.from base64 163 ns/iter (96.92 ns … 375 ns) 99.45 ns 127 ns 220 ns Buffer#write base64 75.48 ns/iter (74.97 ns … 134 ns) 75.17 ns 81.83 ns 96.84 ns ```
2024-07-30refactor(ext/node): create separate ops for node:http module (#24788)Bartek Iwańczuk
This commit duplicates ops from "ext/fetch" to "ext/node" to kick off a bigger rewrite of "node:http". Most of duplication is temporary and will be removed as these ops evolve.
2024-07-29fix(ext/node): prevent panic in http2.connect with uppercase header names ↵Satya Rohith
(#24780) Closes https://github.com/denoland/deno/issues/24678
2024-07-28fix: actually add missing `node:readline/promises` module (#24772)David Sherret
Closes #24768
2024-07-26chore: forward v1.45.4 release commit to main (#24754)denobot
Co-authored-by: David Sherret <dsherret@gmail.com>
2024-07-26fix(ext/node): read correct CPU usage stats on Linux (#24732)MrEconomical
Fixes #24731 <img width="554" alt="deno_fixed" src="https://github.com/user-attachments/assets/691f2f89-d979-4ca5-be9a-cf51446cd9b2"> The total CPU usage row is ignored and info from `cpu0` and `cpu1` is correctly read. --------- Signed-off-by: MrEconomical <47700125+MrEconomical@users.noreply.github.com>
2024-07-25refactor: decouple node resolution from deno_core (#24724)David Sherret
2024-07-25fix(node/worker_threads): support `port.once()` (#24725)Marvin Hagemeister
Support `MessagePort.once` in Node mode and enable relevant `worker_threads` test. Noticed that another Node test was passing as well, so I enabled that too.
2024-07-24fix(ext/fetch): respect authority from URL (#24705)Bartek Iwańczuk
This commit fixes handling of "authority" in the URL by properly sending "Authorization Basic..." header in `fetch` API. This is a regression from https://github.com/denoland/deno/pull/24593 Fixes https://github.com/denoland/deno/issues/24697 CC @seanmonstar
2024-07-24fix(ext/node/net): emit `error` before `close` when connection is refused ↵Yoshiya Hinosawa
(#24656)
2024-07-24fix(node): Run node compat tests listed in the `ignore` field (and fix the ↵Nathan Whitaker
ones that fail) (#24631) The intent is that those tests will be executed, but our check that the files are up to date won't overwrite the contents of the tests. This is useful when a test needs some manual edits to work. It turns out we weren't actually running them. --- This ended up turning into a couple of small bug fixes to get the tests passing: - We weren't canonicalizing the exec path properly (it sometimes still had `..` or `.` in it) - We weren't accepting strings in `process.exit` There was one failure I couldn't figure out quickly, so I disabled the test for now, and filed a follow up issue: #24694
2024-07-24fix(node): better detection for when to surface node resolution errors (#24653)David Sherret
2024-07-23refactor: update to use deno_package_json (#24688)David Sherret
This is in preparation for extracting out node resolution code from ext/node (which is something I'm going to do gradually over time). Uses https://github.com/denoland/deno_package_json
2024-07-23chore: forward v1.45.3 release commit to main (#24681)denobot
This is the release commit being forwarded back to main for 1.45.3 --------- Co-authored-by: bartlomieju <bartlomieju@users.noreply.github.com> Co-authored-by: Bartek Iwańczuk <biwanczuk@gmail.com>
2024-07-19fix: missing `emitWarning` import (#24587)Marvin Hagemeister
We were missing an import of `emitWarning` in our streams implementation. The code prior to this PR assumed that `process` would be available as a global. Fixes https://github.com/denoland/deno/issues/23709 --------- Co-authored-by: Bartek Iwańczuk <biwanczuk@gmail.com>
2024-07-19fix(node): support `tty.hasColors()` and `tty.getColorDepth()` (#24619)Marvin Hagemeister
This PR adds support for [`tty.WriteStream.prototype.hasColors()`](https://nodejs.org/api/tty.html#writestreamhascolorscount-env) and [`tty.WriteStream.prototype.getColorDepth()`](https://nodejs.org/api/tty.html#writestreamgetcolordepthenv). I couldn't find any usage on GitHub which passes parameters to it. Therefore I've skipped adding support for the `env` parameter to keep our snapshot size small. Based on https://github.com/denoland/deno_terminal/pull/3 Fixes https://github.com/denoland/deno/issues/24616
2024-07-19fix(ext/node): do not expose `self` global in node (#24637)Yoshiya Hinosawa
closes #23727
2024-07-17fix(ext/node): stat.mode on windows (#24434)Divy Srivastava
2024-07-18Reland "refactor(fetch): reimplement fetch with hyper instead of reqwest" ↵Bartek Iwańczuk
(#24593) Originally landed in https://github.com/denoland/deno/commit/f6fd6619e708a515831f707438368d81b0c9aa56. Reverted in https://github.com/denoland/deno/pull/24574. This reland contains a fix that sends "Accept: */*" header for calls made from "FileFetcher". Absence of this header made downloading source code from JSR broken. This is tested by ensuring this header is present in the test server that servers JSR packages. --------- Co-authored-by: Sean McArthur <sean@seanmonstar.com>
2024-07-17perf(ext/node): optimize fs.exists[Sync] (#24613)Divy Srivastava
Use `access` on *nix and `GetFileAttributesW` on Windows. [Benchmark](https://paste.divy.work/p/-gq8Ark.js): ``` $ deno run -A bench.mjs # main (568dd) existsSync: 8980.636629ms $ target/release/deno run -A bench.mjs # this PR existsSync: 6448.7604519999995ms $ bun bench.mjs existsSync: 6562.88671ms $ node bench.mjs existsSync: 7740.064653ms ``` Ref https://github.com/denoland/deno/pull/24434#discussion_r1679777912
2024-07-16refactor(node): internally add `.code()` to node resolution errors (#24610)David Sherret
This makes it easier to tell what kind of error something is (even for deeply nested errors) and will help in the future once we add error codes to the JS errors this returns.
2024-07-16fix(ext/node): http request uploads of subarray of buffer should work (#24603)Satya Rohith
Closes https://github.com/denoland/deno/issues/24571
2024-07-15chore: upgrade dependencies (#24565)snek
rustls + hyper + deno_core
2024-07-15fix(std/http2): release window capacity back to remote stream (#24576)Rano | Ranadeep
This PR adds logic to release window capacity after reading the chunks from the stream. Without it, large response (more than `u16::MAX`) may fill up the capacity and the whole response can't be read. Closes https://github.com/denoland/deno/issues/24552 Closes https://github.com/denoland/deno/issues/24305
2024-07-13Revert "refactor(fetch): reimplement fetch with hyper instead of reqwest ↵Ryan Dahl
(#24237)" (#24574) This reverts commit f6fd6619e708a515831f707438368d81b0c9aa56. I'm seeing a difference between canary and 1.45.2. In `deno-docs/reference_gen` I can't download dax when running `deno task types` ``` ~/src/deno-docs/reference_gen# deno upgrade --canary Looking up latest canary version Found latest version f6fd6619e708a515831f707438368d81b0c9aa56 Downloading https://dl.deno.land/canary/f6fd6619e708a515831f707438368d81b0c9aa56/deno-aarch64-apple-darwin.zip Deno is upgrading to version f6fd6619e708a515831f707438368d81b0c9aa56 Archive: /var/folders/9v/kys6gqns6kl8nksyn4l1f9v40000gn/T/.tmpb5lDnq/deno.zip inflating: deno Upgraded successfully ~/src/deno-docs/reference_gen# deno -v deno 1.45.2+f6fd661 ~/src/deno-docs/reference_gen# rm -rf /Users/ry/Library/Caches/deno ~/src/deno-docs/reference_gen# deno task types Task types deno task types:deno && deno task types:node Task types:deno deno run --allow-read --allow-write --allow-run --allow-env --allow-sys deno-docs.ts error: JSR package manifest for '@david/dax' failed to load. expected value at line 1 column 1 at file:///Users/ry/src/deno-docs/reference_gen/deno-docs.ts:2:15 ~/src/deno-docs/reference_gen# deno upgrade --version 1.45.2 Downloading https://github.com/denoland/deno/releases/download/v1.45.2/deno-aarch64-apple-darwin.zip Deno is upgrading to version 1.45.2 Archive: /var/folders/9v/kys6gqns6kl8nksyn4l1f9v40000gn/T/.tmp3R7uhF/deno.zip inflating: deno Upgraded successfully ~/src/deno-docs/reference_gen# rm -rf /Users/ry/Library/Caches/deno ~/src/deno-docs/reference_gen# deno task types Task types deno task types:deno && deno task types:node Task types:deno deno run --allow-read --allow-write --allow-run --allow-env --allow-sys deno-docs.ts Task types:node deno run --allow-read --allow-write=. --allow-env --allow-sys node-docs.ts ```
2024-07-13refactor(fetch): reimplement fetch with hyper instead of reqwest (#24237)Sean McArthur
This commit re-implements `ext/fetch` and all dependent crates using `hyper` and `hyper-util`, instead of `reqwest`. The reasoning is that we want to have greater control and access to low level `hyper` APIs when implementing `fetch` API as well as `node:http` module. --------- Co-authored-by: Bartek Iwańczuk <biwanczuk@gmail.com>
2024-07-12chore: forward v1.45.2 release commit to main (#24564)denobot
Co-authored-by: dsherret <dsherret@users.noreply.github.com>
2024-07-12fix(ext/node): handle prefix mapping for IPv4-mapped IPv6 addresses (#24546)Satya Rohith
Ensure that the prefix is properly adjusted when dealing with IPv4 addresses mapped to IPv6. This fixes inconsistencies in network range calculations for mapped addresses. Closes https://github.com/denoland/deno/issues/24525
2024-07-11fix: unblock fsevents native module (#24542)snek
node-api is fully supported now. Signed-off-by: snek <snek@deno.com>
2024-07-12chore: forward v1.45.1 release commit to main (#24540)denobot
This is the release commit being forwarded back to main for 1.45.1 Co-authored-by: bartlomieju <bartlomieju@users.noreply.github.com>
2024-07-11fix(node): revert invalid package target change (#24539)David Sherret
Reason is that `e` may contain an invalid package target nested deeply in the returned errors. We should probably add a `.code()` to all errors to make matching easier or make the errors flatter.
2024-07-111.45.0 (#24512)denobot
Bumped versions for 1.45.0 --------- Co-authored-by: bartlomieju <bartlomieju@users.noreply.github.com> Co-authored-by: Bartek Iwańczuk <biwanczuk@gmail.com>
2024-07-10fix(check): CJS types importing dual ESM/CJS package should prefer CJS types ↵David Sherret
(#24492) Closes #16370
2024-07-10fix(node/perf_hooks): stub eventLoopUtilization (#24501)Marvin Hagemeister
This PR stubs `perf_hooks.eventLoopUtilization` to make the tests of [hapi](https://github.com/hapijs/hapi) start. Previously, they'd all error because of this function throwing a not implemented error. This brings down the test failures in their suite from 982 to 68 failures.
2024-07-10fix(node/v8): stub serializer methods (#24502)Marvin Hagemeister
Stubbing a few of Node's `v8` serializer methods makes the `tap` test runner at least start. Previously, it would fail to boot up as some instances of `DefaultSerializer` are constructed eagerly :tada: Before: <img width="1041" alt="Screenshot 2024-07-10 at 16 41 30" src="https://github.com/denoland/deno/assets/1062408/58f1157f-cef8-4176-9239-9d724ca0a677"> After: <img width="830" alt="Screenshot 2024-07-10 at 16 39 35" src="https://github.com/denoland/deno/assets/1062408/710ef673-8120-405a-b9d3-a5ca826b4829"> Fixes https://github.com/denoland/deno/issues/24409
2024-07-10fix: Add sys permission kinds for node compat (#24242)Adam Gregory
Fixes #24241 * Support "statfs", "username", "getPriority" and "setPriority" kinds for `--allow-sys`. * Check individual permissions in `node:os.userInfo()` instead of a single "userInfo" permission. * Check for "uid" permission in `node:process.geteuid()` instead of "geteuid". * Add missing "homedir" to `SysPermissionDescriptor.kind` union Co-authored-by: Bartek Iwańczuk <biwanczuk@gmail.com>
2024-07-10fix(node/http): don't send destroyed requests (#24498)Marvin Hagemeister
Make sure that already destroyed requests are not actually sent. This error was discovered in jsdom's test suite.
2024-07-10fix(node/http): don't error if request destroyed before send (#24497)Marvin Hagemeister
A request can be destroyed before it was even made in the Node http API. We errored on that. This issue was discovered in the JSDOM test suite.
2024-07-10feat(node): Support executing npm package lifecycle scripts ↵Nathan Whitaker
(preinstall/install/postinstall) (#24487) Adds support for running npm package lifecycle scripts, opted into via a new `--allow-scripts` flag. With this PR, when running `deno cache` (or `DENO_FUTURE=1 deno install`) you can specify the `--allow-scripts=pkg1,pkg2` flag to run lifecycle scripts attached to the given packages. Note at the moment this only works when `nodeModulesDir` is true (using the local resolver). When a package with un-run lifecycle scripts is encountered, we emit a warning suggesting things may not work and to try running lifecycle scripts. Additionally, if a package script implicitly requires `node-gyp` and it's not found on the system, we emit a warning. Extra things in this PR: - Extracted out bits of `task.rs` into a separate module for reuse - Added a couple fields to `process.config` in order to support `node-gyp` (it relies on a few variables being there) - Drive by fix to downloading new npm packages to test registry --- TODO: - [x] validation for allow-scripts args (make sure it looks like an npm package) - [x] make allow-scripts matching smarter - [ ] figure out what issues this closes --- Review notes: - This adds a bunch of deps to our test registry due to using `node-gyp`, so it's pretty noisy
2024-07-09fix: do not return undefined for missing global properties (#24474)snek
accessing e.g. `Buffer` in `Mode::Deno` mode should throw, not return undefined. --------- Signed-off-by: snek <snek@deno.com>
2024-07-09refactor: use concrete error types for node resolution (#24470)David Sherret
This will help clean up some of the code in the CLI because we'll be able to tell how the resolution failed (not part of this PR).
2024-07-09fix(node/http): support all `.writeHead()` signatures (#24469)Marvin Hagemeister
Implement the missing `.writeHead()` signatures from Node's `ServerResponse` class that we didn't support. Fixes https://github.com/denoland/deno/issues/24468
2024-07-08fix(node/assert): throws not checking error instance (#24466)Marvin Hagemeister
The implementation for `assert.throws()` from `node:assert` didn't work when the expected value was an `Error` constructor. In this case the thrown error should checked if it's an instance of said constructor. Fixes https://github.com/denoland/deno/issues/24464
2024-07-05fix(node): Implement `fs.lchown` (and `process.getegid`) (#24418)Nathan Whitaker
Closes https://github.com/denoland/deno/issues/21260. Part of https://github.com/denoland/deno/issues/18218. Implements `node:fs.lchown`, and enables the node_compat test for it. The test uses `process.getegid`, which we didn't have implemented, so I went ahead and implemented that as well to get the test working.
2024-07-05chore: remove unused class private properties (#24441)Marvin Hagemeister
Noticed that these private class properties are never used. Maybe a leftover from an earlier implementation.
generated by cgit v1.2.3 (git 2.25.1) at 2025-07-24 14:07:27 +0000