In the dev server, we need to call `getStaticInfoIncludingLayouts` for the middleware file to extract its `matchers` field. However, that's currently executed in both app and pages workers. This method is expensive as it depends on the SWC binary to be loaded.
This PR changes it to invoke it as an IPC method, so only the router worker (which runs the compilation) loads SWC, instead of all 3 of them.
This also fixes duplicated console messages of:
```
Using locally built binary of @next/swc
```
For our test app, I'm seeing a 150~250ms improvement:
Haven't measured about the memory impact yet, but it should be a lot.
This makes it really easy to copy/paste the changelog into the PR description after running `sync-react.js`.
Previously, the output would link to incorrect PRs because it didn't know the PR number was from a different repo.
We know `facebook/react` uses squash so there should always be a PR number at the end of the commit title.
Example of the new format: https://github.com/vercel/next.js/pull/52005
This adapts the new route module rendering to support edge as well.
- Added a new `routeModule` export to the Edge SSR Loader
- Updated some tests to validate page state
Fixes NEXT-1327
When looking at [some sites](https://rsc-llm-on-the-edge.vercel.app/) with a large amount of chunks streamed, I noticed that the inlined Flight data array can be optimized quite a lot. Currently we do:
```js
self.__next_f.push([1,"d5:[\"4\",[\"$\",\"$a\",null,..."])
```
1. The `self.` isn't needed (except for the initial bootstrap tag) as React itself has `<script>$RC("B:f","S:f")</script>` too.
2. After the bootstrap script tag, all items are an array with `[1, flight_data]` and `flight_data` is always a string. We can just push only these strings.
3. We use `JSON.stringify(flight_payload)` to inline the payload where the payload itself is a string with a lot of double quotes (`"`), this results in a huge amount of backslashes (`\`). Here we can instead replace it to use a pair of single quotes on the outside and un-escape the double quotes inside.
Here's a side-by-side comparison of a small page:
<img width="1710" alt="CleanShot 2023-06-30 at 11 41 02@2x" src="https://github.com/vercel/next.js/assets/3676859/398356ec-91d5-435c-892d-16fb996029e8">
For a real production page I saw the HTML payload reduced by 11,031 bytes, a 3% improvement.
Note that all the tests are not considering gzip here, so the actual traffic impact will be smaller.
This PR changes some Webpack loaders to be synchronous as they don't have async code inside. Some of them will scale quiet a lot such as `next-flight-client-module-loader` and we don't want to waste some extra ticks there, as well as got potentially queued after some other events like file I/O.
We show the "Application error: a client-side exception has occurred (see the browser console for more information)" error incorrectly when a server-side error occurs (a digest is present) when we should be showing an error saying it is in fact a server side error and should check the logs there. This change will make the error message more accurate for users to look up
Fixes NEXT-1263
the metadata tests have an unnecessarily strict matcher which will fail if the matched metadata isn't the first instance of the element tag type. This will cause a problem when we start to preload bootstrap scripts which will cause preload links in all test cases that did not come from the metadata api itself
Co-authored-by: Jiachi Liu <4800338+huozhi@users.noreply.github.com>
Unlike other routes, these two entries are needed for every request so there is no reason to have them disposed. This avoids wasting extra dev server resource to get them re-compiled.
Increase `maxInactiveAge` to be `60` seconds and `pagesBufferLength` to be `5`. This change makes it infrequent to have expired entry compilations from recent accessed routes when running the dev server.
### What?
The updates `edge-runtime` to the latest version
### Why?
https://github.com/vercel/edge-runtime/pull/428 fixes `consumeUint8ArrayReadableStream` so that when we break iteration early (due to client disconnect), we cleanup the inner stream. That will fire the stream's `cancel` handler, and allow devs to disconnect from an AI service fetch.
### How?
`edge-runtime` maintain a `try {} finally {}` over the inner stream's iteration. When we early break, JS will call `it.return()`, and that will resume `consumeUint8ArrayReadableStream` with an abrupt completion (essentially, the `yield` turns into a `return`). We'll be able to trigger the `finally {}` block with that, and we call `inner.cancel()` to cleanup.
Fixes https://github.com/vercel-labs/ai/issues/90
Filter out the invalid images in metadata og/twitter `images` filter to avoid crash when falsy image slides in.
Add filtering for now as the erroring doesn't show the proper trace pointing to where it's original introduced, might introduce other validation in the future.
This PR documents the `NODE_ENV` environment variable behavior.
This feature was originally added in
152c2c2af3 which was released in `next@5`.
See the current source code here:
e3e76a45ee/packages/next/src/bin/next.ts (L82)
---------
## What?
Currently we use 3 separate webpack compilers:
- server
- client
- edge
All of these were creating their own input filesystem (which is used to
read file, stat, etc.). Changing them to share a single inputFileSystem
allows the `cachedFileSystem` to be reused better, as `stat` and
`readFile` can be cached across the 3 compilers.
For the page on vercel.com we've been testing this shaves off 300-400ms
on a cold compile (no cache, deleted `.next`).
<!-- Thanks for opening a PR! Your contribution is much appreciated.
To make sure your PR is handled as smoothly as possible we request that
you follow the checklist sections below.
Choose the right checklist for the change(s) that you're making:
## For Contributors
### Improving Documentation
- Run `pnpm prettier-fix` to fix formatting issues before opening the
PR.
- Read the Docs Contribution Guide to ensure your contribution follows
the docs guidelines:
https://nextjs.org/docs/community/contribution-guide
### Adding or Updating Examples
- The "examples guidelines" are followed from our contributing doc
https://github.com/vercel/next.js/blob/canary/contributing/examples/adding-examples.md
- Make sure the linting passes by running `pnpm build && pnpm lint`. See
https://github.com/vercel/next.js/blob/canary/contributing/repository/linting.md
### Fixing a bug
- Related issues linked using `fixes #number`
- Tests added. See:
https://github.com/vercel/next.js/blob/canary/contributing/core/testing.md#writing-tests-for-nextjs
- Errors have a helpful link attached, see
https://github.com/vercel/next.js/blob/canary/contributing.md
### Adding a feature
- Implements an existing feature request or RFC. Make sure the feature
request has been accepted for implementation before opening a PR. (A
discussion must be opened, see
https://github.com/vercel/next.js/discussions/new?category=ideas)
- Related issues/discussions are linked using `fixes #number`
- e2e tests added
(https://github.com/vercel/next.js/blob/canary/contributing/core/testing.md#writing-tests-for-nextjs
- Documentation added
- Telemetry added. In case of a feature if it's used or not.
- Errors have a helpful link attached, see
https://github.com/vercel/next.js/blob/canary/contributing.md
## For Maintainers
- Minimal description (aim for explaining to someone not on the team to
understand the PR)
- When linking to a Slack thread, you might want to share details of the
conclusion
- Link both the Linear (Fixes NEXT-xxx) and the GitHub issues
- Add review comments if necessary to explain to the reviewer the logic
behind a change
### What?
### Why?
### How?
Closes NEXT-
Fixes #
-->
---------
Co-authored-by: Shu Ding <g@shud.in>
remaining:
- add tests
caveat: this only works at the route level, we won't inherit from the layout or anything. I think that's fine
Co-authored-by: Florentin / 珞辰 <8146736+ecklf@users.noreply.github.com>
### What?
This PR fixes `next/link`'s `<Link />` missing many `<a />` props when
`experimental.typedRoutes` is enabled.
### How?
It does that by changing `AnchorHTMLAttributes<HTMLAnchorElement>` in
LinkRestProps to
`DetailedHTMLProps<AnchorHTMLAttributes<HTMLAnchorElement>,
HTMLAnchorElement>`, which contains all `<a />` props.
Fixes#51907
---------
Co-authored-by: Shu Ding <g@shud.in>
This PR changes `fs.stat` calls (to check if a file exists or not) to be
aggregated as `fs.readdir` calls and cached during the entire loader
pass. This is because we have a huge amount of file to check, but they
are always in a small amount of directories.
For example, a route that's 5-directory deep (`/a/b/c/d/page.js`) can
have up to 4,000 special segments and metadata files to check. However
they're always under these 5 directories. So instead of 4,000 `fs.stat`
calls let's just do 5 `fs.readdir` calls.
Didn't measure it carefully but a quick local test shows a 20~30%
improvement.
<details>
<summary>Another idea for future improvements</summary>
Currently, we create a MissingModuleDependency for any combination of
possible metadata filename. But in reality, we only need to track them
incrementally by index. For example if `icon1.js` is missing, it's kinda
waste to still track `icon2.js`~`icon9.js` because `icon1.js` will be
added first. This change should at least reduce the number of file
watchers by 80% and the initial compilation time by 0.5~1s, depending on
the actual route.
</details>
---------
Co-authored-by: Tim Neutkens <tim@timneutkens.nl>
