Closes #20613.
Reimplements the serialization on top of the v8 APIs instead of
deno_core. Implements `v8.Serializer`, `v8.DefaultSerializer`,
`v8.Deserializer`, and `v8.DefaultSerializer`.
implement require(esm) using `op_import_sync` from deno_core.
possible future changes:
- cts and mts
- replace Deno.core.evalContext to optimize esm syntax detection
Fixes: https://github.com/denoland/deno/issues/25487
This commit removes all occurrences of `--unstable` flag
from all the tests that are run in CI.
Turns out none of the tests actually required that flag
anymore.
This PR addresses issue #25534
**Code Changes**
- Updated malva version to the latest in cli/Cargo.toml.
- Updated LanguageOptions to match new Malva config.
- Added test case same as the issue to assure changes success.
This commit improves error messages for unstable APIs:
- `--unstable-broadcast-channel`
- `--unstable-cron`
- `--unstable-http`
- `--unstable-kv`
- `--unstable-temporal`
By providing information and hints what went wrong and how the
error can be fixed. It reuses the same infra that was added in
https://github.com/denoland/deno/pull/21764.
This was initially added in #25399 in order to make transitioning over
from package.json to deno.json more easy, but it causes some problems
that are shown in the issue and it also means that the output of `deno
install` would have different resolution than `npm install`. Overall, I
think it's too much complexity to be smarter about this and it's
probably best to not do it. If someone needs an aliased folder then they
should keep using a package.json
Closes #25538
The long form "files" config has been flattened into the parent.
Old:
```json
{
"test": {
"files": {
"include": ["**/*.ts"],
"exclude": ["ignore.ts"]
}
}
}
```
New:
```json
{
"test": {
"include": ["**/*.ts"],
"exclude": ["ignore.ts"]
}
}
```
This was deprecated some time ago, but we're removing it now in Deno
2.0.
Closes #25415
A workaround for the issue #25480
`Deno.Listener` can't be closed synchronously after `accept()` is
called. This PR delays the `accept` call 2 ticks (The listener callback
is called 1 tick later. So the 1 tick delay is not enough), and makes
`net.Server` capable of being closed synchronously.
This unblocks `npm:detect-port` and `npm:portfinder`
closes #18301
closes #25175
Closes https://github.com/denoland/deno/issues/25321
Ended up being a larger refactoring, since we're now juggling
(potentially) two config files in the same `add`, instead of choosing
one. I don't love the shape of the code, but I think it's good enough
Some smaller side improvements:
- `deno remove` supports `jsonc`
- `deno install --dev` will be a really simple change
- if `deno remove` removes the last import/dependency in the
`imports`/`dependencies`/`devDependencies` field, it removes the field
instead of leaving an empty object
The map field has been empty for years now and we don't want the emit
file to be exposed so it allows us to iterate on making the cache
faster. Additionally, it's racy/unreliable to rely on this information.
Instead, people should emit the TS files themselves using tools like
deno_emit, typescript, esbuild, etc.
Closes https://github.com/denoland/deno/issues/17703
```
$ cat exports_error.js
Object.defineProperty(exports, "__esModule", { value: true });
$ deno exports_error.js
error: Uncaught (in promise) ReferenceError: exports is not defined
Object.defineProperty(exports, "__esModule", { value: true });
^
at file:///exports_error.js:1:23
info: Deno doesn't support CommonJS modules without `.cjs` extension.
hint: Rewrite this module to ESM or change the file extension to `.cjs`.
```
This commit adds support for executing top-level `.cjs` files,
as well as import `.cjs` files from within npm packages.
This works only for `.cjs` files, the contents of sibling `package.json`
are not consulted for the `"type"` field.
Closes https://github.com/denoland/deno/issues/25384
---------
Signed-off-by: David Sherret <dsherret@users.noreply.github.com>
Co-authored-by: Luca Casonato <hello@lcas.dev>
Co-authored-by: David Sherret <dsherret@users.noreply.github.com>
This commit adds:
- `addAbortListener` in `node:events`
- `aborted` in `node:util`
- `execPath` and `execvArgs` named export from `node:process`
- `getDefaultHighWaterMark` from `node:stream`
The `execPath` is very hacky - because module namespaces can not have
real getters, `execPath` is an object with a `toString()` method that on
call returns the actual `execPath`, and replaces the `execPath` binding
with the string. This is done so that we don't require the `execPath`
permission on startup.
Running a file like:
```
import "@std/dotenv/load";
```
Without a mapping in `imports` field of `deno.json` or `dependencies` of
`package.json`
will now error out with a hint:
```
error: Relative import path "@std/dotenv/load" not prefixed with / or ./ or ../
hint: Try running `deno add @std/dotenv/load`
at [WILDCARD]bare_specifier_without_import/main.ts:1:8
```
Closes https://github.com/denoland/deno/issues/24699
---------
Co-authored-by: David Sherret <dsherret@users.noreply.github.com>
This commit changes when to cause the hostname substition of `0.0.0.0` ->
`localhost`.
Currently we substitute `localhost` to the hostname on windows before
calling `options.onListen`, which prevents the users to do more advanced
thing using hostname string like
https://github.com/denoland/std/issues/5558. This PR changes it not to
substitute it when the user provide `onListen` callback.
closes #24776
unblocks https://github.com/denoland/std/issues/5558
This change fixes the handling of upgraded socket from `node:http` module.
In `op_node_http_fetch_response_upgrade`, we create DuplexStream paired
with `hyper::upgrade::Upgraded`. When the connection is closed from the
server, the read result from `Upgraded` becomes 0. However because we
don't close the paired DuplexStream at that point, the Socket object in
JS side keeps alive even after the server closed. That caused the issue
#20179
This change fixes it by closing the paired DuplexStream when the
`Upgraded` stream returns 0 read result.
closes #20179
This allows using npm deps of jsr deps without having to add them to the
root package.json.
Works by taking the package requirement and scanning the
`node_modules/.deno` directory for the best matching package, so it
relies on deno's node_modules structure.
Additionally to make the transition from package.json to deno.json
easier, Deno now:
1. Installs npm deps in a deno.json at the same time as installing npm
deps from a package.json.
2. Uses the alias in the import map for `node_modules/<alias>` for
better package.json compatiblity.
These now works:
```
$ deno add @std/dotenv/load
$ deno add npm:preact/hooks
```
Previously we were erroring out, because this is a "package reference"
including
a subpath.
Closes https://github.com/denoland/deno/issues/25385
---------
Signed-off-by: Bartek Iwańczuk <biwanczuk@gmail.com>
Co-authored-by: David Sherret <dsherret@users.noreply.github.com>
`--allow-run` even with an allow list has essentially been
`--allow-all`... this locks it down more.
1. Resolves allow list for `--allow-run=` on startup to an absolute
path, then uses these paths when evaluating if a command can execute.
Also, adds these paths to `--deny-write`
1. Resolves the environment (cwd and env vars) before evaluating
permissions and before executing a command. Then uses this environment
to evaluate the permissions and then evaluate the command.
This PR serves as a part of #22907 .
---------
Signed-off-by: David Sherret <dsherret@users.noreply.github.com>
Co-authored-by: Bartek Iwańczuk <biwanczuk@gmail.com>
Co-authored-by: David Sherret <dsherret@users.noreply.github.com>
Remove `--allow-hrtime` and `--deny-hrtime`. We are doing this because
it is already possible to get access to high resolution timers through
workers and SharedArrayBuffer.
Co-authored-by: Bartek Iwańczuk <biwanczuk@gmail.com>
Note: this is implemented on Deploy. However, according to @magurotuna,
a thin compatibility layer might be in the works that'd prevent
breakages for PRs such as this one.
Towards #22079
`deno bundle` now produces:
```
error: ⚠️ `deno bundle` was removed in Deno 2.
See the Deno 1.x to 2.x Migration Guide for migration instructions: https://docs.deno.com/runtime/manual/advanced/migrate_deprecations
```
`deno bundle --help` now produces:
```
⚠️ `deno bundle` was removed in Deno 2.
See the Deno 1.x to 2.x Migration Guide for migration instructions: https://docs.deno.com/runtime/manual/advanced/migrate_deprecations
Usage: deno bundle [OPTIONS]
Options:
-q, --quiet Suppress diagnostic output
--unstable Enable all unstable features and APIs. Instead of using this flag, consider enabling individual unstable features
To view the list of individual unstable feature flags, run this command again with --help=unstable
```
Before:
```
$ deno upgrade v1.xx
error: Invalid version passed
```
After:
```
$ deno upgrade v1.xx
error: Invalid version passed (v1.xx)
Example usage:
deno upgrade | deno upgrade 1.46 | deno upgrade canary
```
Also updates help text to use "shorthand version" without flags, but a
positional arg.
This commit effectively turns Deno into Deno 2.0.
This is done by forcing `DENO_FUTURE=1` env var, that was available in
the past few months to try Deno 2 changes.
This commit contains several breaking changes scheduled for Deno 2:
- all deprecated JavaScript APIs are not available any more, mostly
`Deno.*` APIs
- `window` global is removed
- FFI, WebGPU and FS APIs are now stable and don't require
`--unstable-*` flags
- import assertions are no longer supported
- "bring your own node modules" is enabled by default
This is the first commit in a series that are scheduled before the Deno
2 release.
Follow up work is tracked in
https://github.com/denoland/deno/issues/25241.
---------
Co-authored-by: Asher Gomez <ashersaupingomez@gmail.com>
Co-authored-by: Nayeem Rahman <nayeemrmn99@gmail.com>
Co-authored-by: Nathan Whitaker <nathan@deno.com>
This is blocking https://github.com/denoland/deno/pull/25213.
Turns out a bunch of FS APIs are completely broken because they
use RIDs (resource IDs) instead of FDs (file descriptors).
Fixes several tests blocking https://github.com/denoland/deno/pull/25213
by replacing `window` global that is gone in Deno 2 with `globalThis`.
Also adjusted a few tests using deprecated `rid` field.
Turns out we only virtualized it so one could have a `Console` property,
and the other one not. We can just make this `console.Console` available
everywhere.
Stores normalized version constraints in the lockfile, which will
improve reproducibility and will fix a bug with duplicate specifiers
ending up in the lockfile. Also, gets rid of some duplicate data in the
specifiers area of the lockfile.
Part of #20613.
If a node addon is using the legacy `napi_module_register` on ctor
approach to module registration, we have to store the registered module
so that other threads can load the addon (because `napi_module_register`
will only be called once per process).
This commit remove `--lock-write` that was deprecated in v1.45 release.
Closes https://github.com/denoland/deno/issues/24167.
---------
Co-authored-by: Asher Gomez <ashersaupingomez@gmail.com>
Fixes #23281. Part of #20613.
We were emitting the `online` event in the constructor, so the caller
could never receive it (since there was no time for them to add a
listener). Instead, emit the event where it's intended – after the
worker is initialized.
---
After this parcel no longer freezes, but still will fail due to other
bugs (which will be fixed in other PRs)
Fix task names containing a colon not being found with `deno run`. We
were only checking for a `module not found` error message, but strings
containing a colon throw a different error.
Fixes https://github.com/denoland/deno/issues/25232
- Add missing exports to `node:cluster`
- Fix default export not being an instance of `EventEmitter`
- Fix aliasing of properties
- Fix `disconnected` -> `disconnect` export naming
This makes `log4js` work in Deno. `karma` starts too, but somehow the
server isn't responding. That looks like a different issue.
Fixes https://github.com/denoland/deno/issues/24858
This commit enables the `log` feature for the `tracing` crate.
This allows us to examine additional detailed logs emitted by third party crates
that use `tracing` crate for logging by setting `RUST_LOG` env var or passing
`-L` option in command line.
Closes #25045
Fixes #24607.
This PR makes the logic that caches top level dependencies (things
present in import map) smarter, so we handle JSR dependencies without
root exports.
This commit deprecates "import assertions" proposal that has been
replaced with "import attributes".
Any time an import assertion is encountered a warning will be printed
to the terminal. This warning will be printed for both local and
remote files (ie. user code and dependencies).
Import assertions support will be removed in Deno 2.
Permission flags are unified in a clearer and concise output.
Unstable flags are hidden by default with exception of the `unstable`
flag itself. the remaining unstable flags can be seen with a
`--help=unstable`.
This also cleans up to show unstable flags only for subcommands that
actually need them.
Also sorts flags alphabetically, and gorups various flags together in a
set of categories.
---------
Co-authored-by: crowlkats <crowlkats@toaxl.com>
This PR addresses a regression introduced in
https://github.com/denoland/deno/pull/25021 that would cause the
`req.url` parameter in Node's http server to always be a single
character instead of the expected value. The regression was caused by
effectively calling `.indexOf()` on an empty string and thus passing the
wrong index for slicing.
```js
"".indexOf("/") // -> -1
request.url.slice(-1) // effectively only giving us the last character
```
Fixes https://github.com/denoland/deno/issues/25080
PrismJS uses `WorkerGlobalScope` and `self` for detecting browser's Web
Worker context:
59e5a34713/prism.js (L11)
Now the detection logic above is broken when it's imported from Deno's
Web Worker context because we only hide `self` (Prism assumes when
`WorkerGlobalScope` is available, `self` is also available).
This change fixes the above by also hiding `WorkerGlobalScope` global in
Node compat mode.
closes #25008
My fix in #25030 was buggy, I forgot to pass the `byteOffset` and
`byteLength`. Whoops.
I also discovered that fs.read was not respecting the `offset` argument,
and we were constructing a new `Buffer` for the callback instead of just
passing the original one (which is what node does, and the @types/node
definitions also indicate the callback should get the same type).
Fixes #25028.
This change improves the Node.js compatibility in managed npm resolution
mode by disabling the discovery of `node_modules` when the
main specifier is inside of `DENO_DIR`.
closes #22732
closes #24589
MacOS tends to not distribute evenly, so just don't assert that requests
were served by > 1 worker on mac.
Also added a check that all workers start, so we at least check that on
macos.
From upgrading `deno_lint`.
Previously if you had a node project that used a bunch of node globals
(`process.env`, etc), you would have to fix the errors by hand. This PR
includes a new lint that detects usages of node globals (`process`,
`setImmediate`, `Buffer`, etc.) and provides an autofix to import the
correct value. For instance:
```ts
// main.ts
const _foo = process.env.FOO;
```
`deno lint` gives you
```ts
error[no-node-globals]: NodeJS globals are not available in Deno
--> /home/foo.ts:1:14
|
1 | const _foo = process.env.FOO;
| ^^^^^^^
= hint: Add `import process from "node:process";`
docs: https://lint.deno.land/rules/no-node-globals
Found 1 problem (1 fixable via --fix)
Checked 1 file
```
And `deno lint --fix` adds the import for you:
```ts
// main.ts
import process from "node:process";
const _foo = process.env.FOO;
```
Linux/macos only currently.
Part of https://github.com/denoland/deno/issues/23524 (fixes it on
platforms other than windows).
Part of #16899 (fixes it on platforms other than windows).
After this PR, playwright is functional on mac/linux.
Adds a `parallel` flag to `deno serve`. When present, we spawn multiple
workers to parallelize serving requests.
```bash
deno serve --parallel main.ts
```
Currently on linux we use `SO_REUSEPORT` and rely on the fact that the
kernel will distribute connections in a round-robin manner.
On mac and windows, we sort of emulate this by cloning the underlying
file descriptor and passing a handle to each worker. The connections
will not be guaranteed to be fairly distributed (and in practice almost
certainly won't be), but the distribution is still spread enough to
provide a significant performance increase.
---
(Run on an Macbook Pro with an M3 Max, serving `deno.com`
baseline::
```
❯ wrk -d 30s -c 125 --latency http://127.0.0.1:8000
Running 30s test @ http://127.0.0.1:8000
2 threads and 125 connections
Thread Stats Avg Stdev Max +/- Stdev
Latency 239.78ms 13.56ms 330.54ms 79.12%
Req/Sec 258.58 35.56 360.00 70.64%
Latency Distribution
50% 236.72ms
75% 248.46ms
90% 256.84ms
99% 268.23ms
15458 requests in 30.02s, 2.47GB read
Requests/sec: 514.89
Transfer/sec: 84.33MB
```
this PR (`with --parallel` flag)
```
❯ wrk -d 30s -c 125 --latency http://127.0.0.1:8000
Running 30s test @ http://127.0.0.1:8000
2 threads and 125 connections
Thread Stats Avg Stdev Max +/- Stdev
Latency 117.40ms 142.84ms 590.45ms 79.07%
Req/Sec 1.33k 175.19 1.77k 69.00%
Latency Distribution
50% 22.34ms
75% 223.67ms
90% 357.32ms
99% 460.50ms
79636 requests in 30.07s, 12.74GB read
Requests/sec: 2647.96
Transfer/sec: 433.71MB
```
This commit adds capability to format HTML, Svelte, Vue, Astro and Angular
files.
"--unstable-html" is required to format HTML files, and "--unstable-component"
flag is needed to format other formats. These can also be specified in the config
file.
Close #25015
This commit adds another test suite that runs all Deno unit tests
with `DENO_FUTURE=1` flag to ensure all APIs are working as
expected, once Deno 2 is released.
---------
Co-authored-by: Divy Srivastava <dj.srivastava23@gmail.com>
Part of #25028.
Our underlying read/write operations in `io` assume the buffer is a
Uint8Array, but we were passing in other typed arrays (in the case above
it was `Int8Array`).
There is no constructor code when creating an inspector `Session`
instance in Node. Also get rid of some symbols which should've been
private properties. This PR doesn't yet add any new implementations
though as these are mostly cosmetic changes.
For some reason we didn't register the `node:inspector` module, which
lead to a panic when trying to import it. This PR registers it.
Related: https://github.com/denoland/deno/issues/25004
This commit stabilizes HMR functionality and renames
`--unstable-hmr` to `--watch-hmr`. The `--unstable-hmr`
flag is still working, but hidden from the help output.
It will be removed in Deno 2.
Once https://github.com/denoland/deno/pull/24958 lands
we should improve grouping of `--watch` and `--watch-hmr`
flags.
This PR integrates [Malva](https://github.com/g-plane/malva) into `deno
fmt`, which introduces the ability to format CSS, SCSS, Sass and Less
files.
On Linux x64 6.10, this PR increases about 800KiB:
```
❯ wc -c target/release/deno
125168728 target/release/deno
❯ wc -c target/release/deno
124349456 target/release/deno
```
This is commonly used to register loading non standard file types. But
some libs also register TS loaders which Deno supports natively, like
the npm `payload` package. This PR unblocks those.
Fixes https://github.com/denoland/deno/issues/24902
This commit adds "--serve" flag to "deno init" subcommand,
that provides a template for quick starting a project using
"deno serve".
---------
Co-authored-by: Asher Gomez <ashersaupingomez@gmail.com>
This commit makes `fetch` error messages include source and destination TCP
socket info i.e. port number and IP address for better debuggability.
Closes #24922
This commit updates permission prompt to add a link
to the documentation for particular flag.
Additionally the box drawings around the prompt have been slightly
altered for better visibility.
<img width="737" alt="Screenshot 2024-08-08 at 12 42 32"
src="https://github.com/user-attachments/assets/6ae748cd-4f29-439d-b0ee-f28f565f211a">
This PR ensures that we forward a `rename` event in our file watcher.
The rust lib we use combines that with the `modify` event.
This fixes a compatibility issue with Node too, which sends the `rename`
event as well.
Fixes https://github.com/denoland/deno/issues/24880
This completely rewrites how we handle key material in ext/node. Changes
in this
PR:
- **Signing**
- RSA
- RSA-PSS 🆕
- DSA 🆕
- EC
- ED25519 🆕
- **Verifying**
- RSA
- RSA-PSS 🆕
- DSA 🆕
- EC 🆕
- ED25519 🆕
- **Private key import**
- Passphrase encrypted private keys 🆕
- RSA
- PEM
- DER (PKCS#1) 🆕
- DER (PKCS#8) 🆕
- RSA-PSS
- PEM
- DER (PKCS#1) 🆕
- DER (PKCS#8) 🆕
- DSA 🆕
- EC
- PEM
- DER (SEC1) 🆕
- DER (PKCS#8) 🆕
- X25519 🆕
- ED25519 🆕
- DH
- **Public key import**
- RSA
- PEM
- DER (PKCS#1) 🆕
- DER (PKCS#8) 🆕
- RSA-PSS 🆕
- DSA 🆕
- EC 🆕
- X25519 🆕
- ED25519 🆕
- DH 🆕
- **Private key export**
- RSA 🆕
- DSA 🆕
- EC 🆕
- X25519 🆕
- ED25519 🆕
- DH 🆕
- **Public key export**
- RSA
- DSA 🆕
- EC 🆕
- X25519 🆕
- ED25519 🆕
- DH 🆕
- **Key pair generation**
- Overhauled, but supported APIs unchanged
This PR adds a lot of new individual functionality. But most importantly
because
of the new key material representation, it is now trivial to add new
algorithms
(as shown by this PR).
Now, when adding a new algorithm, it is also widely supported - for
example
previously we supported ED25519 key pair generation, but we could not
import,
export, sign or verify with ED25519. We can now do all of those things.
This PR updates `deno run` to fallback to executing tasks when there is
no script with the specified name. If there are both script and a task
with the same name then `deno run` will prioritise executing the script.
This moves YAML formatting behind an unstable flag for Deno 1.46. This
will make it opt-in to start and then we can remove the flag to make it
on by default in version of Deno after that.
This can be specified by doing `deno fmt --unstable-yaml` or by
specifying the following in a deno.json file:
```json
{
"unstable": ["fmt-yaml"]
}
```
- upgrade to v8 12.8
- optimizes DataView bigint methods
- fixes global interceptors
- includes CPED methods for ALS
- fix global resolution
- makes global resolution consistent using host_defined_options.
originally a separate patch but due to the global interceptor bug it
needs to be included in this pr for all tests to pass.
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
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.
This change aims to replace all relative import specifiers targeted at
`tests/util/std` with mapped ones (using a `deno.json` file). Towards
updating the `std` git submodule.
Two small changes:
- In our BYONM errors, suggest running `deno install` instead of `npm
install` if `DENO_FUTURE` is set
- Only emit warning about `deno install` changes if you do `deno install
<foo>` with deno_future unset
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
```
> deno upgrade
error: Unsupported lockfile version 'invalid'. Try upgrading Deno or recreating the lockfile.
V:\scratch
> V:\deno\target\debug\deno upgrade
Looking up latest version
Local deno version 1.45.3 is the most recent release
```
Closes #24517
Closes #20729
This pull request addresses an issue where the Error.cause property was
not formatted correctly when printed using console.log, leading to
confusion.
solution:
Implemented a fix to ensure that Error.cause is formatted properly when
printed by console.log, and the fix done by using JSON.stringify
This PR fixes https://github.com/denoland/deno/issues/23416
---------
Signed-off-by: MujahedSafaa <168719085+MujahedSafaa@users.noreply.github.com>
Currently, importing Node-Addons modules in a standalone binary results
in a `missing symbol called` error
(https://github.com/denoland/deno/issues/24614). Because the NAPI
symbols are not exported in this mode. This PR should fix the issue.
Regression from
04f9db5b22
Originally I thought to fix the issue in the PR we needed to explicitly
pass through the `node-modules-dir` flag, but after applying the correct
fix that david pointed out (setting `NPM_PROCESS_STATE`) that wasn't
necessary (or correct).
We had a test for deno task with BYONM, but it only tested with
`"unstable": ["byonm"]` in deno.json, so it didn't catch this.
