| Age | Commit message (Collapse) | Author |
|---|
|
Closes #18477
|
|
This reverts commit a3529d02329e0d2127ad2a5bb78b4c476ddd6984.
This change made debugging Node tests very hard - `AssertionError` is
now printed as `[Circular *1]` giving no visibility what failed.
We need to align two implementations together and remove this one then.
|
|
1. Fixes a cosmetic issue in the repl where it would display lsp warning
messages.
2. Lazily loads dependencies from the package.json on use.
3. Supports using bare specifiers from package.json in the REPL.
Closes #17929
Closes #18494
|
|
(#18505)
We currently run the all test cases in `parallel` category at the same
time, which invokes hundreds process at the same time, and that seems
causing some flakiness in CI. (maybe related to #18487)
This PR limits the concurrency to the number of cpu cores. This is more
aligned to how Node.js run their `parallel` test in their repository.
https://github.com/nodejs/node/blob/42c4a359525f70c322c0df0eac188fa2b05c3f53/Makefile#L356
|
|
Merge of dotland and dotcom caused this test to fail.
|
|
This commit adds associated type to "NodeEnv" trait, called "Fs".
The "Fs" type has a trait bound on "NodeFs", which specifies APIs
required for all ops and resolution APIs to function.
A "RealFs" implementation of "NodeFs" is exported from the "deno_node"
crate, that provides a default implementation for the trait.
All code in "deno_node" extension was changed to use the "NodeFs" trait
to handle file system operations, instead of relying on APIs from the
standard library.
|
|
I think this makes things clearer at the call sites.
|
|
|
|
|
|
|
|
A completely unrelated change in
https://github.com/denoland/deno/pull/18286 threw of this test.
These changes make it more robust and easier to update in such cases.
|
|
This commit changes the type parameter for "deno_node" extension, from
`P: NodePermission` to `Env: NodeEnv`.
`NodeEnv` is a new trait that has associated type `P: NodePermission`.
This is a stepping stone to support swappable file system for the
extension, that will be added as a second associated type to the
`NodeEnv` trait.
|
|
1. Rewrites the tests to be more back and forth rather than getting the
output all at once (which I believe was causing the hangs on linux and
maybe mac)
2. Runs the pty tests on the linux ci.
3. Fixes a bunch of tests that were just wrong.
4. Adds timeouts on the pty tests.
|
|
CacheDB (#18469)
Fast-follow on #18401 -- the reason that some tests were panicking in
the `CacheDB` `impl Drop` was that the cache itself was being dropped
during panic and the runtime may or may not still exist at that point.
We can reduce the actual tokio runtime testing to where it's needed.
In addition, we return the journal mode to `TRUNCATE` to avoid the risk
of data corruption.
|
|
|
|
Co-authored-by: Yoshiya Hinosawa <stibium121@gmail.com>
|
|
|
|
Towards #18455
This commit implements `checkPrimeSync` and `checkPrime` in node:crypto
using the Miller-Rabin primality test (fun fact: it actually is a test
for composite numbers)
It first compares the candidate against many known small primes and if
not, proceeds to run the Miller-Rabin primality test.
http://nickle.org/examples/miller-rabin.5c used as reference
implementation.
|
|
Towards #18455
|
|
Closes #18266
|
|
This gets SQLite off the flamegraph and reduces initialization time by
somewhere between 0.2ms and 0.5ms. In addition, I took the opportunity
to move all the cache management code to a single place and reduce
duplication. While the PR has a net gain of lines, much of that is just
being a bit more deliberate with how we're recovering from errors.
The existing caches had various policies for dealing with cache
corruption, so I've unified them and tried to isolate the decisions we
make for recovery in a single place (see `open_connection` in
`CacheDB`). The policy I chose was:
1. Retry twice to open on-disk caches
2. If that fails, try to delete the file and recreate it on-disk
3. If we fail to delete the file or re-create a new cache, use a
fallback strategy that can be chosen per-cache: InMemory (temporary
cache for the process run), BlackHole (ignore writes, return empty
reads), or Error (fail on every operation).
The caches all use the same general code now, and share the cache
failure recovery policy.
In addition, it cleans up a TODO in the `NodeAnalysisCache`.
|
|
This internal node hook is used by libraries such as `ts-node` when used
as a require hook `node -r ts-node/register`. That combination is often
used with test frameworks like `mocha` or `jasmine`.
We had a reference to `Module._preloadModules` in our code, but the
implementation was missing. While fixing this I also noticed that the
`fakeParent` module that we create internally always threw because of
the `pathDirname` check on the module id in the constructor of `Mdoule`.
So this code path was probably broken for a while.
```txt
✖ ERROR: Error: Empty filepath.
at pathDirname (ext:deno_node/01_require.js:245:11)
at new Module (ext:deno_node/01_require.js:446:15)
at Function.Module._resolveFilename (ext:deno_node/01_require.js:754:28)
at Function.resolve (ext:deno_node/01_require.js:1015:19)
```
|
|
This `Lazy<String>` calls were showing up on flamegraph and there's
really no point in using them.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Drive-by cleanup while I was looking into serialization of
BootstrapOptions.
There's no need to use non-default implementation in these places.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
This commit adds the `crypto.createSecretKey` API.
Key management: This follows the same approach as our WebCrypto
CryptoKey impl where we use WeakMap for storing key material and a
handle is passed around, such that (only internal) JS can access the key
material and we don't have to explicitly close a Rust resource.
As a result, `createHmac` now accepts a secret KeyObject.
Closes https://github.com/denoland/deno/issues/17844
|
|
|
|
The `getMany` method was missing from the implementation of the
`Deno.Kv` class. This patch fixes it.
|
|
These caused a bunch of unnecessary allocations on each startup.
|
|
than 7-bit ASCII (#18372)
This will improve diagnostics and catch any non-ASCII extension code
early.
This will use `debug_assert!` rather than `assert!` to avoid runtime
costs, and ensures (in debug_assert mode only) that all extension source
files are ASCII as we load them.
|
|
Co-authored-by: bartlomieju <bartlomieju@users.noreply.github.com>
Co-authored-by: Bartek Iwańczuk <biwanczuk@gmail.com>
|
|
Bug reported here shortly after merging `--ext` changes
https://github.com/denoland/deno/pull/17172#issuecomment-1480898098
Also found a missing `--check` in integration tests for `--ext` that
would have missed a bug if there was one.
Fixes #18392
|
|
entry text (#18366)
Fixes `Deno.inspect` to make an object literal non-compact when an entry
has multiple lines in it.
|
|
This will make it a bit harder to accidentally use a client url in the
wrong place. I don't fully understand why we do this mapping, but this
will help prevent bugs like #18373
Closes #18374
|
|
No need for two almost identical implementations of the same thing
---------
Co-authored-by: Yoshiya Hinosawa <stibium121@gmail.com>
Co-authored-by: Aapo Alasuutari <aapo.alasuutari@gmail.com>
|
|
There's no point in having `Lazy<Url>`, since the only use case
is string substitution in the "deno init" subcommand.
|
|
slash for workspace dir (#18373)
Closes https://github.com/denoland/vscode_deno/issues/827
|
|
Bumped versions for 1.32.0
---------
Co-authored-by: mmastrac <mmastrac@users.noreply.github.com>
Co-authored-by: Matt Mastracci <matthew@mastracci.com>
Co-authored-by: David Sherret <dsherret@users.noreply.github.com>
|
