Adds support for re-exporting an ES module from a CJS one and then
importing the CJS module from ESM. Also fixes a bug where require esm
wasn't working in deno compile.
We did not serialize the `AssignmentPattern` node inside `ObjectPattern`
properties.
```ts
({ a = b } = {})
```
This is a bit different in SWC and looks like I got confused with the
different AST formats.
Fixes https://github.com/denoland/deno/issues/28399
This adds support for installing `file:` dependencies in a local
package.json.
In order to use these, you must not set `--node-modules-dir=...` when
using a package.json and it should use the default of
`--node-modules-dir=manual`.
Closes https://github.com/denoland/deno/issues/18701
This PR adds support for `:has/:is/:where()` and `:not()`. The latter
was already present, but found a bunch of issues with it and I'd say
that it didn't really work before this PR.
Fixes https://github.com/denoland/deno/issues/28335
Internally, we use a group node for array-like children, which is hidden
from the user. When we're asking for a parent of a node we need to take
group nodes into account and walk over them.
Fixes one issue reported in
https://github.com/denoland/deno/issues/28355
This PR adds support for field selectors (`.<field>`) in the lint plugin
API. This is supported in ESLint as well, but was missing in our
implementation.
```css
/* Only search the test expression of an IfStatement */
IfStatement.test
```
Fixes https://github.com/denoland/deno/issues/28314
The code to support regex matching on attribute values was already
there. I just forgot to wire it up properly in the selector matching
code.
Fixes https://github.com/denoland/deno/issues/28336
Updates to use rust 1.85. Doesn't move to the 2024 edition, as that's a
fair bit more involved.
A nice side benefit is that the new rustc version seems to lead to a
slight reduction in binary size (at least on mac):
```
FILE SIZE
--------------
+4.3% +102Ki __DATA_CONST,__const
[NEW] +69.3Ki __TEXT,__literals
[NEW] +68.5Ki Rebase Info
+5.0% +39.9Ki __TEXT,__unwind_info
+57% +8.85Ki [__TEXT]
[NEW] +8.59Ki Lazy Binding Info
[NEW] +5.16Ki __TEXT,__stub_helper
[NEW] +3.58Ki Export Info
[NEW] +3.42Ki __DATA,__la_symbol_ptr
-0.1% -726 [12 Others]
-21.4% -3.10Ki [__DATA_CONST]
-95.8% -3.39Ki __DATA_CONST,__got
-20.9% -3.43Ki [__DATA]
-0.5% -4.52Ki Code Signature
-100.0% -11.6Ki [__LINKEDIT]
-1.0% -43.5Ki Symbol Table
-1.6% -44.0Ki __TEXT,__gcc_except_tab
-0.2% -48.1Ki __TEXT,__const
-3.3% -78.6Ki __TEXT,__eh_frame
-0.7% -320Ki __TEXT,__text
-1.5% -334Ki String Table
-0.5% -586Ki TOTAL
```
When all Rust-based rules where filtered out we were bailing out early
instead of checking if there are plugin rules we need to run. This meant
we errored out with a "No lint rules to run" message, even though plugin
rules were active.
Fixes https://github.com/denoland/deno/issues/28267
Fixes https://github.com/denoland/deno/issues/28223
This is kind of an ugly fix, but it works, and I think is the easiest
way to handle the fact that when caching the module graph we might
encounter imports that won't actually error at runtime (for instance in
files that will be bundled).
The `:exit` selectors were called at the wrong time during visiting.
They need to be called when going upwards and a node and all its
children have been fully visited. Instead we called it when the node +
all its sibling were visited which is wrong.
Fixes https://github.com/denoland/deno/issues/28227
Fixes https://github.com/denoland/deno/issues/28206.
Basically if you execute a script with `node:vm`, this produces a
"script" with the file name `evalmachine.<anonymous>`, which ends up
producing coverage like
```json
{
"scriptId": "319",
"url": "evalmachine.<anonymous>",
"functions": [
{
"functionName": "",
"ranges": [{ "startOffset": 0, "endOffset": 18, "count": 1 }],
"isBlockCoverage": true
}
]
}
```
We assume that the `url` field here (the specifier of the script) is a
valid URL, and so we error out when processing that coverage.
There are two potential fixes: either don't write the coverage files for
those scripts, or ignore the errors when we process the data. I went
with the former here.
Fixes https://github.com/denoland/deno/issues/27569.
Fixes https://github.com/denoland/deno/issues/27215.
This PR makes it so type resolution falls back to looking for definitely
typed packages (`@types/foo`) if a given NPM package does not contain
type declarations.
One complication is choosing _which_ version of the `@types/*` package
to use, if the project depends on multiple versions. The heuristic here
is to try to match the major and minor versions, falling back to the
latest version. So if you have
```
@types/foo: 0.1.0, 0.2.0, 3.1.0, 3.1.2, 4.0.0
foo: 3.1.0
```
we would choose `@types/foo@3.1.2` when resolving types for `foo`.
---
Note that this only uses `@types/` packages if you _already_ depend on
them. So a follow up to this PR could be to add a diagnostic and
quickfix to install `@types/foo` if we don't find types for `foo`.
Now that ArrayBuffer/ArrayBufferView is a generic Value type, we have to
handle it being passed any value. To do this, thread
FastApiCallbackOptions through the function, and add error raising
logic.
If we run conversion and the value is not valid, we return `isize::MAX`,
and then in cranelift we use this value to know that we should branch to
the error logic.
An example compilation looks like this:
```rust
extern "C" fn print_buffer(ptr: *const u8, len: usize);
```
```clif
function %print_buffer_wrapper(i64, i64, i64, i64) system_v {
sig0 = (i64, i64) system_v
sig1 = (i64) -> i64 system_v
sig2 = (i64) system_v
block0(v0: i64, v1: i64, v2: i64, v3: i64):
v4 = iconst.i64 0x6525_9198_2d00 ; turbocall_ab_contents
v5 = call_indirect sig1, v4(v1)
v6 = iconst.i64 0x7fff_ffff_ffff_ffff
v7 = icmp eq v5, v6
brif v7, block1, block2
block2:
v8 = iconst.i64 0x7558_4c0c_0700 ; sym.ptr
call_indirect sig0, v8(v5, v2)
return
block1 cold:
v9 = iconst.i64 0x6525_9198_2d70 ; turbocall_raise
call_indirect sig2, v9(v3)
return
}
```
Also cleaned up all the `unwrap`s and added some logging.
