Part of #18610.
Our intent is to move `definePlugin`, `defineRule`, and `eslintCompatPlugin` to a separate package `@oxlint/plugins` (plural to avoid confusion with naming convention for plugins `oxlint-plugin-xxx`).
So rename the current entry point which exports these functions from `oxlint/plugin` to `oxlint/plugins` (plural).
As we prepare for alpha release of JS plugins, it feels like we can remove the warning logged to stderr.
```
WARNING: JS plugins are experimental and not subject to semver.
Breaking changes are possible while JS plugins support is under development.
```
They do still remain experimental, but I don't think we need to shout that quite as loudly now.
This is technically breaking (only for types), but custom plugins are experimental.
The ruletester/defineRule/definePlugin are re-exported but with a `@deprecated` tag which I think is a good enough trade off
I want to avoid users bringing in a ton of custom plugin code when doing `import { defineConfig } from 'oxlint'`
This ensures that we print the expected number of rules for users that are using JS Plugins.
Fix#17519.
Based on #17517, because otherwise we'd end up with weird snapshot differences when trying to merge these anyway.
AI Disclosure: The fix here was generated with Copilot + GitHub's Raptor mini model.
For a while, we've been skipping the printing of rule runs unless `--disable-nested-config` was passed. This was a result of the nested_configs value including the base config (see #16356). When we were checking whether that value was empty, it would always have at least 1 item (unless nested configs were entirely disabled).
As part of this, I've updated some configs in the fixtures to disable correctness rules, otherwise we will end up with a bunch of fixture snapshots needing updates every time a new default rule is added.
It's probably easiest to review this one commit at a time.
Format `oxlint` test fixtures. The directory `fixtures` contains not just the fixtures themselves, but also Oxlint plugins, which should be formatted.
Skip 2 specific test fixture files which must not be formatted to maintain correctness of the tests.
Add `loc` to all AST nodes.
`loc` is computed lazily, as it's quite an expensive operation - because it requires searching the source text for line breaks and building a translation table.
We can consider making `loc` a "real" property once we have a faster translation table built on Rust side.
Export AST types directly from `oxlint` package, instead of depending on `@oxc-project/types`. This enables altering the types to reflect that `range` field is always present on AST node in Oxlint's AST (#14354).
Add `range` field to AST nodes, using the support for `range` in raw transfer added in #14319.
Use `range` field rather than `start` + `end` in nodes passed to `context.report` and fixer methods - for closer alignment with ESLint.
Avoid deserializing the AST if it's not required.
It's possible that all rules return an empty visitor, or that all `createOnce` rules return `false` from their `before` hook. In those cases, the AST does not need to be walked, so AST doesn't need to be deserialized. And the source text doesn't need to be decoded from buffer either.
These cases probably aren't so common, but not completely uncommon either.
However, source text and AST are both accessible before traversal via `context.sourceCode`. Enable the usage of those APIs in `create` method / `before` hook by decoding source text / deserializing AST on demand when those APIs are used.
Boshen