refs #10070.

## The behavior to review

This PR makes one choice: when a coarse outer map points to a column before the first mapping on the same line of a more detailed inner map, use that first same-line mapping instead of dropping the generated position.

Rollup makes the same choice when tracing segments. The fallback never crosses a line.

## Concrete before and after

### User input

`main.js` is the source written by the user:

```js
export function app() {
  globalThis.side = 1;
  return 42;
}
```

### Generated output

Rolldown produces this chunk. Region comments and the `sourceMappingURL` line are omitted here, but the code shape is otherwise the fixture's real output:

```js
function app() {
	globalThis.side = 1;
	return 42;
}
export { app };
```

The generated JavaScript is identical before and after this PR. Only its source map changes.

### Before this PR

Asking the composed map where the two generated statements came from returns no original position:

```text
generated 3:1  globalThis.side = 1  ──X──>  null
generated 4:1  return 42             ──X──>  null
```

In a source-map visualizer, these generated statements have no colored counterpart in the Original pane. A debugger or stack-trace consumer cannot take these positions back to `main.js`.

### After this PR

The same generated positions resolve to the corresponding user-written statements:

```text
Generated output                              User input: main.js

3:1  globalThis.side = 1  ─────────────────>  2:2  globalThis.side = 1
4:1  return 42             ─────────────────>  3:2  return 42
```

Source-map consumer lines are one-based and columns are zero-based. These are the actual positions asserted by the integration fixture.

In a visualizer, each generated statement now shares a color with its corresponding statement in the Original pane. Selecting either side can identify the other side. Debuggers and stack traces can return to the correct original line; the column is the best same-line position available from the coarse map.

## Why the lookup changes

The map chain contains two different levels of detail:

```text
detailed map:  intermediate column 2 -> user source
coarse map:    final column 0        -> intermediate column 0
```

Before:

```text
final column 0
  -> intermediate column 0
  -> detailed map has no mapping at or before column 0
  -> drop the generated position
```

After:

```text
final column 0
  -> intermediate column 0
  -> detailed map's first mapping on this line is column 2
  -> use column 2 and preserve the generated position
```

Lookups between or after existing mappings still use the nearest mapping at or before the requested column. Missing and empty lines still fail. One-field explicitly unmapped segments are skipped in this PR, matching Rollup; #10254 separately preserves those boundaries.

#10074 remains necessary because it makes Rolldown-owned mutation maps more precise. This PR covers coarse external maps and other cases where the producer did not provide matching columns.

## Why the snapshots change

The snapshot visualizer writes mappings as:

```text
(original line:column) "original text" --> (generated line:column) "generated text"
```

Unlike the source-map consumer positions above, these snapshot coordinates are zero-based for both lines and columns. The three snapshots only add mappings; no existing mapping is removed or changed.

| Snapshot | Newly restored generated mapping | Previous first mapping on that generated line | Original position used |
| --- | --- | --- | --- |
| `misc/wrapped_esm` | `60:1 "("`, `60:2 "{"` | `60:3 "e} = "` | `foo.js 4:4` |
| `misc/wrapped_esm` | `61:1 "("`, `61:2 "{"` | `61:3 "h: "` | `foo.js 7:2` |
| `misc/wrapped_esm` | `73:1 "("`, `73:2 "{"` | `73:3 "destructuring} = "` | `foo.js 30:6` |
| `wrapped_esm_default_function` | `19:0 "init_foo("` | `19:9 ");"` | `bar.js 0:23` |
| `wrapped_esm_export_named_function` | `18:0 "init_foo("` | `18:9 ");"` | `bar.js 0:27` |

There are eight added rows in total: six in `misc/wrapped_esm`, one in each deconflict snapshot. Every addition follows the behavior under review:

```text
new generated column < previous first mapped column on that line
new original position = previous first mapping's original position
```

For example:

```diff
+(0:23) ";\n" --> (19:0) "init_foo("
 (0:23) ";\n" --> (19:9) ");\n"
```

The coarse map requested generated column 0, while the detailed map's first mapping was at column 9. The fallback associates column 0 with the same original position already used at column 9. It does not move or replace the existing mapping.

## Suggested review order

1. Review the lookup change and `source_id` guards in `crates/rolldown_sourcemap/src/lib.rs`.
2. Review the Rust test where a column-0 coarse token meets a column-2 detailed token.
3. Review the `renderChunk` fixture, which exercises the public plugin path and asserts the exact positions shown above.
4. Check the snapshot table against the eight added snapshot rows.

## User impact

- generated JavaScript, runtime behavior, and bundle size do not change
- users of source maps regain original locations for lines previously dropped during composition
- the change matters when a plugin or another transform provides a less column-detailed map than Rolldown's preceding map
- already resolvable mappings are unchanged

## Validation

- `cargo test -p rolldown_sourcemap`
- `just test-rust`
- `just test-node-rolldown-only fixtures-concurrent.test.ts -t plugin/render-chunk/coarse-sourcemap-composition`
- `cargo clippy -p rolldown_sourcemap --all-targets -- --deny warnings`

## [1.2.0] - 2026-07-15

### 🚀 Features

- dev: skip shipping factories for newly imported top-level modules (#10223) by @h-a-n-a
- dev: per-client ship map for HMR patch sizing (#10208) by @h-a-n-a
- dev: client-side HMR (#10164) by @h-a-n-a
- dev: send a full-reload update to clients when a tsconfig changes (#10262) by @shulaoda
- treat `import.meta['url']` and `import.meta['ROLLUP_FILE_URL_*']` as side-effect free (#10267) by @sapphi-red
- rewrite `import.meta['url']` (#10251) by @sapphi-red
- add `FILE_NOT_FOUND` error (#10220) by @sapphi-red
- treat `import.meta.ROLLUP_FILE_URL_*` as side-effect free (#10217) by @sapphi-red

### 🐛 Bug Fixes

- sourcemap: preserve unmapped boundaries during composition (#10254) by @hyfdev
- `[format]` in `*FileNames` option for ESM format should be `es` instead of `esm` (#10214) by @sapphi-red
- sourcemap: preserve coarse mappings during composition (#10249) by @hyfdev
- rolldown_plugin_vite_import_glob: support tsconfig paths with `import.meta.glob` (#10167) by @sapphi-red
- dev: clear tsconfig caches for bare full builds (#10276) by @shulaoda
- dev: force a full rebuild when a tsconfig changes (#10261) by @shulaoda
- treat rooted drive-less module ids as absolute in preserveModules naming (#10235) by @IWANABETHATGUY
- watch: rebuild when tsconfig files change (#10258) by @shulaoda
- watch: drop tsconfig-merged transform options on each rebuild (#10257) by @shulaoda
- incorrect `EMPTY_IMPORT_META` warning for `import.meta.ROLLUP_FILE_URL_*` for CJS output (#10221) by @sapphi-red
- deconflict: rename CJS locals shadowing wrapped-ESM namespace objects (#9970) by @IWANABETHATGUY
- rolldown: drop the unused runtime module after entry-level external flattening (#10237) by @IWANABETHATGUY
- rolldown: re-propagate has_dynamic_exports to transitive star importers (#10239) by @IWANABETHATGUY
- tree-shaking: tree-shake destructured dynamic import namespace bindings (#10213) by @logaretm
- s390x: use json-escape-simd 3.1.1 for big-endian JSON escaping fix (#10211) by @satyamg1620

### 🚜 Refactor

- dev: move full-reload to client side (#10207) by @h-a-n-a
- readability follow-ups to the ReplaceWith migration (#10286) by @IWANABETHATGUY
- replace take_in-then-write-back with ReplaceWith and by-value moves (#10285) by @Boshen
- share the main resolver's cache with the transformer's tsconfig lookups (#10205) by @shulaoda
- rolldown: extract the ns star-external __reExport emission rule into LinkingMetadata (#10238) by @IWANABETHATGUY
- rolldown: unify link/generate diagnostics into a Diagnostics accumulator (#10234) by @IWANABETHATGUY
- sourcemap_filenames: drop dead sourcemap-filename plumbing (#10189) by @IWANABETHATGUY
- extract external import symbol merging into a method (#10224) by @IWANABETHATGUY
- rolldown: skip CJS namespace merging under strict execution order (#10203) by @hyfdev
- resolve the manual tsconfig per file instead of once at startup (#10200) by @shulaoda
- rolldown: route interop ESM init emission through a shared init-target view (#10202) by @hyfdev
- rolldown: collapse vestigial wrap-kind state and share chunk sort helper (#10201) by @hyfdev

### 📚 Documentation

- show plugin kinds in JSDoc and each hook's description (#10218) by @sapphi-red
- add an explanation about removing imports from external modules without any messages (#10215) by @sapphi-red

### ⚡ Performance

- sourcemap: owned merge in SourceJoiner::join (4005->5 allocs/chunk) (#10250) by @Boshen
- avoid redundant sourcemap string copies in collapse and minify paths (#10093) by @Boshen

### 🧪 Testing

- code-splitting: establish strict-order review baselines (#10287) by @hyfdev
- dev: add hot API test cases (#10181) by @h-a-n-a
- code-splitting: normalize strict execution order variants (#10277) by @hyfdev
- code-splitting: harden strict execution order coverage (#10252) by @hyfdev
- code-splitting: add strict execution order regressions (#10253) by @hyfdev

### ⚙️ Miscellaneous Tasks

- deps: update github actions (#10241) by @renovate[bot]
- deps: update oxc to 0.140.0 (#10274) by @shulaoda
- update Yunfei's GitHub username (#10275) by @hyfdev
- deps: update napi (#10260) by @renovate[bot]
- deps: update test262 submodule for tests (#10266) by @rolldown-guard[bot]
- deps: update dependency vite-plus to v0.2.4 (#10256) by @renovate[bot]
- deps: update napi (#10240) by @renovate[bot]
- deps: update oxc resolver to v11.24.2 (#10245) by @renovate[bot]
- deps: update rust crates (#10244) by @renovate[bot]
- disable Renovate updates for idna_adapter (#10248) by @shulaoda
- deps: update oxc resolver to v11.24.1 (#10232) by @renovate[bot]
- deps: update rust crate oxc_sourcemap to v8.1.1 (#10233) by @renovate[bot]
- deps: update dependency rolldown-plugin-dts to ^0.27.0 (#10206) by @renovate[bot]
- deps: upgrade sugar_path to v3 (#10230) by @hyfdev
- add `dist-*` to `.gitignore` in sourcemap-filenames/hash-final-content fixture (#10216) by @sapphi-red
- deps: update dependency rust to v1.97.0 (#10209) by @renovate[bot]

### ❤️ New Contributors

* @satyamg1620 made their first contribution in [#10211](#10211)

Co-authored-by: shulaoda <[email protected]>

…10368)

### Description

The `node-test-windows` job has been failing on every `main` run since 2026-07-15 (for example [this run](https://github.com/rolldown/rolldown/actions/runs/29795703407/job/88527096422)). The two fixtures added in #10249 and #10254 guard their `load` and `transform` hooks with `id.endsWith('/main.js')`. On Windows, module ids use backslashes, so the guard never matches and the hooks are silently skipped.

With the hooks skipped, `coarse-sourcemap-composition-issue-6399` bundles the on-disk comment-only `main.js` into an empty chunk whose `map` is `null`, so `new TraceMap(null)` throws `Cannot read properties of null (reading '_decodedMemo')`. In `explicit-unmapped-boundary`, the transform never injects `injected()`, so `code.indexOf('injected()')` returns -1 and the assertion fails.

This PR drops the leading slash so the checks become `id.endsWith('main.js')`, which is the pattern every other fixture in the repo already uses and works with both path separators. No production code is touched.