feat: add rootDir option and set __svelte_meta.file like in svelte4

[dominikg]

lots of tests fail because there are still ... in various expects and i havn't found a way to automatically update them.

tbd: not defaulting to process.cwd would make svelte more independant, but means bundler plugins will have to set it to avoid issues with path leaks

it gets complicated with error handling (stacktraces) and sourcemap merging that might need the absolute path still.
In general we should try to use the stripped filename everywhere in output, not just in dev mode like in this PR.

This is basically the minimum way i think it could be added, feel free to completely refactor/change/close this, but for svelte-inspector to work we need

__svelte_meta.file to be src/foo/Bar.svelte and not .../foo/Bar.svelte

Svelte 5 rewrite

Please note that the Svelte codebase is currently being rewritten for Svelte 5. Changes should target Svelte 5, which lives on the default branch (main).

If your PR concerns Svelte 4 (including updates to svelte.dev.docs), please ensure the base branch is svelte-4 and not main.

Before submitting the PR, please make sure you do the following

• It's really useful if your PR references an issue where it is discussed ahead of time. In many cases, features are absent for a reason. For large changes, please create an RFC: https://github.com/sveltejs/rfcs
• Prefix your PR title with feat:, fix:, chore:, or docs:.
• This message body should clearly illustrate what problems it solves.
• Ideally, include a test that fails without this PR but passes with it.

Tests and linting

• Run the tests with pnpm test and lint the project with pnpm lint

[changeset-bot]

🦋 Changeset detected

Latest commit: f767eb5

The changes in this PR will be included in the next version bump.

This PR includes changesets to release 1 package

Name	Type
svelte	Patch

Not sure what this means? Click here to learn what changesets are.

Click here if you're a maintainer who wants to add another changeset to this PR

[Rich-Harris]

lots of tests fail because there are still ... in various expects and i havn't found a way to automatically update them

working on this

[Rich-Harris]

I think this is ready to go, though there are some Windows failures that I don't really understand — forward slashes versus backward slashes — and will leave to someone else

[Rich-Harris]

oh and I guess we should summon @Conduitry in case he wants to veto the fallback to process.cwd()

[dominikg]

I think this is ready to go, though there are some Windows failures that I don't really understand — forward slashes versus backward slashes — and will leave to someone else

looks nice now. I did wonder about using state for it as well but cheaped out 👍

the windows path separator issue is really unfortunate. Do we want to go the lengths to normalize \ to / to avoid it?
To make only windows users pay the price we'd have to spy on process.platfrom too which adds yet another quirk.

[Rich-Harris]

I mean I don't understand how and where it's happening. surely filename and rootDir both use the same separators?

[dummdidumm]

Fixed it by normalizing slashes. The slashes were a mix of \ and / on windows because of this code in the test suite - I could've fixed that there instead, but I figured that such a mixup could happen in userland, too, so it's more robust to normalize slashes instead.