-
Notifications
You must be signed in to change notification settings - Fork 973
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
chore(test): Switch rwjs/auth over to vitest #10423
Merged
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Tobbe
force-pushed
the
tobbe-rw-auth-vitetst
branch
from
April 5, 2024 13:16
9e3622b
to
e1126f0
Compare
ahaywood
pushed a commit
that referenced
this pull request
Apr 5, 2024
dac09
added a commit
to dac09/redwood
that referenced
this pull request
Apr 9, 2024
…auth-provider-p1 * 'main' of github.com:redwoodjs/redwood: fix(middleware): Handle POST requests in middleware router too (redwoodjs#10418) chore(ci): get ci running on next (redwoodjs#10432) RSC: Explain noExternal vite config option (redwoodjs#10429) chore(web): Fix .d.ts overwrite build issue (redwoodjs#10431) chore(web): .js imports to prep for ESM (redwoodjs#10430) chore(refactor): Split rwjs/forms up into several smaller logical units (redwoodjs#10428) chore(rsc): simplify `noExternals` config (redwoodjs#10220) chore(deps): Update vite to 5.2.8 (redwoodjs#10427) chore(auth): Convert `@redwoodjs/auth` to ESM+CJS dual build (redwoodjs#10417) chore(framework-tools): Warn about missing metafile (redwoodjs#10426) chore(test): Switch rwjs/auth over to vitest (redwoodjs#10423) chore(whatwg-fetch): Switch to importing instead of requiring (redwoodjs#10424) chore(deps): bump undici from 5.28.3 to 5.28.4 in /.github/actions/check_changesets (redwoodjs#10421) chore(route-manifest): Add relativeFilePath to route manifest (redwoodjs#10416)
dac09
added a commit
that referenced
this pull request
Apr 11, 2024
…th-mw-auth * 'main' of github.com:redwoodjs/redwood: (21 commits) fix(auth): Handle when authorization header is lowercased (#10442) Update rbac.md - code match (#10405) chore: make crwa e2e test work across branches (#10437) feat: [Auth] Common AuthProvider & use* changes for middleware auth (#10420) fix(cli): only show webpack options for dev if `bundler = "webpack"` (#10359) fix(vercel): specify build env vars as a string (#10436) fix(vercel): write `vercel.json` as a part of setup (#10355) fix(middleware): Handle POST requests in middleware router too (#10418) chore(ci): get ci running on next (#10432) RSC: Explain noExternal vite config option (#10429) chore(web): Fix .d.ts overwrite build issue (#10431) chore(web): .js imports to prep for ESM (#10430) chore(refactor): Split rwjs/forms up into several smaller logical units (#10428) chore(rsc): simplify `noExternals` config (#10220) chore(deps): Update vite to 5.2.8 (#10427) chore(auth): Convert `@redwoodjs/auth` to ESM+CJS dual build (#10417) chore(framework-tools): Warn about missing metafile (#10426) chore(test): Switch rwjs/auth over to vitest (#10423) chore(whatwg-fetch): Switch to importing instead of requiring (#10424) chore(deps): bump undici from 5.28.3 to 5.28.4 in /.github/actions/check_changesets (#10421) ...
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Labels
changesets-ok
Override the changesets check
release:chore
This PR is a chore (means nothing for users)
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Switches
@redwoodjs/auth
over from jest to vitest.The one weird thing is I had to make
RWJS_API_GRAPHQL_URL
a full URL, or mock-service-worker would throw an error about invalid URL.require()
orimport
of whatwg-fetch also doesn't seem to make a difference, so I removed it. @jtoar and I spent some time looking into it yesterday, but couldn't get a full understanding exactly what was going on. It's something about whatRequest
is being used.MDN notes (emphasis mine)
So it sounds like what we had should have been allowed. And it was with jest. But now, with vitest, for some reason it isn't anymore. My hunch is that
require('whatwg-fetch')
orimport 'whatwg-fetch'
isn't polyfillingRequest
anymore (for some reason), so msw is using some other version ofRequest
that doesn't allow relative URLs. I haven't (yet) been able to verify that though.There's a StackOverflow question here that sounds like it might be related. But difficult to know for sure. https://stackoverflow.com/questions/76046546/fetch-error-typeerror-err-invalid-url-invalid-url-for-requests-made-in-test
The answer there is to update to the latest version of msw
Found this in the msw docs
The thing is though, that
document.baseURI
seems to already be polyfilled. So still not sure what's going on.On main
With vitest
So definitely a different one. Not sure why though.
I can "fix" it by doing this
But I would still very much like to understand what's actually going on here.
UPDATE:
vmContext
that it loads its modules invmContext
isglobalThis
for modules. Basically what "window
" is in a browser, butwindow.fetch
(reallyglobalThis.fetch
) is undefinedglobalThis.fetch
it polyfillsfetch
by attaching its own implementation offetch
toglobalThis
, and its own implementation ofRequest
etcimport
statement.globalThis.fetch
globalThis.fetch
, so it doesn't try to polyfill anythingSo that's the reason why there are two different versions of
Request
. And this is the reason why we need to manually setglobalThis.Request = RequestPolyfill
for vitest.Furthermore, you can easily reproduce the "invalid url" error for relative urls in the Node version of
Request
by simply runningnode
in your terminal and doingnew Request('/relative')
. As MDN states, relative urls are relative to the document'sbaseURI
, but there is no document in node, so that's why you need full urls.vitest does provide a value for
document.baseURI
, as we saw earlier (http://localhost:3000
). But my guess here is that node'sRequest
implementation doesn't even bother checkingdocument
, because it knows there will never be a document in the node environment. However, if you use a polyfill, like whatwg-fetch, they do look fordocument.baseURI
and so it all works.To keep the test (-environment) as close as possible to the jest environment we had I'm going to explicitly use the whatwg-fetch implementations of all its exports (rather than switching to a full url).