### What does this PR do?
`JSPromise.rejectedPromiseValue` does not notify the VM about the promise it creates, meaning unhandled rejections created this way do not trigger `unhandledRejection`. This is leading to accidental error suppression in (likely) a lot of places. Additionally it returns a `JSValue` when really it should be returning a `*JSPromise`, making Zig bindings more type-safe.
This PR renames `rejectedPromiseValue` to `dangerouslyCreateRejectedPromiseValueWithoutNotifyingVM` and marks it as deprecated. It does _not_ modify code calling this function, meaning no behavior changes should occur. We should slowly start replacing its usages with `rejectedPromise`
## Changelog
- Rename `rejectedPromiseValue` to `dangerouslyCreateRejectedPromiseValueWithoutNotifyingVM`
- Mark `JSPromise.asValue` as deprecated. It takes a `*JSGlobalObject` but never uses it. New code should use `toJS()`
- Refactors `blob` to make null checks over `destination_blob.source` a release assertion
- `ErrorBuilder.reject` uses `rejectedPromiseValue` when 1.3 feature flag is enabled
* oooooh magic
* stuff
* run format
* ok
* yippee
* run the formatter back
* finish things up
* fix webkit
* more
* [autofix.ci] apply automated fixes
* fix compile
* fix compilation on windows, it seems to not work though :(
* update
* a
* v
* ok
* [autofix.ci] apply automated fixes
* OOPS
* bump bun to reduce ci bugs
* a
* js2native is done!
* improve array binding
* rebase
* some final stuff
* wasi fixes
* os
---------
Co-authored-by: Jarred Sumner <jarred@jarredsumner.com>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
* Simplify string handling code
* add extra + external
* Update fs.test.ts
* Bump
* woopsie
* prettier
* Rename stats() to resourceUsage()
* Fix leak
* Fix more leaks
* Setup malloc heap breakdown
* Thread safety
* Fix bug when creating buffer from utf-16 string
cc @dylan-conway
* Use global allocator
* More new
* Update fs.test.ts
* Update setTimeout.test.js
* Fix UAF in HTMLRewriter
* More bun.new
* Remove logs
* Un-skip test which no longer is flaky
* Even more `bun.new`
* Fix memory leak in HTMLRewriter.
Fixes#2325
* Don't accept Buffer for now
* Fix issue with node-fetch polyfill
* Don't destruct the response value too soon
---------
Co-authored-by: Jarred Sumner <709451+Jarred-Sumner@users.noreply.github.com>
* chore: upgrade zig to 0.12.0-dev.1828+225fe6ddb
* open as iterable
* fix building identifier cache
* fix windows build
* fix linux build
* fix linux build
* really lame prototype
* uses threads but badly
* it works i guess
* unwatchFile but lame
* it works
* test
* a
* aomitcs
* fix unwatching race condition
* use hasPendingActivity and GC stuff better
* test
* revert this
* half working disaster code
* this
* async copyFile
* .
* its failing symlink tests
* asdfg
* asdf
* hmm
* okay i think ti works
* small edits
* fix test on linux
* i hate atomics / atomics hate me back <3
* add a message in the builtins bundler that 0.8 is needed. it breaks on older versions lol.
* fixed
* rebase
* Fix unsafe GC behavior on large arrays returned by fs
* Fix crash in large arrays of strings
* async readdir
* Add tests for large number of files returned by readdir
* Move this down
* Fix encoding edgecase in path.join
* Async stat & lstat
* add test
---------
Co-authored-by: Jarred Sumner <709451+Jarred-Sumner@users.noreply.github.com>
* initial support
* add types
* fix comment
* fix types
* bigfix up
* more fixes
* fix some encoding support for watch
* fix rename event
* fixup
* fix latin1
* add fs_events, still failing some tests
* fixuup
* remove unecesary check
* readd tests ops
* this is necessary? just testing CI/CD weird errors
* just use dupe here
* cleanup and fix deinit
* fix zig upgrade
* Fix a bunch of bugs
* undo that one
* Fix crash in readdir()
* woops
* woops
* Add comment
* ✂️
* Make `readlink()` and `realpath` use much less memory
* Update BunString.cpp
* woopsie
* Unnecessary
* Don't commit these
---------
Co-authored-by: Jarred Sumner <709451+Jarred-Sumner@users.noreply.github.com>
