|
* initial test case
* fix segfault from JSObjectMakeDeferredPromise
* pass exceptions through from FetchHeader.createFromJS
* not resolved, but getting close
* implement review suggestions
* fix exception check, tests
* Change how header filtering is accomplished
Previously the FetchHeaders implementation relied on converting names and values
to IDLByteString to catch non-ASCII data, though not always reliably. This
resulted in message-less TypeErrors when headers contained invalid characters.
This commit shifts everything to IDLDOMString for the conversion and relies on
the actual error checking in FetchHeaders.canWriteHeader, resulting in nicer
error messages.
To ensure that all headers are written as ASCII/UTF8 rather than UTF16, the
copyTo bindings function checks the encoding and converts if necessary.
* wrapping up FetchHeader fixes
* since utf8 allocates only do so when needed
* Update src/bun.js/bindings/bindings.cpp
Co-authored-by: Jarred Sumner <jarred@jarredsumner.com>
* WebCore__FetchHeaders__has should return on exception path
* strip out log calls from test
---------
Co-authored-by: Jarred Sumner <jarred@jarredsumner.com>
|