Summary: | "new EventSource()" should throw SYNTAX_ERR when an invalid argument is passed | ||||||
---|---|---|---|---|---|---|---|
Product: | WebKit | Reporter: | Kentaro Hara <haraken> | ||||
Component: | DOM | Assignee: | Nobody <webkit-unassigned> | ||||
Status: | RESOLVED DUPLICATE | ||||||
Severity: | Normal | CC: | abarth, ap, dominicc | ||||
Priority: | P2 | ||||||
Version: | 528+ (Nightly build) | ||||||
Hardware: | All | ||||||
OS: | All | ||||||
Attachments: |
|
Description
Kentaro Hara
2011-08-15 21:56:25 PDT
Created attachment 104011 [details]
Patch
Comment on attachment 104011 [details] Patch This patch does not appear correct, see: http://www.w3.org/TR/WebIDL/#es-DOMString In the null-or-undefined case, we hit step 3, which gives use "null" or "undefined", which is a valid relative URL. Therefore, we shouldn't throw a SYNTAX_ERROR in that case. Comment on attachment 104011 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=104011&action=review > LayoutTests/fast/eventsource/eventsource-constructor-exceptions-expected.txt:1 > +This tests an exception thrown by the EventSource constructor. This test cannot run interactively. Nit: Why can’t this run interactively? (In reply to comment #2) > (From update of attachment 104011 [details]) > This patch does not appear correct, see: > > http://www.w3.org/TR/WebIDL/#es-DOMString > > In the null-or-undefined case, we hit step 3, which gives use "null" or "undefined", which is a valid relative URL. Therefore, we shouldn't throw a SYNTAX_ERROR in that case. Thank you for the information. I confirmed that this is not a bug and thus should not be fixed. |