Different browsers return different origin values for such URLs.
For example, the results of parsing blob:blob:http://host:123/path are:
Browser or standard
origin
URL standard
http://host:123
Chromium
null
Firefox
http://host:123
Safari
blob://
I would prefer Chromium's behavior, even though it differs from the URL standard. There are two reasons for this:
Using the recursive origin's algorithm of the URL standard, it is easy to cause out of memory error by, for example, providing a URL with millions of blob: in it: blob:blob:blob: ... blob:http://host:123/path
Maybe I am wrong, but I think that a URL with many embedded blob: has no practical use.
Different browsers return different
origin
values for such URLs. For example, the results of parsingblob:blob:http://host:123/path
are:origin
http://host:123
null
http://host:123
blob://
I would prefer Chromium's behavior, even though it differs from the URL standard. There are two reasons for this:
blob:
in it:blob:blob:blob: ... blob:http://host:123/path
blob:
has no practical use.If you agree, I can prepare PR and tests.