Kevin, you said: IMHO we should really keep the list that is checked to a minimum (being the list Windows checks) as this code is basically doing "If a web server reports a MIME type in this list, ignore it and make our own guess" which is totally wrong standards-wise and a major cause of many problems under IE on Windows..
(Apologies for the lack of quoting.) According to that appendix, IE should only be overriding MIME types that are text/plain, application/octet-stream, or empty. It does so because it considers these "ambiguous". I think your patch may be useful for the small set of types you suggest, but shouldn't you only override in these cases too?
--Juan
__________________________________ Do you Yahoo!? Yahoo! Mail - More reliable, more storage, less spam http://mail.yahoo.com
IE overrides MIME types for more then just ambiguous types, to quote one test mentioned in the appendix
"If the server-provided MIME type is either known or ambiguous, the buffer is scanned in an attempt to verify or obtain a MIME type from the actual content. If a positive match is found (one of the hard-coded tests succeeded), this MIME type is immediately returned as the final determination, overriding the server-provided MIME type (this type of behavior is necessary to identify a .gif file being sent as text/html)."
On Thursday 18 March 2004 02:58 pm, Juan Lang wrote:
(Apologies for the lack of quoting.) According to that appendix, IE should only be overriding MIME types that are text/plain, application/octet-stream, or empty. It does so because it considers these "ambiguous". I think your patch may be useful for the small set of types you suggest, but shouldn't you only override in these cases too?
Kevin Koltzau wrote:
IE overrides MIME types for more then just ambiguous types, to quote one test mentioned in the appendix
"If the server-provided MIME type is either known or ambiguous, the buffer is scanned in an attempt to verify or obtain a MIME type from the actual content. If a positive match is found (one of the hard-coded tests succeeded), this MIME type is immediately returned as the final determination, overriding the server-provided MIME type (this type of behavior is necessary to identify a .gif file being sent as text/html)."
On Thursday 18 March 2004 02:58 pm, Juan Lang wrote:
(Apologies for the lack of quoting.) According to that appendix, IE should only be overriding MIME types that are text/plain, application/octet-stream, or empty. It does so because it considers these "ambiguous". I think your patch may be useful for the small set of types you suggest, but shouldn't you only override in these cases too?
I really think we should not immitate the Windows behaviour here. It is causing nothing but security problems. If the server reports a MIME type, just use that type.
Shachar
-
Juan Lang -
Kevin Koltzau -
Shachar Shemesh