Hi omr,

Aha, yes you are correct. Using characters which are in ISO-8859-1 was
the key.

IE6
http://jscudtest.appspot.com/form_encoding_test?q=voil%E0%2C+%E7a+marche+pas

Firefox
http://jscudtest.appspot.com/form_encoding_test?q=voil%C3%A0%2C+%C3%A7a+marche+pas

I also tried your earlier suggestion of including

<meta http-equiv="content-type" content="text/html; charset=utf-8">

with an intentionally mismatched Content-Type HTTP header (still using
ISO-8859-1) and it seems that IE6 is still using ISO-8859-1 for
encoding. At this point I think the safest course of action is to
abandon form encoding and rely on JavaScript to escape the query
string.

Thank you,

Jeff

On Jul 1, 2:47 pm, omr <[email protected]> wrote:
> Jeff,
>
> Thanks for investigating this.
>
> Please try this query:
>
>    voilà, ça marche pas
>
> Try that in IE8, IE7, or IE6 on your example search page and accept-
> charset test page:
>
>    http://ajax-apis.appspot.com/html/two_page_search.html
>
>    http://jscudtest.appspot.com/form_encoding_test
>
> In this test IE encodes the characters à and ç in ISO-8859-1 (%E0,
> %E7), not UTF-8.
>
> This demonstrates that (as noted in my previous post) in such cases
> "Internet Explorer may not apply the encoding specified in a form's
> accept-charset attribute."
>
> .
>
> (Credit:
> In my previous post, I should have credited "FabryShock", a CSE user
> whose post in the Custom Search forum called attention to this issue.)
>
> -- omr

-- 
You received this message because you are subscribed to the Google Groups 
"Google AJAX APIs" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/google-ajax-search-api?hl=en.

Reply via email to