Comment #12 on issue 25790 by [email protected]: HTTP Basic / Digest auth fails when password contains non-ASCII characters http://code.google.com/p/chromium/issues/detail?id=25790
Thank you, Eric, for the summary. Also, sorry for missing your email. And, I agree with Wan-Teh that we have to talk to HTTPbis WG. I think Firefox "works"(??) as long as all the characters are *printable* characters <= U+00FF AND the server has password in Latin-1 before base64-encoding it. That's why the reporter said that Firefox "works" for him. Hmm, however, my theory and Edumund's observation are not consistent with your empirical findings. In case of IE, it appears to use the system code page. So, it will not work if the encoding used on the server-side does not match the OS default codepage as Eric wrote. In terms of going forward, I like what Chrome/Opera do (using UTF-8). In terms of compat with existing web sites (how critical this is is another question), doing one of three (or a series of them) may work better: 1. referrer's encoding (if one is available) 2. default charset of Chrome (which is UI-language dependent and a user can change) 3. OS default charset A rather crazy idea: Can we try all 3 of them plus UTF-8 until one of them works? Perhaps, it's not a good idea. If this is really critical in terms of compat, we *mi....ght* consider that, but even in that case, we'd better be careful with a security implication, I guess. -- You received this message because you are listed in the owner or CC fields of this issue, or because you starred this issue. You may adjust your issue notification preferences at: http://code.google.com/hosting/settings -- Automated mail from issue updates at http://crbug.com/ Subscription options: http://groups.google.com/group/chromium-bugs
