Authentication Failure When Password Contains £ Character

This issue is because we pass “as is” the password received from client, to the LDAP server. In some cases we need to convert password from Unicode-codepoints to UTF-8 to fix this issue.

Unicode codepoints for the range 0x00000000 – 0x0000007F pose no trouble: for characters in this range, codepoint == UTF-8 encoding. All ASCII numerals, English alphabets fall in this range. This is the reason why plain English usernames/passwords are interpreted correctly by the receiving auth(LDAP) server.

Conversion is needed for characters in the range beginning with 0x00000080. Umlauts (like ä, ë, ï ) and Arabic characters fall in this range. For these characters, codepoint != UTF-8 encoding. So we need to convert the codepoint to UTF-8 before passing on to auth server.

Related:

Leave a Reply