Thanks to Masami for his work tracking down this bug. I'm concerned about the patch applied to resolve this bug; the code should not segfault just because of bad datafiles; as long as the API is followed, errors should be detected (indeed, this will help detect invalid datafiles in a nicer way).
If you agree, please reopen this bug with severity: normal, and someone will get around to checking the return values as necessary. Justin References [0] http://bugs.debian.org/397312 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
