Parsing N Tokens
rbrenton opened this issue · 1 comments
rbrenton commented
From http://ircd.bircd.org/bewarep10.txt
{83} many P10 services, including srvx, break on parsing N tokens with umode +h (sethost) because they don't parse the numeric and base64 IP counting from the end.
entrope commented
Strictly speaking, P10 doesn't have such a thing as +h user mode, so there's no reason to count backwards for those other fields. However, srvx has counted backwards since we migrated the code from arch in early 2004 (which was srvx 1.3).