# These are just directly copied from soup-content-sniffer.c
# and could surely be improved.
# UTF-16 BOM
"\xFE\xFF"
# UTF-8 BOM
"\xEF\xBB\xBF"
# webm
"\x1A\x45\xDF\xA3"
# audio
".snd"
# aiff
"FORM\x00\x00\x00\x00AIFF"
# mpeg
"ID3"
# ogg
"OggS\x00"
# midi
"MThd\x00\x00\x00\x06"
# wave
"RIFF\x00\x00\x00\x00WAVE"
# avi
"RIFF\x00\x00\x00\x00AVI "
# HTML
"<!DOCTYPE HTML"
"<HTML"
# XML
"<?xml"