summaryrefslogtreecommitdiff
path: root/ISSUES/charset-sniffing.md
blob: 4ed70316438c408bfcc35bcc6fadb71ea43c3255 (plain)
1
2
3
# Optimize Charset Sniffing

Almost all charsets are supersets of ASCII, so when sniffing the charset for files which don't specify the encoding in their MIMEtype I can treat all the preceding text as ASCII. Though I suppose for this trick to work on UTF16 or UTF32 I'd need to remove any 0 bytes.