── WHAT COUNTS AS NON-ASCII?────────────────────────────────
ASCII covers exactly 128 characters: the English letters, digits, common punctuation and a set of control codes. Everything else — accented letters (é), curly quotes (“ ”), em dashes (—), the non-breaking space, every emoji — is non-ASCII. These characters are perfectly valid Unicode, but plenty of older or stricter systems still expect pure ASCII.
── WHY THEY BREAK THINGS──────────────────────────────────
The classic failure: you draft code or config in a word processor, which silently "beautifies" your straight quotes into curly ones — and the compiler rejects ‘smart quotes’ it can't parse. Same story with CSVs that won't import because of a non-breaking space, JSON that fails validation over an invisible zero-width character, email headers, legacy databases, and URLs.
[ AD SLOT · IN-ARTICLE · 728×90 RESERVED ]
── THE USUAL SUSPECTS──────────────────────────────────────
CHARCODE POINTUSUALLY COMES FROMSAFE REPLACEMENT
’ ‘U+2018/19Word, Google Docs autocorrect' straight quote
“ ”U+201C/1Dsame — "smart" double quotes" straight quote
—U+2014em dash autocorrect from ---- or -
␣U+00A0non-breaking space — copied from web pagesregular space
∅U+200Bzero-width space — invisible, from rich-text editorsdelete it
∅U+FEFFBOM — sneaks into the first byte of filessave without BOM
── FAQ────────────────────────────────────────────────────────
[+] Should I always remove non-ASCII characters?
No — in prose for humans, é and — are correct typography. Strip them only where a system demands ASCII: identifiers, config files, data pipelines, legacy protocols. The checker above shows you what's there; you decide.
[+] Is UTF-8 the same as ASCII?
UTF-8 is backwards-compatible with ASCII: the first 128 code points encode to identical bytes. That's why pure-ASCII text works everywhere — and why one smart quote is all it takes to make a file "not ASCII" anymore.
── RELATED TOOLS──────────────────────────────────────────────
[ AD SLOT · FOOTER · 728×90 RESERVED ]