ASCII table with every control code explained

ASCII defines 128 characters numbered 0 to 127, while UTF-8 is an encoding that reproduces all 128 of them byte for byte and then keeps going to 1,114,112 code points. That compatibility was the design goal: a file containing only ASCII characters is already valid UTF-8, unchanged. The difference appears above 127, where ASCII simply stops and UTF-8 spends two to four bytes per character — é is a single byte 233 in Latin-1 but the two bytes 0xC3 0xA9 in UTF-8.

Because DEL is 127, all seven bits set, and on paper tape that meant punching out every hole in a column. A hole cannot be un-punched, so the only code that could overwrite any other was the one with every bit on — whatever was there became 1111111. That is why delete sits alone at the end of the chart instead of alongside the other control codes at the start.

No — 127 is DEL, a control character, so ASCII has 95 printable characters (32 to 126) and 33 control codes (0 to 31 plus 127). Terminals treat DEL as an editing instruction rather than a glyph, which is also why the Backspace key on most keyboards actually transmits 127 and not 8, the code named Backspace.

Because a terminal turns Ctrl plus a letter into that letter's code with the top three bits cleared, and C is 67, so Ctrl+C transmits 3 — ETX, End of Text. The same arithmetic explains the rest of the muscle memory: Ctrl+D is 4 (EOT, so end of input), Ctrl+S is 19 (XOFF, pause the stream, which is why it appears to freeze a terminal), Ctrl+Q is 17 (XON, resume) and Ctrl+[ is 27 (ESC).

Only five genuinely need escaping: & as &amp;, < as &lt;, > as &gt;, the double quote as &quot; inside a double-quoted attribute and the apostrophe as &apos; inside a single-quoted one. HTML5 defines named references for far more than that — this chart lists 155 of them, including &excl; for the exclamation mark and &Tab; for code 9 — but those are conveniences, not requirements. The numeric forms &#65; and &#x41; work for every code point whether a name exists or not.

They are not ASCII at all, so their meaning depends entirely on the encoding in use. ISO-8859-1 (Latin-1) puts Western European letters and punctuation in 160 to 255 and reserves 128 to 159 for C1 control codes that print nothing; Windows-1252 fills those 32 slots with curly quotes, the en and em dash, the euro sign and the ellipsis instead. The HTML standard instructs browsers to decode any page labelled ISO-8859-1 as Windows-1252 precisely because the mislabelling is so common, so this chart shows the Windows-1252 characters.

The bytes are correct UTF-8 being read as Latin-1 or Windows-1252. UTF-8 stores é as the two bytes 0xC3 0xA9; read one byte at a time under Latin-1, 0xC3 is à and 0xA9 is ©, so é appears on screen. Fix the declared encoding rather than the visible text, because re-saving the mangled version encodes the mistake a second time and makes the original much harder to recover.