← X character counter

49 conformance vectors for X’s weighted length

Open test data for anyone building a character counter for X — including the cases almost everybody gets wrong. Take them, run them against your own code, and tell us if we are the ones who are wrong.

X does not count characters. It assigns each character a weight and allows 280 weight per post. Latin letters, digits and common punctuation weigh 1; nearly everything else — CJK, Cyrillic, Arabic, Devanagari, every emoji — weighs 2. Every URL costs a flat 23 whatever its real length.

That much is widely known and still not enough. The rule is grapheme clusters, not code points: a complete emoji is one unit at weight 2 however many code points it is built from. A four-person family emoji is seven code points joined by zero-width joiners. Counting code points and weighting them gives 11. X charges 2.

We shipped that bug. Our counter over-counted joined emoji for weeks, and the test we had comparing our browser copy against our server copy passed every single run — because both copies were wrong in the same way. Two implementations agreeing with each other is evidence of consistency, not of correctness.

Which is the reason this page exists. Every counter in this category claims accuracy and none of them publish anything you can check. So here is ours, in a form that can falsify it.

Get the data

curl -s https://postsiete.com/tools/x-vectors.json | jq '.vectors | length'
# 49

Served as JSON with Access-Control-Allow-Origin: *, so a browser-based test suite can fetch it directly. MIT licensed. Each vector carries a why giving the arithmetic, so a disagreement points at a specific claim rather than at a number.

const { vectors } = await (await fetch(
  'https://postsiete.com/tools/x-vectors.json'
)).json();

for (const v of vectors) {
  const got = yourWeightedLength(v.text);
  if (got !== v.expected) {
    console.log(v.name, 'expected', v.expected, 'got', got, '—', v.why);
  }
}

How to falsify these

The expected values are derived from X’s published rules and stated as arithmetic — they are not snapshots of any implementation, ours included. That matters: a suite generated from an implementation can only ever prove that the implementation agrees with itself.

If you believe a vector is wrong, the fastest disproof is to post the string to X and see whether it is accepted at the length claimed. We would rather hear it than be wrong — tell us.

The vectors

Sources

• X Developer Platform — Counting Characters: the 280 limit, the flat 23 for URLs, the weight table, and NFC normalisation.
• twitter-text configuration: emojiParsingEnabled — the weighted length “considers all emoji as a single code point … including longer grapheme clusters combined by zero-width joiners.”