Extract hashtags from any text (Arabic supported) with frequency counts and platform limits
Text to Thread
Split a long text into a thread under the 280 limit using X’s weighting (links = 23, emoji and CJK = 2) with 1/n numbering counted inside the limit, breaking at sentences then words.
Workspace
Thread
How the result is calculated
The 280 limit is weighted as in the twitter-text library: letters in U+0000–10FF and general punctuation = 1, any other character (CJK, symbols) = 2, an emoji (whole grapheme) = 2, and every http(s):// or www. link = 23 however long. The text is split at paragraph, then sentence (. ! ? ؟ …) then word boundaries, and a word longer than the budget is split by characters. The i/n numbering length is counted in the budget by iterating until the post count settles.
Limits:
Link detection is simplified (links with a scheme or www. only, not bare domains such as example.com), which X may also count as 23. Mentions and hashtags are not specially treated. Premium account limits are not applied. The tool does not check that the meaning stays clear after splitting.
FAQ
How are links counted?
Every link counts as 23 units regardless of length.
Why does numbering shorten a post?
Because “1/5” is part of the text and counts inside 280.
