TextLens · No. 1 of the Korean-first shelf
Korean forms still cut people off by characters with spaces and EUC-KR bytes. Paste your text and both numbers appear at once, with how many characters are left before the limit. Blank lines, broken line breaks and Markdown marks clean up in one click, and sentences over 80 characters and clichés get marked. The draft never leaves this tab.
EUC-KR is measured with the browser's own CP949 table, not guessed from a rule of thumb.
No sentence runs past 80 characters.
One Korean prose should keep one ending: a resume in -습니다 stays in -습니다.
Nothing repeats more than three times yet.
No stock phrase from the list appears.
Double spaces 0 · trailing spaces 0 · tabs 0 · full-width spaces 0 · zero-width 0
There is no endpoint for text to be sent to. The counter at the bottom of this page reports every request the page makes, and it stays at zero for anything outside this site.
UTF-8 comes from the browser's own encoder. EUC-KR is measured by decoding the CP949 table with the browser's euc-kr decoder once, then adding each character's real width — 1 byte or 2 — instead of assuming every Hangul is 2 bytes and everything else 1.
Sentence length, ending style, repeated words and stock phrases are counted by fixed rules. Nothing here decides whether a text was written by a person or a machine, and nothing is rewritten for you.
Most job sites count characters with spaces, while public-sector and some university forms count EUC-KR bytes because their systems were built on that encoding. Both numbers sit side by side here, and the target gauge can follow either one — pick the basis the form itself states.
The page asks the browser to decode every valid CP949 two-byte pair once, keeps the characters that come back and adds each one's real width: 1 byte for ASCII, 2 bytes for Hangul, Hanja, kana and full-width marks. Characters that CP949 cannot hold — emoji, for instance — are counted as 3 bytes and reported separately, because a form that expects EUC-KR will mangle them.
It only joins lines that were broken mid-sentence, the way a PDF or a web page copy arrives. A line that ends in a full stop, a question mark or a closing quote keeps its break, and blank lines between paragraphs are preserved. Every clean-up can be undone once with the button, and the full text is never rewritten beyond those joins.
No. The draft lives in the textarea and in memory only — there is no localStorage copy, no cookie carrying it, no request that could carry it. Only the interface language and your target setting are remembered, and both are numbers and language codes, not text.
Note. A counter is a ruler, not an editor: it cannot know that a sentence is wrong, only that it is long. Check the form's own rule for what counts — characters or bytes — and keep the version you submitted, since some portals silently trim what is pasted.