String Length Calculator

How to Use the String Length Calculator

1. Paste or type your string into the input textarea. Single tokens, lines of code, or long paragraphs all work.
2. Watch the metrics update in the cards above the output: .length, UTF-8 bytes, UTF-16 bytes, graphemes, code points, and lines. Every metric refreshes on each keystroke.
3. Look for the divergence warning. When .length differs from the grapheme count, a yellow note appears - your string contains characters that will break naive length checks.
4. Use the UTF-8 byte count when sizing database columns, HTTP headers, or file sizes. Use the UTF-16 byte count when thinking about in-memory strings in Java, C#, or JavaScript.
5. Copy any specific metric by selecting the number directly from the card; the input textarea keeps your full string available for further tweaking.

What Each Metric Means Under the Hood

.length reads the JavaScript string property, which returns the number of UTF-16 code units - 16-bit chunks as the string is stored internally. UTF-8 bytes is computed by new TextEncoder().encode(str).length, which runs the WHATWG UTF-8 encoder and returns a Uint8Array whose byte length is what a UTF-8 file would have. UTF-16 bytes is simply str.length * 2, because every UTF-16 code unit is 2 bytes. Code points use the spread operator [...str].length, which iterates by Unicode scalar value and counts surrogate pairs as one. Grapheme count relies on Intl.Segmenter with granularity: "grapheme", the only correct way to count user-perceived characters in modern JavaScript. Line count splits on /\\r?\\n/. Every metric is computed locally with no network call.

When These Metrics Matter

• Validating a user nickname against a 20-character limit where users assume "character" means grapheme, not UTF-16 code unit.
• Sizing a database VARCHAR(255) column - PostgreSQL counts Unicode code points, MySQL with utf8mb4 counts bytes, and SQL Server NVARCHAR counts UTF-16 code units.
• Calculating exact SMS payload fees, where carriers charge by GSM septet count for ASCII and UCS-2 character count for anything else.
• Checking whether a tweet fits under 280 "characters" as Twitter defines them, which blends code points with weighted Latin vs CJK scoring.
• Confirming that your API response body stays under a Content-Length limit expressed in bytes.
• Spotting a bug where your form validator reports "too long" for a perfectly normal-looking name because the backend counts bytes and the UI counts characters.

Common Pitfalls and Edge Cases

• Single emoji spanning 4 bytes. The grinning face 😀 is one grapheme, one code point (U+1F600), two UTF-16 code units, and four UTF-8 bytes. Any validator that disagrees about which of those is "length" will surprise someone.
• Family emoji with ZWJ. A family-of-four emoji is seven code points joined by zero-width joiners - one grapheme but up to 25 UTF-8 bytes. The divergence here is dramatic and routinely breaks mobile input validators.
• Hangul syllable blocks. Korean syllables can be precomposed (one code point) or decomposed into Jamo (multiple code points). The two forms render identically but have different lengths in every metric except grapheme count.
• Combining accents. é as U+00E9 is 1 code point and 2 UTF-8 bytes; as e + combining acute (U+0065 U+0301) it is 2 code points and 3 UTF-8 bytes. Always normalize with str.normalize("NFC") before length comparison.
• BOM and invisible characters. A leading U+FEFF byte-order mark adds 3 UTF-8 bytes that the user cannot see. Run suspicious inputs through the Invisible Character Detector on this site.
• Normalization forms. NFC, NFD, NFKC, and NFKD produce different byte lengths for the same visual string; choose one consistently.

Why There Are Six Different Answers

The Unicode standard defines abstract characters and assigns them code points from U+0000 to U+10FFFF. UTF-8 (RFC 3629) encodes each code point as 1 to 4 bytes; UTF-16 (RFC 2781) as 1 or 2 16-bit code units; UTF-32 as a single 32-bit unit. JavaScript uses UTF-16 internally, so string.length counts code units, and a supplementary-plane emoji counts as 2. Grapheme clusters, defined by Unicode Standard Annex #29, group multiple code points into what the user perceives as one character - what you see on screen when you put the caret to the left and press right arrow once. The difference matters because form validation, database limits, SMS billing, and API length caps are all defined in different units by different systems.

Comparison to Alternatives

In Node.js the same metrics are reachable with str.length, Buffer.byteLength(str, "utf8"), and [...str].length; for graphemes you still need Intl.Segmenter or the grapheme-splitter npm package. Python's len() on a str counts code points, str.encode("utf-8") plus len gives bytes, and the grapheme package gives grapheme counts. Ruby's string.length counts grapheme clusters (since 2.6 with each_grapheme_cluster), which is why Ruby web apps often "just work" where Node apps surprise users. VS Code and JetBrains IDEs show code-unit count in the status bar; your editor's char count almost always means code units. Use this tool when you need all six numbers at once to debug a length-mismatch bug across a full stack.