Inspect and convert Unicode Tamil
- Paste Unicode Tamil or mixed-script text and choose encoding, or paste a supported representation and choose decoding.
- Select the same representation used by the input or desired for the output.
- Convert and review the result, then download or copy.
தமிழ் has five scalar values: U+0BA4 U+0BAE U+0BBF U+0BB4 U+0BCD. Code points describe stored characters, rather than grapheme clusters, glyph shapes or syllables. Dependent vowel signs and pulli remain in their original logical order. Composed and decomposed sequences remain different; no normalization, transliteration, spelling correction or legacy-font mapping is applied. Unicode Tamil chart.
Supported representations
U+ tokens use uppercase U+ followed by four to six hexadecimal digits, for example U+0BA4. Tokens require ASCII whitespace separators. Brace escapes use complete sequences such as \u{BA4}, concatenated or separated by ASCII spaces, tabs or CR/LF. UTF-8 hex uses two hexadecimal digits per byte, separated by ASCII whitespace; த is E0 AE A4. Hexadecimal digits accept either case. Separator whitespace is notation formatting; represent actual whitespace explicitly, such as U+0020 or UTF-8 byte 20.
Decoding accepts Unicode scalar values U+0000 through U+10FFFF, excluding surrogate values, and preserves their sequence. Valid scalar values need not be assigned characters; the tool does not verify assignment. Brace escapes are parsed as data, never evaluated as JavaScript. Named or numeric HTML references, JSON backslash escapes, percent encoding, plain legacy bytes and mixed literal/escaped text are outside these formats.
UTF-8 decoding is strict: incomplete, overlong, surrogate or out-of-range byte sequences fail without inserting replacement characters. An initial UTF-8 BOM is preserved as U+FEFF, not removed. No legacy fallback is attempted. UTF-8 decoding reference.
Limits and privacy
Metrics describe the underlying decoded Unicode text in either direction. The Tamil-block metric counts scalars in U+0B80–U+0BFF, including any unassigned values in that range; it is not a Tamil-language detector or letter count, and excludes Tamil Supplement. All scripts and mixed text are supported. Font availability controls how output appears.
Input is capped at 100,000 UTF-16 units in either direction, including notation characters. Encoded output can exceed that input cap; use smaller portions if converting it back. Local worker, ten-second timeout, cancellation, stale-result guards and valid-Unicode export validation. No uploads, draft storage, external APIs or executable markup. Original TXT preserves the current textarea value; browsers normalize textarea CR/CRLF to LF, so pasted line endings reflect browser input. Clipboard copying is optional; use downloads or selection if blocked. Clear retains direction and format. Examples match both settings. Import, example, large paste, Clear and Convert back offer one undo step, including direction and format; typing or changing settings discards it. TXT import preserves an initial BOM and normalizes CRLF/CR to LF for the editor. Decode and download preserve encoded carriage returns; Convert back is unavailable for that result because a textarea would change it. Character inspection shows every scalar in pages of 20, with combining signs isolated for inspection rather than reshaped into words.