Characters Are Not Glyphs
Unicode assigns code points to encoded characters. A glyph is the shape a font and text renderer use to draw text. One character can have different glyphs; a glyph can also represent several characters, as with a ligature. Unicode does not encode every possible glyph or install fonts. See the Unicode character and glyph model for the distinction.
A font file supplies shapes and metrics. Choosing another font can change how the same letters look. A Unicode text generator instead substitutes characters or adds combining marks; the receiving app still needs fonts to draw them.
Check the Actual Code Points
The following assignments were generated and checked from the characters, rather than inferred from how they look.
| Character | Code point | Unicode name |
|---|---|---|
| A | U+0041 | LATIN CAPITAL LETTER A |
| 𝐀 | U+1D400 | MATHEMATICAL BOLD CAPITAL A |
| 𝗔 | U+1D5D4 | MATHEMATICAL SANS-SERIF BOLD CAPITAL A |
| 𝒜 | U+1D49C | MATHEMATICAL SCRIPT CAPITAL A |
| 𝗛 | U+1D5DB | MATHEMATICAL SANS-SERIF BOLD CAPITAL H |
| 𝗲 | U+1D5F2 | MATHEMATICAL SANS-SERIF BOLD SMALL E |
| 𝗹 | U+1D5F9 | MATHEMATICAL SANS-SERIF BOLD SMALL L |
| 𝗼 | U+1D5FC | MATHEMATICAL SANS-SERIF BOLD SMALL O |
For Sans Bold, Hello becomes 𝗛𝗲𝗹𝗹𝗼. That is five code points with four distinct values, since l occurs twice. Each of these supplementary characters occupies two UTF-16 code units. A grapheme counter can still report five visible characters. These measures do not establish a social platform’s allowance. X uses weighted counting.
What Happens on Paste
Copying the output copies the character sequence. An app can reject, normalize or filter it. If it accepts the sequence, its font selection controls the glyphs. A missing glyph can become a box. “Valid Unicode,” “accepted by the field,” “drawn by the device,” and “accessible to the reader” are separate checks.
| Choice | What changes | Main limitation |
|---|---|---|
| Real font styling | Presentation of ordinary characters | Plain-text paste can discard styling |
| Unicode substitution | Stored character sequence | Limited alphabets and destination validation |
| Combining effect | Marks added after characters | Mark placement and extra code points |
Coverage and Exceptions
Mathematical alphabets cover basic Latin letters. Some styles have digits; there are no separate mathematical italic digit alphabets. Several script, Fraktur and double-struck letters occupy older Letterlike Symbols positions. A converter must handle those exceptions rather than incrementing every letter by the same offset. The Gothic generator’s Fraktur and double-struck tables show the distinct capital-letter exceptions side by side.
Accents, punctuation and non-Latin scripts do not have a matching styled equivalent in every alphabet. This generator preserves unmapped input. Existing accents remain, so mixed shapes are possible. The small text generator’s coverage comparison identifies uppercase letters and lowercase gaps that remain plain.
Accessibility and Search
A mathematical letter has a different identity from an ordinary Latin letter. Screen readers can announce names or read sequences differently depending on voice and context. Search systems need not normalize the two to the same string. Keep names, keywords and essential information in ordinary text.
Use actual fonts, semantic HTML and CSS when you control a website or document. A font choice alone does not create semantic emphasis; HTML strong and em elements convey meaning. Decorative Unicode can serve a short visual experiment in a plain-text field, but cannot guarantee accessible reading.
Try a Controlled Comparison
Enter a mixture of Latin letters, digits, accents and emoji in the generator. Compare what changes and what stays plain. Copy into a private draft and inspect the saved result. Keep the original so you can undo the substitution accurately. For a combining-mark comparison, the Zalgo generator’s intensity examples show how the same base letters acquire different numbers of marks.
