Kannada

texish sets the Kannada script with the syllable shaping it needs. Like the other Indic scripts it is written left to right, so it needs no bidirectional reordering; what it needs is work inside each syllable, and Kannada asks for more of it than any other script texish sets. The engine segments a run into orthographic syllables and shapes each one — splitting the composed vowel signs, moving the sign that belongs to the base, building the subscripts, and lifting out the reph — all automatically from text typed in the usual Unicode order.

Kannada is Telugu‘s closest relative here, and the two differ in exactly two places: Kannada forms a reph and Telugu does not, and Kannada writes far more of its vowel signs as combinations.

The Kannada font

A Kannada face, Noto Serif Kannada, is bundled in a regular and a bold cut. Select it with \font kannada:

{\font kannada 14 regular
ಕನ್ನಡ ಒಂದು ಸುಂದರವಾದ ಭಾಷೆ.
}

{\font kannada 18 bold ನಮಸ್ಕಾರ!}

(As with the bundled Hebrew, Arabic, CJK and other Indic faces, the in-browser Scala.js build does not ship the Kannada font, to keep the download small.)

What the engine shapes

Everything below happens automatically for any run set in a Kannada font — you type the characters in reading order and the engine does the rest.

A consonant and its vowel sign fuse into one glyph. Where Devanagari sets a letter and a sign beside it, Kannada usually draws the pair as a single form: ಕಿ is one glyph, not two, and so is ಕೆ.

Joined consonants hang beneath the base. A consonant joined to the next by the virama is drawn as the ottakshara, a subscript under the base — the second ಕ of ಕ್ಕ, the ತ of ಸ್ತ. This inverts which consonant is the base: Devanagari’s is the last consonant of a conjunct, Kannada’s the first, with everything after it underneath. Words like ಪುಸ್ತಕ (book) and ಚಿಹ್ನೆ (sign) stack this way.

The vowel sign is set beside the base, not after the subscripts. A sign is typed after the whole conjunct but belongs to the base alone, so the engine moves it back across the subscripts to reach it — ಕ್ಕಾ sets the ka, its aa sign, and only then the subjoined ka. Without that move the base and its sign never meet and the font cannot fuse them. The signs drawn below the syllable rather than on the base — the vocalic r and rr, the vocalic L and LL — are the exception and keep their place after the subscripts, as do the two length marks.

Composed vowel signs are split, and the split repeats. Five of the twelve signs are written as combinations of others: ೀ is the i sign plus a length mark, ೇ the e sign plus a length mark, ೈ the e sign plus an ai length mark, ೊ the e sign plus a uu sign — and ೋ is the ೊ sign plus a length mark, so it splits twice and reaches the font as three glyphs. These are Unicode’s own canonical decompositions. The au sign ೌ is the one composite Kannada draws with a glyph of its own, so it does not split.

A syllable-opening ra becomes the arkavattu. A ra joined by the virama at the head of a syllable is not drawn in place: it is lifted out and set after the syllable, after any vowel sign the base carries — so ರ್ಕಾ sets ka, the aa sign, then the arkavattu. ಧರ್ಮ (duty), ಕರ್ಮ (action), ಸೂರ್ಯ (sun) and ಅರ್ಥ (meaning) are the everyday cases. It lands where Devanagari’s reph lands, but it is not drawn the same way: Devanagari’s floats above its base with no width of its own, while Kannada’s is written to the upper right and takes its own room on the line.

The anusvara and visarga follow the base. They are set after the syllable, outside whatever is subjoined beneath it: ಕಂ, ಕಃ.

Setting a paragraph

Kannada sets in ordinary running text: the line breaker finds its breaks at the interword spaces exactly as for Latin, and only the shaping within each word is special. A whole Kannada paragraph needs nothing beyond selecting the font.

{\font kannada 12 regular
ಇದು ಕನ್ನಡ ಲಿಪಿಯಲ್ಲಿ ಬರೆಯಲಾಗಿದೆ. ಪ್ರತಿ ಅಕ್ಷರ ಒಂದು ವ್ಯಂಜನ ಮತ್ತು ಅದರ ಸ್ವರವನ್ನು
ಒಟ್ಟಿಗೆ ಹೊಂದಿರುತ್ತದೆ.
}

Because Kannada sets left to right like a Latin text face, it uses the same TeX-style punctuation shorthands: `` and '' become curly quotes, --/--- en/em dashes, and ... an ellipsis. (The right-to-left Hebrew and Arabic faces do not — their quote conventions differ, so those documents type the marks literally.)

The bundled scripts/kannada-demo.script sets a full page covering each of these features.