For best results, select letter first, then diacritic &
vowel (top row), then cantillation (bottom two rows). Performance
Alternately an integrated input device could be built, which
might ensure optimal performance without the necessity of
The meteg checkbox is inactive if any positive (except pretonic
ole and tsinorit) or postpositive cantillation mark is selected.
Some care will have to be taken so that the search input exactly
corresponds to the revised coding standard, as
pertaining to the order of signs, most notably pretonic or
prepositive signs and meteg. A solution is needed to the
problem of holam, holam male, and vav holam haser.
* Vowels, diacritics, and cantillation marks (mostly) appear in ascending order of Unicode from left to right. Meteg (top row, fourth from right) doubles as silluk. A zero-width non-joiner precedes holam (top row, seventh from right) to ensure postpositive alignment, but performance is inconsistent. Prepositive cantillation marks (dekhi, geresh mukdam, yetiv, great telisha) normally occur on the first letter of words, as postpositive signs (segolta, pashta, zarka/tsinor, little telisha) normally occur on the last. If not, alignment problems may ensue depending on fonts. (The same problems may also occur with shin/sin dot.) If possible, left-to-right output in Numeric Character Reference (NCR) should also be implemented (cf. http://www.wandel.person.dk/hebrew.html). If possible, keycaps tooltips should be implemented from inline comments.