tetrak_translit¶

Detect, transliterate and fold text in scripts that have several romanisations.

Three functions, one each for the three jobs:

detect() says which script a string is in and, where the script has variants, which – for Armenian, the classical (pre-1922, still used in the diaspora) or reformed orthography.
transliterate() renders a script in a named romanisation scheme and reads the reversible ones back. The schemes are for display.
fold() maps any of the above – the script itself, Latin in any of its schemes – onto one canonical index form, so that a search typed one way finds a record written another. The fold is for finding; its output is a key, not a spelling.

Two scripts so far, each a package of tables the engine runs over: Armenian (tetrak_translit.hy: ALA-LC, ISO 9985, BGN/PCGN, Hübschmann-Meillet and the Western pronunciation spelling) and Georgian (tetrak_translit.ka: ALA-LC, ISO 9984 and the national system).

>>> import tetrak_translit as tt
>>> tt.detect("Երեւան").orthography
'classical'
>>> tt.transliterate("Երևան", "ala_lc")
'Erevan'
>>> tt.transliterate("თბილისი", "national")
'tbilisi'
>>> tt.fold("Yerevan", script="hy") == tt.fold("Երեւան") == tt.fold("Erewan", script="hy")
True
>>> tt.fold("Chavchavadze", script="ka") == tt.fold("ჭავჭავაძე")
True

No runtime dependencies. Built for the Armenian heritage aggregator described in Tetrak’s brief 015 and for Tetrak’s own OCR output, where a romanisation beside a transcript is both a catalogue-ready field and an accessibility one.

Functions

`schemes`([script])	The available romanisation schemes, for one script or for all.
`scripts`()	The registered scripts, in documentation order.

Modules

`cli`	Command-line interface: `tetrak-translit`.
`detect`(text)	Detect the script of text and, for a registered script, its variant.
`engine`	The transliteration engine: pure functions over a `Scheme`.
`fold`(text, *[, script, scheme])	Fold text onto its canonical index form.
`hy`	Armenian: the script, its five romanisation schemes, orthography and fold.
`ka`	Georgian: the script, three romanisation schemes and the fold.
`latin`	Shared pieces of the Latin side of a fold.
`registry`	The scripts this package handles, and how to find the one a string is in.
`script`	The two kinds of data everything else runs over: a script and a scheme.
`transliterate`(text[, scheme, script, to])	Transliterate text between a script and scheme.