Scripts¶

A script is one entry of Wiktionary's Module:scripts/data: Latin (Latn), Arabic (Arab), Arabic in its Nastaliq form (Aran), Simplified Han (Hans). There are 233 of them, each with its name, the kind of writing system it is, the way it runs, and the characters it is made of.

KotlinSwiftPython

import com.fenakhay.polyglossa.Language
import com.fenakhay.polyglossa.Script

val arabic = Script.of("Arab")
arabic.system                      // "abjad"
arabic.direction                   // RIGHT_TO_LEFT
Script.named("Arabic")             // Arabic (Arab)
Script.of("Hans").parent           // Han (Hani)
Script.of("Latn").aliases          // [Roman]
Script["Xxxx"]                     // null

import Polyglossa

let arabic = try Script.of("Arab")
arabic.system                      // "abjad"
arabic.direction                   // .rightToLeft
Script.named("Arabic")             // Arabic (Arab)
try Script.of("Hans").parent       // Han (Hani)
try Script.of("Latn").aliases      // ["Roman"]
Script["Xxxx"]                     // nil

from polyglossa import Language, Script

arabic = Script["Arab"]
arabic.system                      # 'abjad'
arabic.direction                   # <ScriptDirection.RIGHT_TO_LEFT: 'rtl'>
Script.named("Arabic").code        # 'Arab'
Script["Hans"].parent.code         # 'Hani'
Script["Latn"].aliases             # ('Roman',)
Script.get("Xxxx")                 # None

Kotlin	Swift	Python	What it is
`code`, `name`, `wikidata`, `aliases`	the same	the same	As for a language
`system`	`system`	`system`	The kind of writing system: `alphabet`, `abjad`, `abugida`, `logography`
`direction`	`direction`	`direction`	Left to right, right to left, or vertical (Mongolian)
`parent`	`parent`	`parent`	The script it is a variety of, such as Han for Simplified Han
`nameFor(language)`	`name(for:)`	`name_for(language)`	The name a language gives it
`localNames`	`localNames`	`local_names`	Those names, by language
`ranges`	`ranges`	`ranges`	Its characters, as ranges of code points
`marks`	`marks`	`marks`	The characters that tell it from a script with the same `ranges`
`contains(codePoint)`	`contains(_:)`	`code_point in script`	Whether a code point is one of its characters
`count(text)`	`count(_:)`	`count(text)`	How many of a text's characters are its own

Names a language gives a script¶

Some languages call a script by a name of their own: Punjabi writes Arabic in its Nastaliq form (Aran) and calls it Shahmukhi. name is the general name, and nameFor the one a language uses:

KotlinSwiftPython

Script.of("Aran").name                          // "Arabic"
Script.of("Aran").nameFor(Language.of("pa"))    // "Shahmukhi"
Script.of("Aran").nameFor(Language.of("ur"))    // "Arabic"

try Script.of("Aran").name                          // "Arabic"
try Script.of("Aran").name(for: Language.of("pa"))  // "Shahmukhi"
try Script.of("Aran").name(for: Language.of("ur"))  // "Arabic"

Script["Aran"].name                            # 'Arabic'
Script["Aran"].name_for(Language["pa"])        # 'Shahmukhi'
Script["Aran"].name_for(Language["ur"])        # 'Arabic'

Characters¶

ranges are the code points a script is made of, as Module:scripts/data lists them, sorted and joined. A script Unicode doesn't encode, such as Afaka, has none:

KotlinSwiftPython

Script.of("Latn").ranges.take(3)        // [65..90, 97..122, 170..170]
0x416 in Script.of("Cyrl")              // true: Ж
Script.of("Arab").count("كتاب abc")      // 4
Script.of("Afak").ranges                // []

try Script.of("Latn").ranges.prefix(3)  // [65...90, 97...122, 170...170]
try Script.of("Cyrl").contains(0x416)   // true: Ж
try Script.of("Arab").count("كتاب abc")  // 4
try Script.of("Afak").ranges            // []

Script["Latn"].ranges[:3]               # (range(65, 91), range(97, 123), range(170, 171))
0x416 in Script["Cyrl"]                 # True: Ж
Script["Arab"].count("كتاب abc")         # 4
Script["Afak"].ranges                   # ()

A character can belong to several scripts: every Latin letter is also Fraktur's, Gaelic's and Japanese's, since Japanese text mixes in Latin. Detecting scripts says how one is chosen.

marks matter for two scripts only. Traditional Han and Simplified Han have Han's characters, and their marks are the characters that tell them apart, from Unicode's Unihan: 6,861 characters Unihan simplifies, and 6,410 it gives a traditional form.