Feat: text (#1)

Cargo.toml

@@ -66,6 +66,7 @@ fast_poisson = { version = "0.5.2", optional = true, features = [
     "single_precision",
 ] } # Higher versions have problems with x86 due to 'kiddo'.
 workerpool = "1.2.1"
+translit = { version = "0.5.0", optional = true }
 
 [features]
 default = [
@@ -101,6 +102,7 @@ all = [
     "rustls_tls",
     "sanitize",
     "sql",
+    "text",
     "time",
     "toml",
     "url",
@@ -160,6 +162,7 @@ pathfinder = ["num-integer", "pathfinding", "serde", "serde_json"]
 poissonnoise = ["fast_poisson"]
 redis_pubsub = ["flume", "redis", "serde", "serde_json"]
 redis_reliablequeue = ["flume", "redis", "serde", "serde_json"]
+text = ["translit"]
 unzip = ["zip", "jobs"]
 worleynoise = ["rand", "rayon"]
 

dmsrc/text.dm

@@ -0,0 +1,2 @@
+#define rustg_cyrillic_to_latin(text) call(RUST_G, "cyrillic_to_latin")("[text]")
+#define rustg_latin_to_cyrillic(text) call(RUST_G, "latin_to_cyrillic")("[text]")

src/lib.rs

@@ -44,6 +44,8 @@ pub mod redis_reliablequeue;
 pub mod sanitize;
 #[cfg(feature = "sql")]
 pub mod sql;
+#[cfg(feature = "text")]
+pub mod text;
 #[cfg(feature = "time")]
 pub mod time;
 #[cfg(feature = "toml")]

src/text.rs

@@ -0,0 +1,210 @@
+use translit::{CharsMapping, Transliterator};
+
+byond_fn!(fn cyrillic_to_latin (string) {
+    Some(cyr_to_latin(string))
+});
+
+byond_fn!(fn latin_to_cyrillic (string) {
+    Some(latin_to_cyr(string))
+});
+
+fn cyr_to_latin(string: &str) -> String {
+    Transliterator::new(table_to_latin()).convert(string, false)
+}
+
+fn latin_to_cyr(string: &str) -> String {
+    Transliterator::new(table_from_latin()).convert(string, true)
+}
+
+fn table_to_latin() -> CharsMapping {
+    [
+        ("А", "A"),
+        ("Б", "B"),
+        ("В", "V"),
+        ("Г", "G"),
+        ("Д", "D"),
+        ("Е", "E"),
+        ("Ё", "Yo"),
+        ("Ж", "Zh"),
+        ("З", "Z"),
+        ("И", "I"),
+        ("Й", "I"),
+        ("К", "K"),
+        ("Л", "L"),
+        ("М", "M"),
+        ("Н", "N"),
+        ("О", "O"),
+        ("П", "P"),
+        ("Р", "R"),
+        ("С", "S"),
+        ("Т", "T"),
+        ("У", "U"),
+        ("Ф", "F"),
+        ("Х", "Kh"),
+        ("Х", "H"),
+        ("Ц", "Ts"),
+        ("Ч", "Ch"),
+        ("Ш", "Sh"),
+        ("Щ", "Shch"),
+        ("Ъ", "Ie"),
+        ("Ы", "Y"),
+        ("Ь", "'"),
+        ("Э", "E"),
+        ("Ю", "Iu"),
+        ("Я", "Ia"),
+        ("а", "a"),
+        ("б", "b"),
+        ("в", "v"),
+        ("г", "g"),
+        ("д", "d"),
+        ("е", "e"),
+        ("ё", "yo"),
+        ("ж", "zh"),
+        ("з", "z"),
+        ("и", "i"),
+        ("й", "i"),
+        ("к", "k"),
+        ("л", "l"),
+        ("м", "m"),
+        ("н", "n"),
+        ("о", "o"),
+        ("п", "p"),
+        ("р", "r"),
+        ("с", "s"),
+        ("т", "t"),
+        ("у", "u"),
+        ("ф", "f"),
+        ("х", "kh"),
+        ("ц", "ts"),
+        ("ч", "ch"),
+        ("ш", "sh"),
+        ("щ", "shch"),
+        ("ъ", "ie"),
+        ("ы", "y"),
+        ("ь", "'"),
+        ("э", "e"),
+        ("ю", "iu"),
+        ("я", "ia"),
+        ("№", "#"),
+    ]
+    .to_vec()
+}
+
+fn table_from_latin() -> CharsMapping {
+    [
+        ("А", "A"),
+        ("Б", "B"),
+        ("В", "V"),
+        ("В", "W"),
+        ("Г", "G"),
+        ("Д", "D"),
+        ("Дж", "J"),
+        ("Э", "E"),
+        ("Ё", "Yo"),
+        ("Ж", "Zh"),
+        ("З", "Z"),
+        ("З", "Th"),
+        ("Зэ", "The"),
+        ("И", "I"),
+        ("Й", "I"),
+        ("К", "C"),
+        ("К", "K"),
+        ("К", "Q"),
+        ("К", "Ck"),
+        ("Кс", "X"),
+        ("Л", "L"),
+        ("М", "M"),
+        ("Н", "N"),
+        ("О", "O"),
+        ("Оу", "Ow"),
+        ("П", "P"),
+        ("Р", "R"),
+        ("С", "S"),
+        ("Т", "T"),
+        ("У", "U"),
+        ("Ф", "F"),
+        ("Х", "Kh"),
+        ("Х", "H"),
+        ("Ц", "Ts"),
+        ("Ч", "Ch"),
+        ("Ш", "Sh"),
+        ("Щ", "Shch"),
+        ("Ъ", "Ie"),
+        ("Ы", "Y"),
+        ("Ь", "'"),
+        ("Е", "E"),
+        ("Ю", "Iu"),
+        ("Я", "Ia"),
+        ("а", "a"),
+        ("б", "b"),
+        ("в", "v"),
+        ("в", "w"),
+        ("г", "g"),
+        ("д", "d"),
+        ("дж", "j"),
+        ("е", "e"),
+        ("ё", "yo"),
+        ("ж", "zh"),
+        ("з", "z"),
+        ("з", "th"),
+        ("зэ", "the"),
+        ("и", "i"),
+        ("й", "i"),
+        ("к", "c"),
+        ("к", "k"),
+        ("к", "q"),
+        ("к", "ck"),
+        ("кс", "x"),
+        ("л", "l"),
+        ("м", "m"),
+        ("н", "n"),
+        ("о", "o"),
+        ("оу", "ow"),
+        ("п", "p"),
+        ("р", "r"),
+        ("с", "s"),
+        ("т", "t"),
+        ("у", "u"),
+        ("ф", "f"),
+        ("х", "kh"),
+        ("х", "h"),
+        ("ц", "ts"),
+        ("ч", "ch"),
+        ("ш", "sh"),
+        ("щ", "shch"),
+        ("ъ", "ie"),
+        ("ы", "y"),
+        ("ь", "'"),
+        ("э", "e"),
+        ("ю", "iu"),
+        ("я", "ia"),
+        ("№", "#"),
+    ]
+    .to_vec()
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn text_test() {
+        let result = cyr_to_latin("Съешь же ещё этих мягких французских булок, да выпей чаю!");
+        assert_eq!(
+            result,
+            "Sieesh' zhe eshchyo etikh miagkikh frantsuzskikh bulok, da vypei chaiu!"
+        );
+
+        let result2 = cyr_to_latin("Привет мир! Hello world!");
+        assert_eq!(result2, "Privet mir! Hello world!");
+
+        let result3 = latin_to_cyr("Privet mir! Hello world! Zhypyr perotin kuroden.");
+        assert_eq!(result3, "Привет мир! Хелло ворлд! Жыпыр перотин куроден.");
+
+        let result4 = latin_to_cyr("Привет мир! Hello world!");
+        assert_eq!(result4, "Привет мир! Хелло ворлд!");
+
+        let result5 = latin_to_cyr("The quick brown fox jumps over the lazy dog!");
+        assert_eq!(result5, "Зэ куик броун фокс джумпс овер зэ лазы дог!");
+    }
+}