From ecc4b3d81b75a5fe3a4911eb1cc68c43fdae1e2d Mon Sep 17 00:00:00 2001
From: Alfan Nur Fauzan <alpancs@gmail.com>
Date: Thu, 21 Mar 2024 06:39:38 +0700
Subject: [PATCH] transliteration improvement for v0.11.3 (#38)

* add more test cases

* adjust the transliteration

* handle "kitabi la roiba"

* compress contextual_map patterns

* remove some test cases

* add test cases

* bump version
---
 Cargo.toml                       |  2 +-
 src/quranize.rs                  | 42 +++++++++++++++++++++++++-------
 src/quranize/transliterations.rs | 17 ++++---------
 3 files changed, 39 insertions(+), 22 deletions(-)

diff --git a/Cargo.toml b/Cargo.toml
index ddd941b..f1a50f5 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -1,6 +1,6 @@
 [package]
 name = "quranize"
-version = "0.11.2"
+version = "0.11.3"
 authors = ["Alfan Nur Fauzan <alpancs@gmail.com>"]
 edition = "2021"
 description = "Encoding transliterations into Quran forms."
diff --git a/src/quranize.rs b/src/quranize.rs
index e9e8aeb..83e1420 100644
--- a/src/quranize.rs
+++ b/src/quranize.rs
@@ -239,6 +239,8 @@ mod tests {
         assert_eq!(q.e("yukhodiun"), vec!["يُخٰدِعونَ"]);
         assert_eq!(q.e("indallah"), vec!["عِندَ اللَّهِ", "عِندِ اللَّهِ"]);
         assert_eq!(q.e("alimul ghoibi"), vec!["عٰلِمُ الغَيبِ"]);
+        assert_eq!(q.e("kaana dhoifa"), vec!["كانَ ضَعيفًا"]);
+        assert_eq!(q.e("kitabi la roiba"), vec!["الكِتٰبِ لا رَيبَ"]);
     }
 
     #[test]
@@ -290,10 +292,7 @@ mod tests {
     #[test]
     fn test_quranize_misc() {
         let q = Quranize::new(70);
-        assert_eq!(q.encode("bismillah")[0].1.len(), 13);
         assert_eq!(q.encode("bismillah")[0].2, 3);
-        assert_eq!(q.encode("arrohman").len(), 3);
-        assert_eq!(q.encode("arrohman")[0].1.len(), 10);
         assert_eq!(
             q.encode("alhamdu")[0].1,
             vec!["a", "l", "h", "a", "m", "d", "u"]
@@ -302,12 +301,37 @@ mod tests {
             q.encode("arrohman")[0].1,
             vec!["a", "", "r", "r", "o", "h", "m", "a", "n", ""]
         );
-        let result = &q.encode("masyaallah")[0];
-        assert_eq!(result.0.chars().count(), result.1.len());
-        assert_eq!(
-            result.1,
-            vec!["m", "a", "", "sy", "a", "a", "", "", "", "", "l", "l", "a", "h", ""]
-        );
+        {
+            let r = &q.encode("masyaallah")[0];
+            assert_eq!(r.0.chars().count(), r.1.len());
+            assert_eq!(
+                r.1,
+                vec!["m", "a", "", "sy", "a", "a", "", "", "", "", "l", "l", "a", "h", ""]
+            );
+        }
+        {
+            let r = &q.encode("birobbinnas")[0];
+            assert_eq!(
+                r.1.iter().zip(r.0.chars()).collect::<Vec<_>>(),
+                vec![
+                    (&"b", 'ب',),
+                    (&"i", '\u{650}',),
+                    (&"r", 'ر',),
+                    (&"o", '\u{64e}',),
+                    (&"b", 'ب',),
+                    (&"b", '\u{651}',),
+                    (&"i", '\u{650}',),
+                    (&"", ' ',),
+                    (&"", 'ا',),
+                    (&"", 'ل',),
+                    (&"n", 'ن',),
+                    (&"n", '\u{651}',),
+                    (&"a", 'ا',),
+                    (&"s", 'س',),
+                    (&"", '\u{650}',),
+                ]
+            );
+        }
     }
 
     #[test]
diff --git a/src/quranize/transliterations.rs b/src/quranize/transliterations.rs
index 6717d9d..8de34ef 100644
--- a/src/quranize/transliterations.rs
+++ b/src/quranize/transliterations.rs
@@ -64,18 +64,11 @@ pub(super) fn map(c: char) -> &'static [&'static str] {
 
 pub(super) fn contextual_map(c0: char, c1: char) -> &'static [&'static str] {
     match (c0, c1) {
-        (SPACE, LETTER_ALEF)
-        | (LETTER_HAMZA, LETTER_ALEF)
-        | (LETTER_ALEF, LETTER_LAM)
-        | (LETTER_AIN, LETTER_WAW)
-        | (LETTER_AIN, LETTER_SUPERSCRIPT_ALEF)
-        | (LETTER_WAW, LETTER_ALEF)
-        | (FATHATAN, LETTER_ALEF)
-        | (DAMMA, LETTER_WAW)
-        | (KASRA, LETTER_ALEF)
-        | (KASRA, LETTER_LAM)
-        | (HAMZA_ABOVE, LETTER_ALEF) => &[""],
-        ('\0', LETTER_ALEF) => &["u", "i"],
+        (SPACE | LETTER_HAMZA | LETTER_WAW | FATHATAN | KASRA | HAMZA_ABOVE, LETTER_ALEF)
+        | (LETTER_ALEF | KASRA, LETTER_LAM)
+        | (LETTER_AIN, LETTER_WAW | LETTER_YEH | LETTER_SUPERSCRIPT_ALEF)
+        | (DAMMA, LETTER_WAW) => &[""],
+        ('\0', LETTER_ALEF) => &["u", "i", ""],
         (_, SHADDA) => map(c0),
         _ => &[],
     }