Add hyphenation_character method to Lang (#31)

Author: santhoshtr

README.md

@@ -18,6 +18,8 @@ hypher = "0.1"
   disable the `alloc` feature, but then overly long words lead to a panic.
 - Support for many languages.
 - No unsafe code, no dependencies, no std.
+- Hyphenation character awareness: `Lang::hyphenation_character()` returns
+  `None` for Indic scripts where visual hyphenation is not conventional.
 
 ## Example
 ```rust

src/lang.rs

@@ -368,6 +368,40 @@ impl Lang {
         }
     }
 
+    /// The default character used to join syllables.
+    ///
+    /// Returns `Some('\u{ad}')` (SOFT HYPHEN) for most languages, but `None`
+    /// for Indic scripts where visual hyphenation is not conventional.
+    pub fn hyphenation_character(self) -> Option<char> {
+        match self {
+            #[cfg(feature = "assamese")]
+            Self::Assamese => None,
+            #[cfg(feature = "bengali")]
+            Self::Bengali => None,
+            #[cfg(feature = "gujarati")]
+            Self::Gujarati => None,
+            #[cfg(feature = "hindi")]
+            Self::Hindi => None,
+            #[cfg(feature = "kannada")]
+            Self::Kannada => None,
+            #[cfg(feature = "malayalam")]
+            Self::Malayalam => None,
+            #[cfg(feature = "marathi")]
+            Self::Marathi => None,
+            #[cfg(feature = "oriya")]
+            Self::Oriya => None,
+            #[cfg(feature = "panjabi")]
+            Self::Panjabi => None,
+            #[cfg(feature = "sanskrit")]
+            Self::Sanskrit => None,
+            #[cfg(feature = "tamil")]
+            Self::Tamil => None,
+            #[cfg(feature = "telugu")]
+            Self::Telugu => None,
+            _ => Some('\u{ad}'),
+        }
+    }
+
     fn root(self) -> State<'static> {
         match self {
             #[cfg(feature = "afrikaans")]

tests/generate.rs

@@ -160,6 +160,30 @@ fn write_lang(
     writeln!(w, "    }}")?;
     writeln!(w)?;
 
+    // Implementation of `hyphenation_character`.
+    writeln!(w, "    /// The default character used to join syllables.")?;
+    writeln!(w, "    ///")?;
+    writeln!(w, "    /// Returns `Some('\\u{{ad}}')` (SOFT HYPHEN) for most languages, but `None`")?;
+    writeln!(
+        w,
+        "    /// for Indic scripts where visual hyphenation is not conventional."
+    )?;
+    writeln!(w, "    pub fn hyphenation_character(self) -> Option<char> {{")?;
+    writeln!(w, "        match self {{")?;
+    for &(name, _, _, script, ..) in languages {
+        if !is_indic_script(script) {
+            continue;
+        }
+        let feature = name.to_lowercase();
+        write!(w, "            ")?;
+        write_cfg(w, &feature)?;
+        writeln!(w, "            Self::{name} => None,")?;
+    }
+    writeln!(w, "            _ => Some('\\u{{ad}}'),")?;
+    writeln!(w, "        }}")?;
+    writeln!(w, "    }}")?;
+    writeln!(w)?;
+
     // Implementation of `root`.
     writeln!(w, "    fn root(self) -> State<'static> {{")?;
     writeln!(w, "        match self {{")?;
@@ -175,6 +199,14 @@ fn write_lang(
     writeln!(w, "}}")
 }
 
+/// Returns true for Indic scripts where visual hyphenation is not conventional.
+fn is_indic_script(script: &str) -> bool {
+    matches!(
+        script,
+        "Beng" | "Deva" | "Gujr" | "Guru" | "Knda" | "Mlym" | "Orya" | "Taml" | "Telu"
+    )
+}
+
 fn write_cfg(w: &mut String, feature: &str) -> fmt::Result {
     writeln!(w, r#"#[cfg(feature = "{feature}")]"#)
 }