Integrated some operations into main loop

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "simple_unicode_normalization_forms"
-version = "0.1.0"
+version = "0.1.1"
 edition = "2021"
 
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html

src/emoji.rs

@@ -0,0 +1,165 @@
+pub const BY_NAME: &'static [(&'static str, &'static [(char, char)])] = &[
+    ("Emoji_Modifier", EMOJI_MODIFIER),
+    ("Emoji_Modifier_Base", EMOJI_MODIFIER_BASE),
+    ("Emoji_Presentation", EMOJI_PRESENTATION),
+];
+
+pub const EMOJI_MODIFIER: &'static [(char, char)] = &[('🏻', '🏿')];
+pub const EXTRA_CHARS: &'static [(char, char)] =
+    &[('\u{FE0E}', '\u{FE0F}'), ('\u{20E2}', '\u{20E4}')];
+
+pub const EMOJI_MODIFIER_BASE: &'static [(char, char)] = &[
+    ('☝', '☝'),
+    ('⛹', '⛹'),
+    ('✊', '✍'),
+    ('🎅', '🎅'),
+    ('🏂', '🏄'),
+    ('🏇', '🏇'),
+    ('🏊', '🏌'),
+    ('👂', '👃'),
+    ('👆', '👐'),
+    ('👦', '👸'),
+    ('👼', '👼'),
+    ('💁', '💃'),
+    ('💅', '💇'),
+    ('💏', '💏'),
+    ('💑', '💑'),
+    ('💪', '💪'),
+    ('🕴', '🕵'),
+    ('🕺', '🕺'),
+    ('🖐', '🖐'),
+    ('🖕', '🖖'),
+    ('🙅', '🙇'),
+    ('🙋', '🙏'),
+    ('🚣', '🚣'),
+    ('🚴', '🚶'),
+    ('🛀', '🛀'),
+    ('🛌', '🛌'),
+    ('🤌', '🤌'),
+    ('🤏', '🤏'),
+    ('🤘', '🤟'),
+    ('🤦', '🤦'),
+    ('🤰', '🤹'),
+    ('🤼', '🤾'),
+    ('🥷', '🥷'),
+    ('🦵', '🦶'),
+    ('🦸', '🦹'),
+    ('🦻', '🦻'),
+    ('🧍', '🧏'),
+    ('🧑', '🧝'),
+    ('🫃', '🫅'),
+    ('🫰', '🫸'),
+];
+
+pub const EMOJI_PRESENTATION: &'static [(char, char)] = &[
+    ('⌚', '⌛'),
+    ('⏩', '⏬'),
+    ('⏰', '⏰'),
+    ('⏳', '⏳'),
+    ('◽', '◾'),
+    ('☔', '☕'),
+    ('♈', '♓'),
+    ('♿', '♿'),
+    ('⚓', '⚓'),
+    ('⚡', '⚡'),
+    ('⚪', '⚫'),
+    ('⚽', '⚾'),
+    ('⛄', '⛅'),
+    ('⛎', '⛎'),
+    ('⛔', '⛔'),
+    ('⛪', '⛪'),
+    ('⛲', '⛳'),
+    ('⛵', '⛵'),
+    ('⛺', '⛺'),
+    ('⛽', '⛽'),
+    ('✅', '✅'),
+    ('✊', '✋'),
+    ('✨', '✨'),
+    ('❌', '❌'),
+    ('❎', '❎'),
+    ('❓', '❕'),
+    ('❗', '❗'),
+    ('➕', '➗'),
+    ('➰', '➰'),
+    ('➿', '➿'),
+    ('⬛', '⬜'),
+    ('⭐', '⭐'),
+    ('⭕', '⭕'),
+    ('🀄', '🀄'),
+    ('🃏', '🃏'),
+    ('🆎', '🆎'),
+    ('🆑', '🆚'),
+    ('🇦', '🇿'),
+    ('🈁', '🈁'),
+    ('🈚', '🈚'),
+    ('🈯', '🈯'),
+    ('🈲', '🈶'),
+    ('🈸', '🈺'),
+    ('🉐', '🉑'),
+    ('🌀', '🌠'),
+    ('🌭', '🌵'),
+    ('🌷', '🍼'),
+    ('🍾', '🎓'),
+    ('🎠', '🏊'),
+    ('🏏', '🏓'),
+    ('🏠', '🏰'),
+    ('🏴', '🏴'),
+    ('🏸', '🐾'),
+    ('👀', '👀'),
+    ('👂', '📼'),
+    ('📿', '🔽'),
+    ('🕋', '🕎'),
+    ('🕐', '🕧'),
+    ('🕺', '🕺'),
+    ('🖕', '🖖'),
+    ('🖤', '🖤'),
+    ('🗻', '🙏'),
+    ('🚀', '🛅'),
+    ('🛌', '🛌'),
+    ('🛐', '🛒'),
+    ('🛕', '🛗'),
+    ('🛜', '🛟'),
+    ('🛫', '🛬'),
+    ('🛴', '🛼'),
+    ('🟠', '🟫'),
+    ('🟰', '🟰'),
+    ('🤌', '🤺'),
+    ('🤼', '🥅'),
+    ('🥇', '🧿'),
+    ('🩰', '🩼'),
+    ('🪀', '🪈'),
+    ('🪐', '🪽'),
+    ('🪿', '🫅'),
+    ('🫎', '🫛'),
+    ('🫠', '🫨'),
+    ('🫰', '🫸'),
+];
+
+pub trait IsEmoji {
+    fn is_emoji(&self) -> bool;
+}
+impl IsEmoji for char {
+    fn is_emoji(&self) -> bool {
+        for (lc, hc) in EMOJI_PRESENTATION {
+            if self >= lc && self <= hc {
+                return true;
+            }
+        }
+        for (lc, hc) in EMOJI_MODIFIER {
+            if self >= lc && self <= hc {
+                return true;
+            }
+        }
+        for (lc, hc) in EMOJI_MODIFIER_BASE {
+            if self >= lc && self <= hc {
+                return true;
+            }
+        }
+        for (lc, hc) in EXTRA_CHARS {
+            if self >= lc && self <= hc {
+                return true;
+            }
+        }
+        false
+    }
+}

src/lib.rs

@@ -1,47 +1,78 @@
 // Copyright (c) 2024 Future Internet Consulting and Development Solutions S.L.
+mod emoji;
 
+use emoji::IsEmoji;
 use lazy_static::lazy_static;
-use regex::Regex;
 use pyo3::prelude::*;
-use std::collections::HashSet;
+use regex::Regex;
 use unicode_normalization::char::decompose_compatible;
 use unicode_normalization::UnicodeNormalization;
 
 lazy_static! {
-    static ref WHITESPACE_RE: Regex = Regex::new(r"\s+").unwrap();
     static ref EMOJI_RE: Regex = Regex::new(r"[\p{Emoji_Presentation}\p{Emoji_Modifier}\p{Emoji_Modifier_Base}\{Cc}\uFE0E\uFE0F\u20E2\u20E3\u20E4]").unwrap();
 }
 
 /// Gives the normalized form of a string skipping some characters.
-fn nfkc_normalization(str: String, allow_chars: HashSet<char>) -> String {
+fn custom_normalization(
+    str: String,
+    allow_chars: Vec<char>,
+    collapse_whitespace: bool,
+    remove_emojis: bool,
+) -> String {
     let mut result = String::with_capacity(str.len());
+    let mut previous_whitespace = false;
     for c in str.chars() {
-        if allow_chars.contains(&c) {
-            result.push(c)
+        custom_character_normalization(
+            &mut result,
+            c,
+            &allow_chars,
+            collapse_whitespace,
+            previous_whitespace,
+            remove_emojis,
+        );
+        previous_whitespace = c.is_whitespace();
+    }
+    result.nfc().collect::<String>()
+}
+
+fn custom_character_normalization(
+    str: &mut String,
+    c: char,
+    allow_chars: &Vec<char>,
+    collapse_whitespace: bool,
+    previous_whitespace: bool,
+    remove_emojis: bool,
+) {
+    if allow_chars.contains(&c) {
+        str.push(c)
+    } else if c.is_whitespace() {
+        if collapse_whitespace && previous_whitespace {
+            return;
         } else {
-            decompose_compatible(c, |r| {
-                // Ignore characters outside the Basic Multilingual Plane and in the disallow_chars set
-                if r <= '\u{FFFF}' {
-                    result.push(r)
-                }
-            })
+            str.push(' ')
         }
+    } else if remove_emojis && c.is_emoji() {
+        return;
+    } else {
+        decompose_compatible(c, |r| {
+            // Ignore characters outside the Basic Multilingual Plane and in the disallow_chars set
+            if r <= '\u{FFFF}' {
+                str.push(r)
+            }
+        })
     }
-
-    result.nfc().collect::<String>()
 }
 
 #[pyfunction]
 fn basic_string_clean(value: String) -> PyResult<String> {
-    Ok(nfkc_normalization(value, HashSet::from(['º', 'ª'])).trim().to_string())
+    Ok(custom_normalization(value, vec!['º', 'ª'], false, false)
+        .trim()
+        .to_string())
 }
 
 #[pyfunction]
 fn remove_emojis(value: String) -> PyResult<String> {
-    let cleaned_value = nfkc_normalization(value, HashSet::from(['º', 'ª']));
-    let whitespace_cleaned_value = WHITESPACE_RE.replace_all(&cleaned_value, " ");
-    let result = EMOJI_RE.replace_all(&whitespace_cleaned_value, "");
-
+    let result = custom_normalization(value, vec!['º', 'ª'], true, true);
     Ok(result.trim().to_string())
 }
 
@@ -52,3 +83,17 @@ fn simple_unicode_normalization_forms(m: &Bound<'_, PyModule>) -> PyResult<()> {
     m.add_function(wrap_pyfunction!(remove_emojis, m)?)?;
     Ok(())
 }
+
+#[cfg(test)]
+mod tests {
+    use super::remove_emojis;
+    use std::time::Instant;
+
+    #[test]
+    fn timeit() {
+        let t1 = Instant::now();
+        remove_emojis("  a\t   name with ❤️✳️0️⃣#️⃣  #©*1   ".to_string());
+        let t2 = Instant::now();
+        println!("{:?}", t2 - t1);
+    }
+}