Generate our own unicode block constants

Add a script to generate a custom unicode_block module with constants
representing unicode blocks. This replaces the blocks from
unic-ucd-blocks.

Also refactor CharacterType::Range to use a RangeInclusive so we can
also remove the dependency on unic-char-range.

Cargo.toml

@@ -22,8 +22,6 @@ tree-sitter = "0.22.6"
 tree-sitter-javascript = "0.21.2"
 tree-sitter-python = "0.21.0"
 unic-ucd-name = "0.9.0"
-unic-ucd-block = "0.9.0"
-unic-char-range = "0.9.0"
 toml = "0.8.14"
 serde = { version = "1.0.203", features = ["derive"] }
 walkdir = "2.5.0"

hack/genblocks

@@ -0,0 +1,39 @@
+#!/usr/bin/env python
+import io
+import urllib.request
+import re
+
+BLOCKDEF = re.compile(
+    r"^(?P<low>[0-9A-Fa-f]+)\.\.(?P<high>[0-9A-Fa-f]+); (?P<name>.*)$"
+)
+
+
+resp = urllib.request.urlopen("https://www.unicode.org/Public/UNIDATA/Blocks.txt")
+blocks = []
+for line in io.TextIOWrapper(resp):
+    if match := BLOCKDEF.match(line.strip()):
+        name = match.group("name")
+        if name in {"Low Surrogates", "High Surrogates", "High Private Use Surrogates"}:
+            continue
+        low = match.group("low")
+        high = match.group("high")
+        blocks.append((name, low, high))
+
+
+def constname(blockname):
+    return blockname.replace(" ", "_").replace("-", "_").upper()
+
+
+print("#![cfg_attr(rustfmt, rustfmt_skip)]")
+print("// Code generated by hack/genblocks. DO NOT EDIT.")
+print()
+for name, low, high in blocks:
+    rustrange = f"'\\u{{{low}}}'..='\\u{{{high}}}'"
+    print(f"pub const {constname(name)}: std::ops::RangeInclusive<char> = {rustrange};")
+
+print(
+    """pub static UNICODE_BLOCKS: phf::Map<&'static str, std::ops::RangeInclusive<char>> = phf::phf_map! {"""
+)
+for name, _, _ in blocks:
+    print(f'    "{name}" => {constname(name)},')
+print("};")

src/config.rs

@@ -35,15 +35,13 @@ impl FromStr for CharacterType {
         if s == "*" {
             return Ok(Self::Anything);
         }
-        for block in unic_ucd_block::BlockIter::new() {
-            if block.name == s {
-                return Ok(Self::Block(block));
-            }
+        if let Some(range) = crate::unicode_blocks::UNICODE_BLOCKS.get(s) {
+            return Ok(Self::Block(range));
         }
         if let Some((low, high)) = s.split_once("..") {
             let low = unicode_notation_to_char(low)?;
             let high = unicode_notation_to_char(high)?;
-            return Ok(Self::Range(unic_char_range::CharRange { low, high }));
+            return Ok(Self::Range(low..=high));
         }
         unicode_notation_to_char(s).map(Self::CodePoint)
     }
@@ -158,9 +156,6 @@ pub struct Config {
 
 #[cfg(test)]
 mod tests {
-    use unic_char_range::CharRange;
-    use unic_ucd_block::BlockIter;
-
     use super::*;
     use crate::rules::*;
 
@@ -212,8 +207,6 @@ deny = ["Tibetan"]
         )
         .unwrap();
 
-        let tibetan_block = BlockIter::new().find(|b| b.name == "Tibetan").unwrap();
-
         let expected_config = Config {
             global: ConfigRules {
                 default: RuleSet {
@@ -235,19 +228,16 @@ deny = ["Tibetan"]
                     rules: ConfigRules {
                         default: RuleSet {
                             allow: vec![
-                                CharacterType::Block(tibetan_block),
+                                CharacterType::Block(&crate::unicode_blocks::TIBETAN),
                                 CharacterType::CodePoint('\u{9000}'),
                             ],
-                            deny: vec![CharacterType::Range(CharRange {
-                                low: '\u{5000}',
-                                high: '\u{5004}',
-                            })],
+                            deny: vec![CharacterType::Range('\u{5000}'..='\u{5004}')],
                         },
                         code_type_rules: HashMap::from([(
                             CodeType::StringLiteral,
                             RuleSet {
                                 allow: vec![],
-                                deny: vec![CharacterType::Block(tibetan_block)],
+                                deny: vec![CharacterType::Block(&crate::unicode_blocks::TIBETAN)],
                             },
                         )]),
                     },

src/main.rs

@@ -15,6 +15,7 @@ use unic_ucd_name::Name;
 
 mod config;
 mod rules;
+mod unicode_blocks;
 
 // Replaces the previous idea of "RuleChain"s.
 struct RuleDispatcher {
@@ -188,13 +189,10 @@ fn get_user_config() -> anyhow::Result<Option<Config>> {
 /// Comments and string literals allow all unicode except Bidi characters,
 /// all other kinds of code deny all unicode.
 fn get_default_config() -> Config {
-    let ascii = unic_ucd_block::BlockIter::new()
-        .find(|b| b.name == "Basic Latin")
-        .unwrap();
     Config {
         global: config::ConfigRules {
             default: RuleSet {
-                allow: vec![rules::CharacterType::Block(ascii)],
+                allow: vec![rules::CharacterType::Block(&unicode_blocks::BASIC_LATIN)],
                 deny: vec![],
             },
             code_type_rules: [

src/rules.rs

@@ -45,11 +45,11 @@ pub enum CharacterType {
     /// Single character (eg. "U+9000")
     CodePoint(char),
     /// An inclusive range of characters (eg. "U+1400..U+1409")
-    Range(unic_char_range::CharRange),
+    Range(std::ops::RangeInclusive<char>),
     /// All bidirectional control characters (right to left etc)
     Bidi,
     /// Named ranges of characters (eg. "Tibetan", "Box Drawing")
-    Block(unic_ucd_block::Block),
+    Block(&'static std::ops::RangeInclusive<char>),
     /// Any possible character.
     Anything,
 }
@@ -58,14 +58,14 @@ impl CharacterType {
     fn matches(&self, c: char) -> bool {
         match self {
             Self::CodePoint(rule_char) => *rule_char == c,
-            Self::Range(range) => range.contains(c),
+            Self::Range(range) => range.contains(&c),
             Self::Bidi => [
                 // List of bidirectional formatting characters from https://en.wikipedia.org/wiki/Trojan_Source
                 '\u{202A}', '\u{202b}', '\u{202c}', '\u{202d}', '\u{202e}', '\u{2066}', '\u{2067}',
                 '\u{2068}', '\u{2069}',
             ]
             .contains(&c),
-            Self::Block(block) => block.range.contains(c),
+            Self::Block(range) => range.contains(&c),
             Self::Anything => true,
         }
     }
@@ -88,7 +88,7 @@ impl PartialEq for CharacterType {
             (CodePoint(self_c), CodePoint(other_c)) => self_c == other_c,
             (Range(self_r), Range(other_r)) => self_r == other_r,
             (Bidi, Bidi) => true,
-            (Block(self_block), Block(other_block)) => self_block.name == other_block.name,
+            (Block(self_range), Block(other_range)) => self_range == other_range,
             (Anything, Anything) => true,
             _ => false,
         }