Automate generation of the char_range.inc file

Co-authored-by: Danil Alexeev <dalexeev12@yandex.ru>

core/string/char_range.inc

@@ -28,6 +28,8 @@
 /* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.                 */
 /**************************************************************************/
 
+// This file was generated using the `misc/scripts/char_range_fetch.py` script.
+
 #ifndef CHAR_RANGE_INC
 #define CHAR_RANGE_INC
 
@@ -43,7 +45,7 @@ struct CharRange {
 
 constexpr inline CharRange xid_start[] = {
 	{ 0x41, 0x5a },
-	{ 0x5f, 0x5f }, // Underscore technically isn't in XID_Start, but for our purposes it's included.
+	{ 0x5f, 0x5f },
 	{ 0x61, 0x7a },
 	{ 0xaa, 0xaa },
 	{ 0xb5, 0xb5 },

misc/scripts/char_range_fetch.py

@@ -0,0 +1,138 @@
+#!/usr/bin/env python3
+
+# Script used to dump char ranges for specific properties from
+# the Unicode Character Database to the `char_range.inc` file.
+# NOTE: This script is deliberately not integrated into the build system;
+# you should run it manually whenever you want to update the data.
+
+import os
+import sys
+from typing import Final, List, Tuple
+from urllib.request import urlopen
+
+if __name__ == "__main__":
+    sys.path.insert(1, os.path.join(os.path.dirname(__file__), "../../"))
+
+from methods import generate_copyright_header
+
+URL: Final[str] = "https://www.unicode.org/Public/16.0.0/ucd/DerivedCoreProperties.txt"
+
+
+xid_start: List[Tuple[int, int]] = []
+xid_continue: List[Tuple[int, int]] = []
+uppercase_letter: List[Tuple[int, int]] = []
+lowercase_letter: List[Tuple[int, int]] = []
+unicode_letter: List[Tuple[int, int]] = []
+
+
+def merge_ranges(ranges: List[Tuple[int, int]]) -> None:
+    if len(ranges) < 2:
+        return
+
+    last_start: int = ranges[0][0]
+    last_end: int = ranges[0][1]
+    original_ranges: List[Tuple[int, int]] = ranges[1:]
+
+    ranges.clear()
+
+    for curr_range in original_ranges:
+        curr_start: int = curr_range[0]
+        curr_end: int = curr_range[1]
+        if last_end + 1 != curr_start:
+            ranges.append((last_start, last_end))
+            last_start = curr_start
+        last_end = curr_end
+
+    ranges.append((last_start, last_end))
+
+
+def parse_unicode_data() -> None:
+    lines: List[str] = [line.decode("utf-8") for line in urlopen(URL)]
+
+    for line in lines:
+        if line.startswith("#") or not line.strip():
+            continue
+
+        split_line: List[str] = line.split(";")
+
+        char_range: str = split_line[0].strip()
+        char_property: str = split_line[1].strip().split("#")[0].strip()
+
+        range_start: str = char_range
+        range_end: str = char_range
+        if ".." in char_range:
+            range_start, range_end = char_range.split("..")
+
+        range_tuple: Tuple[int, int] = (int(range_start, 16), int(range_end, 16))
+
+        if char_property == "XID_Start":
+            xid_start.append(range_tuple)
+        elif char_property == "XID_Continue":
+            xid_continue.append(range_tuple)
+        elif char_property == "Uppercase":
+            uppercase_letter.append(range_tuple)
+        elif char_property == "Lowercase":
+            lowercase_letter.append(range_tuple)
+        elif char_property == "Alphabetic":
+            unicode_letter.append(range_tuple)
+
+    # Underscore technically isn't in XID_Start, but for our purposes it's included.
+    xid_start.append((0x005F, 0x005F))
+    xid_start.sort(key=lambda x: x[0])
+
+    merge_ranges(xid_start)
+    merge_ranges(xid_continue)
+    merge_ranges(uppercase_letter)
+    merge_ranges(lowercase_letter)
+    merge_ranges(unicode_letter)
+
+
+def make_array(array_name: str, range_list: List[Tuple[int, int]]) -> str:
+    result: str = f"constexpr inline CharRange {array_name}[] = {{\n"
+
+    for start, end in range_list:
+        result += f"\t{{ 0x{start:x}, 0x{end:x} }},\n"
+
+    result += "};\n\n"
+
+    return result
+
+
+def generate_char_range_inc() -> None:
+    parse_unicode_data()
+
+    source: str = generate_copyright_header("char_range.inc")
+
+    source += f"""
+// This file was generated using the `misc/scripts/char_range_fetch.py` script.
+
+#ifndef CHAR_RANGE_INC
+#define CHAR_RANGE_INC
+
+#include "core/typedefs.h"
+
+// Unicode Derived Core Properties
+// Source: {URL}
+
+struct CharRange {{
+\tchar32_t start;
+\tchar32_t end;
+}};\n\n"""
+
+    source += make_array("xid_start", xid_start)
+    source += make_array("xid_continue", xid_continue)
+    source += make_array("uppercase_letter", uppercase_letter)
+    source += make_array("lowercase_letter", lowercase_letter)
+    source += make_array("unicode_letter", unicode_letter)
+
+    source += "#endif // CHAR_RANGE_INC\n"
+
+    char_range_path: str = os.path.join(os.path.dirname(__file__), "../../core/string/char_range.inc")
+    with open(char_range_path, "w", newline="\n") as f:
+        f.write(source)
+
+    print("`char_range.inc` generated successfully.")
+
+
+if __name__ == "__main__":
+    generate_char_range_inc()