Take Mac OS text encoding constants from headers

Among other uses, these constants are used on HFS+ volumes to choose
which encoding to use when converting Unicode filenames back to an 8-bit
string for old Mac OS APIs.

This is documented in TN 1150 under "Text Encodings".

https://developer.apple.com/library/archive/technotes/tn/tn1150.html

scripts/encoding.csv

@@ -0,0 +1,41 @@
+Name,Value
+kTextEncodingMacRoman,0
+kTextEncodingMacJapanese,1
+kTextEncodingMacChineseTrad,2
+kTextEncodingMacKorean,3
+kTextEncodingMacArabic,4
+kTextEncodingMacHebrew,5
+kTextEncodingMacGreek,6
+kTextEncodingMacCyrillic,7
+kTextEncodingMacDevanagari,9
+kTextEncodingMacGurmukhi,10
+kTextEncodingMacGujarati,11
+kTextEncodingMacOriya,12
+kTextEncodingMacBengali,13
+kTextEncodingMacTamil,14
+kTextEncodingMacTelugu,15
+kTextEncodingMacKannada,16
+kTextEncodingMacMalayalam,17
+kTextEncodingMacSinhalese,18
+kTextEncodingMacBurmese,19
+kTextEncodingMacKhmer,20
+kTextEncodingMacThai,21
+kTextEncodingMacLaotian,22
+kTextEncodingMacGeorgian,23
+kTextEncodingMacArmenian,24
+kTextEncodingMacChineseSimp,25
+kTextEncodingMacTibetan,26
+kTextEncodingMacMongolian,27
+kTextEncodingMacEthiopic,28
+kTextEncodingMacCentralEurRoman,29
+kTextEncodingMacVietnamese,30
+kTextEncodingMacExtArabic,31
+kTextEncodingMacSymbol,33
+kTextEncodingMacDingbats,34
+kTextEncodingMacTurkish,35
+kTextEncodingMacCroatian,36
+kTextEncodingMacIcelandic,37
+kTextEncodingMacRomanian,38
+kTextEncodingMacCeltic,39
+kTextEncodingMacGaelic,40
+kTextEncodingMacKeyboardGlyphs,41

scripts/extract.py

@@ -1,12 +1,21 @@
 """Extract script and region constants from Script.h."""
 import csv
+import os
 import re
 import sys
 
-from typing import List, Tuple
+from typing import Iterator, List, Tuple
 
 Item = Tuple[str, int]
 
+def list_enums(filename: str) -> Iterator[Item]:
+    """List enum definitions in a file."""
+    with open(filename, 'rb') as fp:
+        data = fp.read()
+    for item in re.finditer(rb'^\s*(\w+)\s*=\s*(\d+)', data, re.MULTILINE):
+        name, value = item.groups()
+        yield name.decode('ASCII'), int(value)
+
 def index_of(data: List[Item], key: str) -> int:
     for i, (name, _) in enumerate(data):
         if name == key:
@@ -17,28 +26,51 @@ def slice(data: List[Item], first: str, last: str) -> List[Item]:
     return data[index_of(data, first):index_of(data, last)+1]
 
 def write_csv(fname: str, data: List[Item]) -> None:
+    print('Writing', fname, file=sys.stderr)
     with open(fname, 'w') as fp:
         w = csv.writer(fp)
         w.writerow(['Name', 'Value'])
         for item in data:
             w.writerow(item)
 
-def main(argv: List[str]) -> None:
+def process_script(filename: str) -> None:
-    if len(argv) != 1:
-        print('usage: script_gen.py <Script.h>', file=sys.stderr)
-        raise SystemExit(2)
-    with open(argv[0], 'rb') as fp:
-        data = fp.read()
     scripts: List[Item] = []
     regions: List[Item] = []
-    for item in re.finditer(rb'^\s*(\w+)\s*=\s*(\d+)', data, re.MULTILINE):
+    for name, value in list_enums(filename):
-        name, value = item.groups()
+        if name.startswith('sm'):
-        if name.startswith(b'sm'):
+            scripts.append((name, value))
-            scripts.append((name.decode('ASCII'), int(value)))
+        elif name.startswith('ver'):
-        elif name.startswith(b'ver'):
+            regions.append((name, value))
-            regions.append((name.decode('ASCII'), int(value)))
     write_csv('script.csv', slice(scripts, 'smRoman', 'smUninterp'))
     write_csv('region.csv', slice(regions, 'verUS', 'verGreenland'))
 
+def process_textcommon(filename: str) -> None:
+    encodings: List[Item] = []
+    for name, value in list_enums(filename):
+        if name.startswith('kTextEncoding'):
+            encodings.append((name, value))
+    write_csv('encoding.csv',
+              slice(encodings, 'kTextEncodingMacRoman',
+                    'kTextEncodingMacKeyboardGlyphs'))
+
+def process(filename: str) -> None:
+    name = os.path.basename(filename).lower()
+    if name == 'script.h':
+        process_script(filename)
+    elif name == 'textcommon.h':
+        process_textcommon(filename)
+    else:
+        print('Error: unknown header file:', repr(filename), file=sys.stderr)
+        raise SystemExit(1)
+
+def main(argv: List[str]) -> None:
+    if not argv:
+        sys.stderr.write(
+            'Usage: script_gen.py [<file.h>...]\n'
+            'This will read Script.h and TextCommon.h\n')
+        raise SystemExit(2)
+    for arg in argv:
+        process(arg)
+
 if __name__ == '__main__':
     main(sys.argv[1:])