From 728b948e7f621bb4a710ad09c5bdcd1e3d0afa77 Mon Sep 17 00:00:00 2001
From: salt-die <salt-die@protonmail.com>
Date: Mon, 13 Jan 2025 16:39:05 -0600
Subject: [PATCH] Refactor char_widths generation code.

---
 setup.py | 50 ++++++++++++++++++++++++--------------------------
 1 file changed, 24 insertions(+), 26 deletions(-)

diff --git a/setup.py b/setup.py
index ada2de40..345038c4 100644
--- a/setup.py
+++ b/setup.py
@@ -1,46 +1,44 @@
 """Build cython extensions and generated files."""
 
 import sys
-from itertools import groupby
 
 from Cython.Build import cythonize
 from setuptools import setup
 from setuptools.command.build_py import build_py
 from wcwidth import wcwidth
 
-WIDTHS_FILE_HEADER = """\
-\"""
+_CHAR_WIDTHS_DOC = '''\
+"""
 Character widths as generated by wcwidth.
 
 Each tuple in `CHAR_WIDTHS` represents an interval of ords with common width, i.e.,
-`(1, 31, 0)` means ords 1 through 31 (inclusive) have width 0. Intervals with width 1
+`(0, 31, 0)` means ords 0 through 31 (inclusive) have width 0. Intervals with width 1
 are omitted so if an ord doesn't belong to any interval we can assume it has width 1.
-\"""
 """
 
+'''
+
+
+def _create_char_widths():
+    """Build ``_char_widths.py``."""
+    groups = []
+    start = end = 0
+    group_width = 1
+    for codepoint in range(sys.maxunicode + 1):
+        char_width = max(wcwidth(chr(codepoint)), 0)
+        if char_width != group_width:
+            if group_width != 1:
+                groups.append((start, end, group_width))
+            group_width = char_width
+            start = end = codepoint
+        else:
+            end = codepoint
 
-def make_widths_table():
-    """Build char widths file."""
-    widths = (
-        (codepoint, wcwidth(chr(codepoint))) for codepoint in range(sys.maxunicode + 1)
-    )
-    grouped = groupby(widths, key=lambda tup: tup[1])
     with open("src/batgrl/_char_widths.py", "w") as file:
-        file.write(WIDTHS_FILE_HEADER)
-        file.write("\n")
+        file.write(_CHAR_WIDTHS_DOC)
         file.write("CHAR_WIDTHS = (\n")
-
-        for width, group in grouped:
-            if width == 1:
-                continue
-            if width == -1:
-                width = 0
-            first_codepoint, _ = next(group)
-            last_codepoint = first_codepoint
-            for last_codepoint, _ in group:
-                pass
-
-            file.write(f"    ({first_codepoint}, {last_codepoint}, {width}),\n")
+        for group in groups:
+            file.write("    ({}, {}, {}),\n".format(*group))
         file.write(")\n")
 
 
@@ -50,7 +48,7 @@ class build_py_with_char_widths(build_py):
     def run(self):
         """Generate width file on build."""
         super().run()
-        make_widths_table()
+        _create_char_widths()
 
 
 setup(