Fix breaking up of long words so it doesn't mess up ANSI escape codes.

Author: devdanzin

tabulate/__init__.py

@@ -2540,16 +2540,19 @@ def _handle_long_word(self, reversed_chunks, cur_line, cur_len, width):
             # take each charcter's width into account
             chunk = reversed_chunks[-1]
             i = 1
+            # Only count printable characters, so strip_ansi first, index later.
             while len(_strip_ansi(chunk)[:i]) <= space_left:
                 i = i + 1
             # Consider escape codes when breaking words up
             total_escape_len = 0
+            last_group = 0
             if _ansi_codes.search(chunk) is not None:
                 for group, _, _, _ in _ansi_codes.findall(chunk):
                     escape_len = len(group)
-                    # FIXME: Needs to keep track of found groups and search from there
-                    if group in chunk[: i  + total_escape_len + escape_len - 1]:
+                    if group in chunk[last_group: i + total_escape_len + escape_len - 1]:
                         total_escape_len += escape_len
+                        found = _ansi_codes.search(chunk[last_group:])
+                        last_group += found.end()
             cur_line.append(chunk[: i  + total_escape_len - 1])
             reversed_chunks[-1] = chunk[i + total_escape_len - 1 :]
 

test/test_textwrapper.py

@@ -177,23 +177,23 @@ def test_wrap_color_line_longword():
 
 
 def test_wrap_color_line_multiple_escapes():
-    data = '012345(\x1b[32ma\x1b[0mbc\x1b[32mdefghij\x1b[0m)'
+    data = "012345(\x1b[32ma\x1b[0mbc\x1b[32mdefghij\x1b[0m)"
     expected = [
-        "012345(\x1b[32ma\x1b[0mbc\x1b[32mdefg\x1b[0m",
-        "\x1b[32mhij\x1b[0m)",
+        "012345(\x1b[32ma\x1b[0mbc\x1b[32m\x1b[0m",
+        "\x1b[32mdefghij\x1b[0m)",
     ]
     wrapper = CTW(width=10)
     result = wrapper.wrap(data)
     assert_equal(expected, result)
+
     clean_data = _strip_ansi(data)
     for width in range(2, len(clean_data)):
-        # Currently fails with 14, 15 and 16, because a escape code gets split at the end
         wrapper = CTW(width=width)
         result = wrapper.wrap(data)
-        # print(width, result)
         # Comparing after stripping ANSI should be enough to catch broken escape codes
         assert_equal(clean_data, _strip_ansi("".join(result)))
 
+
 def test_wrap_datetime():
     """TextWrapper: Show that datetimes can be wrapped without crashing"""
     data = [