replace Python standard textwrap by MBCS sensitive one for i18n text

Mercurial has problem around text wrapping/filling in MBCS encoding
environment, because standard 'textwrap' module of Python can not
treat it correctly. It splits byte sequence for one character into two
lines.

According to unicode specification, "east asian width" classifies
characters into:

   W(ide), N(arrow), F(ull-width), H(alf-width), A(mbiguous)


W/N/F/H can be always recognized as 2/1/2/1 bytes in byte sequence,
but 'A' can not. Size of 'A' depends on language in which it is used.

Unicode specification says:

   If the context(= language) cannot be established reliably they
   should be treated as narrow characters by default

but many of class 'A' characters are full-width, at least, in Japanese
environment.

So, this patch treats class 'A' characters as full-width always for
safety wrapping.

This patch focuses only on MBCS safe-ness, not on writing/printing
rule strict wrapping for each languages

MBCS sensitive textwrap class is originally implemented
by ITO Nobuaki <[e-mail]>.

mercurial/commands.py

@@ -1697,7 +1697,9 b' def help_(ui, name=None, with_version=Fa'
                 commands = cmds[f].replace("|",", ")
                 ui.write(" %s:\n      %s\n"%(commands, h[f]))
             else:
-                ui.write(' %-*s   %s\n' % (m, f, util.wrap(h[f], m + 4)))
+                ui.write('%s\n' % (util.wrap(h[f],
+                                             initindent=' %-*s   ' % (m, f),
+                                             hangindent=' ' * (m + 4))))
 
         if not ui.quiet:
             addglobalopts(True)
@@ -1824,8 +1826,11 b' def help_(ui, name=None, with_version=Fa'
         opts_len = max([len(line[0]) for line in opt_output if line[1]] or [0])
         for first, second in opt_output:
             if second:
-                second = util.wrap(second, opts_len + 3)
-                ui.write(" %-*s  %s\n" % (opts_len, first, second))
+                initindent = ' %-*s  ' % (opts_len, first)
+                hangindent = ' ' * (opts_len + 3)
+                ui.write('%s\n' % (util.wrap(second,
+                                             initindent=initindent,
+                                             hangindent=hangindent)))
             else:
                 ui.write("%s\n" % first)
 

mercurial/encoding.py

@@ -72,6 +72,6 b' def colwidth(s):'
     d = s.decode(encoding, 'replace')
     if hasattr(unicodedata, 'east_asian_width'):
         w = unicodedata.east_asian_width
-        return sum([w(c) in 'WF' and 2 or 1 for c in d])
+        return sum([w(c) in 'WFA' and 2 or 1 for c in d])
     return len(d)
 

mercurial/minirst.py

@@ -35,8 +35,8 b' It only supports a small subset of reStr'
 - inline literals (no other inline markup is not recognized)
 """
 
-import re, sys, textwrap
-
+import re, sys
+import util
 
 def findblocks(text):
     """Find continuous blocks of lines in text.
@@ -304,9 +304,9 b' def formatblock(block, width):'
         hang = len(block['lines'][-1]) - len(block['lines'][-1].lstrip())
         defindent = indent + hang * ' '
         text = ' '.join(map(str.strip, block['lines'][1:]))
-        return "%s\n%s" % (term, textwrap.fill(text, width=width,
-                                               initial_indent=defindent,
-                                               subsequent_indent=defindent))
+        return '%s\n%s' % (term, util.wrap(text, width=width,
+                                           initindent=defindent,
+                                           hangindent=defindent))
     subindent = indent
     if block['type'] == 'bullet':
         if block['lines'][0].startswith('| '):
@@ -338,9 +338,9 b' def formatblock(block, width):'
         subindent = indent + (len(option) + len(arg)) * ' '
 
     text = ' '.join(map(str.strip, block['lines']))
-    return textwrap.fill(text, width=width,
-                         initial_indent=indent,
-                         subsequent_indent=subindent)
+    return util.wrap(text, width=width,
+                     initindent=indent,
+                     hangindent=subindent)
 
 
 def format(text, width, indent=0, keep=None):

mercurial/templatefilters.py

@@ -5,7 +5,7 b''
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
-import cgi, re, os, time, urllib, textwrap
+import cgi, re, os, time, urllib
 import util, encoding
 
 def stringify(thing):
@@ -61,15 +61,17 b' def fill(text, width):'
         while True:
             m = para_re.search(text, start)
             if not m:
-                w = len(text)
-                while w > start and text[w - 1].isspace():
+                uctext = unicode(text[start:], encoding.encoding)
+                w = len(uctext)
+                while 0 < w and uctext[w - 1].isspace():
                     w -= 1
-                yield text[start:w], text[w:]
+                yield (uctext[:w].encode(encoding.encoding),
+                       uctext[w:].encode(encoding.encoding))
                 break
             yield text[start:m.start(0)], m.group(1)
             start = m.end(1)
 
-    return "".join([space_re.sub(' ', textwrap.fill(para, width)) + rest
+    return "".join([space_re.sub(' ', util.wrap(para, width=width)) + rest
                     for para, rest in findparas()])
 
 def firstline(text):

mercurial/util.py

@@ -16,7 +16,7 b' hide platform-specific details from the '
 from i18n import _
 import error, osutil, encoding
 import cStringIO, errno, re, shutil, sys, tempfile, traceback
-import os, stat, time, calendar, textwrap, signal
+import os, stat, time, calendar, textwrap, unicodedata, signal
 import imp
 
 # Python compatibility
@@ -1257,21 +1257,49 b' def uirepr(s):'
     # Avoid double backslash in Windows path repr()
     return repr(s).replace('\\\\', '\\')
 
-def wrap(line, hangindent, width=None):
+#### naming convention of below implementation follows 'textwrap' module
+
+class MBTextWrapper(textwrap.TextWrapper):
+    def __init__(self, **kwargs):
+        textwrap.TextWrapper.__init__(self, **kwargs)
+
+    def _cutdown(self, str, space_left):
+        l = 0
+        ucstr = unicode(str, encoding.encoding)
+        w = unicodedata.east_asian_width
+        for i in xrange(len(ucstr)):
+            l += w(ucstr[i]) in 'WFA' and 2 or 1
+            if space_left < l:
+                return (ucstr[:i].encode(encoding.encoding),
+                        ucstr[i:].encode(encoding.encoding))
+        return str, ''
+
+    # ----------------------------------------
+    # overriding of base class
+
+    def _handle_long_word(self, reversed_chunks, cur_line, cur_len, width):
+        space_left = max(width - cur_len, 1)
+
+        if self.break_long_words:
+            cut, res = self._cutdown(reversed_chunks[-1], space_left)
+            cur_line.append(cut)
+            reversed_chunks[-1] = res
+        elif not cur_line:
+            cur_line.append(reversed_chunks.pop())
+
+#### naming convention of above implementation follows 'textwrap' module
+
+def wrap(line, width=None, initindent='', hangindent=''):
     if width is None:
         width = termwidth() - 2
-    if width <= hangindent:
+    maxindent = max(len(hangindent), len(initindent))
+    if width <= maxindent:
         # adjust for weird terminal size
-        width = max(78, hangindent + 1)
-    padding = '\n' + ' ' * hangindent
-    # To avoid corrupting multi-byte characters in line, we must wrap
-    # a Unicode string instead of a bytestring.
-    try:
-        u = line.decode(encoding.encoding)
-        w = padding.join(textwrap.wrap(u, width=width - hangindent))
-        return w.encode(encoding.encoding)
-    except UnicodeDecodeError:
-        return padding.join(textwrap.wrap(line, width=width - hangindent))
+        width = max(78, maxindent + 1)
+    wrapper = MBTextWrapper(width=width,
+                            initial_indent=initindent,
+                            subsequent_indent=hangindent)
+    return wrapper.fill(line)
 
 def iterlines(iterator):
     for chunk in iterator:

tests/test-encoding.out

@@ -121,18 +121,18 b' tip 5:db5'
 ?                                  3:770b9b11621d
 % hg tags (latin-1)
 tip                                5:db5520b4645f
-�                                  3:770b9b11621d
+�                                 3:770b9b11621d
 % hg tags (utf-8)
 tip                                5:db5520b4645f
-é                                  3:770b9b11621d
+é                                 3:770b9b11621d
 % hg branches (ascii)
 ?                              5:db5520b4645f
 default                        4:9cff3c980b58 (inactive)
 % hg branches (latin-1)
-�                              5:db5520b4645f
+�                             5:db5520b4645f
 default                        4:9cff3c980b58 (inactive)
 % hg branches (utf-8)
-é                              5:db5520b4645f
+é                             5:db5520b4645f
 default                        4:9cff3c980b58 (inactive)
 % hg log (utf-8)
 changeset:   5:db5520b4645f