Mercurial > genshi > mirror
comparison examples/trac/trac/util/text.py @ 39:93b4dcbafd7b trunk
Copy Trac to main branch.
| author | cmlenz |
|---|---|
| date | Mon, 03 Jul 2006 18:53:27 +0000 |
| parents | |
| children |
comparison
equal
deleted
inserted
replaced
| 38:ee669cb9cccc | 39:93b4dcbafd7b |
|---|---|
| 1 # -*- coding: utf-8 -*- | |
| 2 # | |
| 3 # Copyright (C) 2003-2006 Edgewall Software | |
| 4 # Copyright (C) 2003-2004 Jonas Borgström <jonas@edgewall.com> | |
| 5 # Copyright (C) 2006 Matthew Good <trac@matt-good.net> | |
| 6 # Copyright (C) 2005-2006 Christian Boos <cboos@neuf.fr> | |
| 7 # All rights reserved. | |
| 8 # | |
| 9 # This software is licensed as described in the file COPYING, which | |
| 10 # you should have received as part of this distribution. The terms | |
| 11 # are also available at http://trac.edgewall.com/license.html. | |
| 12 # | |
| 13 # This software consists of voluntary contributions made by many | |
| 14 # individuals. For exact contribution history, see the revision | |
| 15 # history and logs, available at http://projects.edgewall.com/trac/. | |
| 16 # | |
| 17 # Author: Jonas Borgström <jonas@edgewall.com> | |
| 18 # Matthew Good <trac@matt-good.net> | |
| 19 # Christian Boos <cboos@neuf.fr> | |
| 20 | |
| 21 import locale | |
| 22 import os | |
| 23 import sys | |
| 24 from urllib import quote, unquote, urlencode | |
| 25 | |
| 26 | |
| 27 CRLF = '\r\n' | |
| 28 | |
| 29 # -- Unicode | |
| 30 | |
| 31 def to_unicode(text, charset=None, lossy=True): | |
| 32 """Convert a `str` object to an `unicode` object. | |
| 33 | |
| 34 If `charset` is not specified, we'll make some guesses, | |
| 35 first trying the UTF-8 encoding then trying the locale | |
| 36 preferred encoding (this differs from the `unicode` function | |
| 37 which only tries with the locale preferred encoding, in 'strict' | |
| 38 mode). | |
| 39 | |
| 40 If the `lossy` argument is `True`, which is the default, then | |
| 41 we use the 'replace' mode: | |
| 42 | |
| 43 If the `lossy` argument is `False`, we fallback to the 'iso-8859-15' | |
| 44 charset in case of an error (encoding a `str` using 'iso-8859-15' | |
| 45 will always work, as there's one Unicode character for each byte of | |
| 46 the input). | |
| 47 """ | |
| 48 if not isinstance(text, str): | |
| 49 if isinstance(text, Exception): | |
| 50 # two possibilities for storing unicode strings in exception data: | |
| 51 try: | |
| 52 # custom __str__ method on the exception (e.g. PermissionError) | |
| 53 return unicode(text) | |
| 54 except UnicodeError: | |
| 55 # unicode arguments given to the exception (e.g. parse_date) | |
| 56 return ' '.join([to_unicode(arg) for arg in text.args]) | |
| 57 return unicode(text) | |
| 58 errors = lossy and 'replace' or 'strict' | |
| 59 try: | |
| 60 if charset: | |
| 61 return unicode(text, charset, errors) | |
| 62 else: | |
| 63 try: | |
| 64 return unicode(text, 'utf-8') | |
| 65 except UnicodeError: | |
| 66 return unicode(text, locale.getpreferredencoding(), errors) | |
| 67 except UnicodeError: | |
| 68 return unicode(text, 'iso-8859-15') | |
| 69 | |
| 70 def unicode_quote(value): | |
| 71 """A unicode aware version of urllib.quote""" | |
| 72 return quote(value.encode('utf-8')) | |
| 73 | |
| 74 def unicode_unquote(value): | |
| 75 """A unicode aware version of urllib.unquote. | |
| 76 | |
| 77 Take `str` value previously obtained by `unicode_quote`. | |
| 78 """ | |
| 79 return unquote(value).decode('utf-8') | |
| 80 | |
| 81 def unicode_urlencode(params): | |
| 82 """A unicode aware version of urllib.urlencode""" | |
| 83 if isinstance(params, dict): | |
| 84 params = params.items() | |
| 85 return urlencode([(k, isinstance(v, unicode) and v.encode('utf-8') or v) | |
| 86 for k, v in params]) | |
| 87 | |
| 88 def to_utf8(text, charset='iso-8859-15'): | |
| 89 """Convert a string to UTF-8, assuming the encoding is either UTF-8, ISO | |
| 90 Latin-1, or as specified by the optional `charset` parameter. | |
| 91 | |
| 92 ''Deprecated in 0.10. You should use `unicode` strings only.'' | |
| 93 """ | |
| 94 try: | |
| 95 # Do nothing if it's already utf-8 | |
| 96 u = unicode(text, 'utf-8') | |
| 97 return text | |
| 98 except UnicodeError: | |
| 99 try: | |
| 100 # Use the user supplied charset if possible | |
| 101 u = unicode(text, charset) | |
| 102 except UnicodeError: | |
| 103 # This should always work | |
| 104 u = unicode(text, 'iso-8859-15') | |
| 105 return u.encode('utf-8') | |
| 106 | |
| 107 | |
| 108 # -- Plain text formatting | |
| 109 | |
| 110 def shorten_line(text, maxlen=75): | |
| 111 if len(text or '') < maxlen: | |
| 112 return text | |
| 113 shortline = text[:maxlen] | |
| 114 cut = shortline.rfind(' ') + 1 or shortline.rfind('\n') + 1 or maxlen | |
| 115 shortline = text[:cut]+' ...' | |
| 116 return shortline | |
| 117 | |
| 118 def wrap(t, cols=75, initial_indent='', subsequent_indent='', | |
| 119 linesep=os.linesep): | |
| 120 try: | |
| 121 import textwrap | |
| 122 t = t.strip().replace('\r\n', '\n').replace('\r', '\n') | |
| 123 wrapper = textwrap.TextWrapper(cols, replace_whitespace=0, | |
| 124 break_long_words=0, | |
| 125 initial_indent=initial_indent, | |
| 126 subsequent_indent=subsequent_indent) | |
| 127 wrappedLines = [] | |
| 128 for line in t.split('\n'): | |
| 129 wrappedLines += wrapper.wrap(line.rstrip()) or [''] | |
| 130 return linesep.join(wrappedLines) | |
| 131 | |
| 132 except ImportError: | |
| 133 return t | |
| 134 | |
| 135 | |
| 136 # -- Conversion | |
| 137 | |
| 138 def pretty_size(size): | |
| 139 if size is None: | |
| 140 return '' | |
| 141 | |
| 142 jump = 512 | |
| 143 if size < jump: | |
| 144 return '%d bytes' % size | |
| 145 | |
| 146 units = ['kB', 'MB', 'GB', 'TB'] | |
| 147 i = 0 | |
| 148 while size >= jump and i < len(units): | |
| 149 i += 1 | |
| 150 size /= 1024. | |
| 151 | |
| 152 return '%.1f %s' % (size, units[i - 1]) |