1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227 | import warnings
from io import StringIO
from django.template.base import TRANSLATOR_COMMENT_MARK, Lexer, TokenType
from django.utils.regex_helper import _lazy_re_compile
from . import TranslatorCommentWarning, trim_whitespace
dot_re = _lazy_re_compile(r'\S')
def blankout(src, char):
"""
Change every non-whitespace character to the given char.
Used in the templatize function.
"""
return dot_re.sub(char, src)
context_re = _lazy_re_compile(r"""^\s+.*context\s+((?:"[^"]*?")|(?:'[^']*?'))\s*""")
inline_re = _lazy_re_compile(
# Match the trans/translate 'some text' part.
r"""^\s*trans(?:late)?\s+((?:"[^"]*?")|(?:'[^']*?'))"""
# Match and ignore optional filters
r"""(?:\s*\|\s*[^\s:]+(?::(?:[^\s'":]+|(?:"[^"]*?")|(?:'[^']*?')))?)*"""
# Match the optional context part
r"""(\s+.*context\s+((?:"[^"]*?")|(?:'[^']*?')))?\s*"""
)
block_re = _lazy_re_compile(r"""^\s*blocktrans(?:late)?(\s+.*context\s+((?:"[^"]*?")|(?:'[^']*?')))?(?:\s+|$)""")
endblock_re = _lazy_re_compile(r"""^\s*endblocktrans(?:late)?$""")
plural_re = _lazy_re_compile(r"""^\s*plural$""")
constant_re = _lazy_re_compile(r"""_\(((?:".*?")|(?:'.*?'))\)""")
def templatize(src, origin=None):
"""
Turn a Django template into something that is understood by xgettext. It
does so by translating the Django translation tags into standard gettext
function invocations.
"""
out = StringIO('')
message_context = None
intrans = False
inplural = False
trimmed = False
singular = []
plural = []
incomment = False
comment = []
lineno_comment_map = {}
comment_lineno_cache = None
# Adding the u prefix allows gettext to recognize the string (#26093).
raw_prefix = 'u'
def join_tokens(tokens, trim=False):
message = ''.join(tokens)
if trim:
message = trim_whitespace(message)
return message
for t in Lexer(src).tokenize():
if incomment:
if t.token_type == TokenType.BLOCK and t.contents == 'endcomment':
content = ''.join(comment)
translators_comment_start = None
for lineno, line in enumerate(content.splitlines(True)):
if line.lstrip().startswith(TRANSLATOR_COMMENT_MARK):
translators_comment_start = lineno
for lineno, line in enumerate(content.splitlines(True)):
if translators_comment_start is not None and lineno >= translators_comment_start:
out.write(' # %s' % line)
else:
out.write(' #\n')
incomment = False
comment = []
else:
comment.append(t.contents)
elif intrans:
if t.token_type == TokenType.BLOCK:
endbmatch = endblock_re.match(t.contents)
pluralmatch = plural_re.match(t.contents)
if endbmatch:
if inplural:
if message_context:
out.write(' npgettext({p}{!r}, {p}{!r}, {p}{!r},count) '.format(
message_context,
join_tokens(singular, trimmed),
join_tokens(plural, trimmed),
p=raw_prefix,
))
else:
out.write(' ngettext({p}{!r}, {p}{!r}, count) '.format(
join_tokens(singular, trimmed),
join_tokens(plural, trimmed),
p=raw_prefix,
))
for part in singular:
out.write(blankout(part, 'S'))
for part in plural:
out.write(blankout(part, 'P'))
else:
if message_context:
out.write(' pgettext({p}{!r}, {p}{!r}) '.format(
message_context,
join_tokens(singular, trimmed),
p=raw_prefix,
))
else:
out.write(' gettext({p}{!r}) '.format(
join_tokens(singular, trimmed),
p=raw_prefix,
))
for part in singular:
out.write(blankout(part, 'S'))
message_context = None
intrans = False
inplural = False
singular = []
plural = []
elif pluralmatch:
inplural = True
else:
filemsg = ''
if origin:
filemsg = 'file %s, ' % origin
raise SyntaxError(
"Translation blocks must not include other block tags: "
"%s (%sline %d)" % (t.contents, filemsg, t.lineno)
)
elif t.token_type == TokenType.VAR:
if inplural:
plural.append('%%(%s)s' % t.contents)
else:
singular.append('%%(%s)s' % t.contents)
elif t.token_type == TokenType.TEXT:
contents = t.contents.replace('%', '%%')
if inplural:
plural.append(contents)
else:
singular.append(contents)
else:
# Handle comment tokens (`{# ... #}`) plus other constructs on
# the same line:
if comment_lineno_cache is not None:
cur_lineno = t.lineno + t.contents.count('\n')
if comment_lineno_cache == cur_lineno:
if t.token_type != TokenType.COMMENT:
for c in lineno_comment_map[comment_lineno_cache]:
filemsg = ''
if origin:
filemsg = 'file %s, ' % origin
warn_msg = (
"The translator-targeted comment '%s' "
"(%sline %d) was ignored, because it wasn't "
"the last item on the line."
) % (c, filemsg, comment_lineno_cache)
warnings.warn(warn_msg, TranslatorCommentWarning)
lineno_comment_map[comment_lineno_cache] = []
else:
out.write('# %s' % ' | '.join(lineno_comment_map[comment_lineno_cache]))
comment_lineno_cache = None
if t.token_type == TokenType.BLOCK:
imatch = inline_re.match(t.contents)
bmatch = block_re.match(t.contents)
cmatches = constant_re.findall(t.contents)
if imatch:
g = imatch[1]
if g[0] == '"':
g = g.strip('"')
elif g[0] == "'":
g = g.strip("'")
g = g.replace('%', '%%')
if imatch[2]:
# A context is provided
context_match = context_re.match(imatch[2])
message_context = context_match[1]
if message_context[0] == '"':
message_context = message_context.strip('"')
elif message_context[0] == "'":
message_context = message_context.strip("'")
out.write(' pgettext({p}{!r}, {p}{!r}) '.format(
message_context, g, p=raw_prefix
))
message_context = None
else:
out.write(' gettext({p}{!r}) '.format(g, p=raw_prefix))
elif bmatch:
for fmatch in constant_re.findall(t.contents):
out.write(' _(%s) ' % fmatch)
if bmatch[1]:
# A context is provided
context_match = context_re.match(bmatch[1])
message_context = context_match[1]
if message_context[0] == '"':
message_context = message_context.strip('"')
elif message_context[0] == "'":
message_context = message_context.strip("'")
intrans = True
inplural = False
trimmed = 'trimmed' in t.split_contents()
singular = []
plural = []
elif cmatches:
for cmatch in cmatches:
out.write(' _(%s) ' % cmatch)
elif t.contents == 'comment':
incomment = True
else:
out.write(blankout(t.contents, 'B'))
elif t.token_type == TokenType.VAR:
parts = t.contents.split('|')
cmatch = constant_re.match(parts[0])
if cmatch:
out.write(' _(%s) ' % cmatch[1])
for p in parts[1:]:
if p.find(':_(') >= 0:
out.write(' %s ' % p.split(':', 1)[1])
else:
out.write(blankout(p, 'F'))
elif t.token_type == TokenType.COMMENT:
if t.contents.lstrip().startswith(TRANSLATOR_COMMENT_MARK):
lineno_comment_map.setdefault(t.lineno, []).append(t.contents)
comment_lineno_cache = t.lineno
else:
out.write(blankout(t.contents, 'X'))
return out.getvalue()
|