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
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
|
from jinja2 import nodes
from jinja2.ext import Extension
from jinja2.exceptions import TemplateSyntaxError
from django.conf import settings
from coffin.template import Library
class LoadExtension(Extension):
"""The load-tag is a no-op in Coffin. Instead, all template libraries
are always loaded.
Note: Supporting a functioning load-tag in Jinja is tough, though
theoretically possible. The trouble is activating new extensions while
parsing is ongoing. The ``Parser.extensions`` dict of the current
parser instance needs to be modified, but apparently the only way to
get access would be by hacking the stack.
"""
tags = set(['load'])
def parse(self, parser):
while not parser.stream.current.type == 'block_end':
parser.stream.next()
return []
"""class AutoescapeExtension(Extension):
""#"
Template to output works in three phases in Jinja2: parsing,
generation (compilation, AST-traversal), and rendering (execution).
Unfortunatly, the environment ``autoescape`` option comes into effect
during traversal, the part where we happen to have basically no control
over as an extension. It determines whether output is wrapped in
``escape()`` calls.
Solutions that could possibly work:
* This extension could preprocess it's childnodes and wrap
everything output related inside the appropriate
``Markup()`` or escape() call.
* We could use the ``preprocess`` hook to insert the
appropriate ``|safe`` and ``|escape`` filters on a
string-basis. This is very unlikely to work well.
There's also the issue of inheritance and just generally the nesting
of autoescape-tags to consider.
Other things of note:
* We can access ``parser.environment``, but that would only
affect the **parsing** of our child nodes.
* In the commented-out code below we are trying to affect the
autoescape setting during rendering. As noted, this could be
necessary for rare border cases where custom extension use
the autoescape attribute.
Both the above things would break Environment thread-safety though!
Overall, it's not looking to good for this extension.
""#"
tags = ['autoescape']
def parse(self, parser):
lineno = parser.stream.next().lineno
old_autoescape = parser.environment.autoescape
parser.environment.autoescape = True
try:
body = parser.parse_statements(
['name:endautoescape'], drop_needle=True)
finally:
parser.environment.autoescape = old_autoescape
# Not sure yet if the code below is necessary - it changes
# environment.autoescape during template rendering. If for example
# a CallBlock function accesses ``environment.autoescape``, it
# presumably is.
# This also should use try-finally though, which Jinja's API
# doesn't support either. We could fake that as well by using
# InternalNames that output the necessary indentation and keywords,
# but at this point it starts to get really messy.
#
# TODO: Actually, there's ``nodes.EnvironmentAttribute``.
#ae_setting = object.__new__(nodes.InternalName)
#nodes.Node.__init__(ae_setting, 'environment.autoescape', lineno=lineno)
#temp = parser.free_identifier()
#body.insert(0, nodes.Assign(temp, ae_setting))
#body.insert(1, nodes.Assign(ae_setting, nodes.Const(True)))
#body.insert(len(body), nodes.Assign(ae_setting, temp))
return body
"""
class URLExtension(Extension):
"""Returns an absolute URL matching given view with its parameters.
This is a way to define links that aren't tied to a particular URL
configuration::
{% url path.to.some_view arg1,arg2,name1=value1 %}
Known differences to Django's url-Tag:
- In Django, the view name may contain any non-space character.
Since Jinja's lexer does not identify whitespace to us, only
characters that make up valid identifers, plus dots and hyphens
are allowed. Note that identifers in Jinja 2 may not contain
non-ascii characters.
As an alternative, you may specifify the view as a string,
which bypasses all these restrictions. It further allows you
to apply filters:
{% url "меткаda.some-view"|afilter %}
"""
tags = set(['url'])
def parse(self, parser):
stream = parser.stream
tag = stream.next()
# get view name
if stream.current.test('string'):
viewname = parser.parse_primary()
else:
# parse valid tokens and manually build a string from them
bits = []
name_allowed = True
while True:
if stream.current.test_any('dot', 'sub'):
bits.append(stream.next())
name_allowed = True
elif stream.current.test('name') and name_allowed:
bits.append(stream.next())
name_allowed = False
else:
break
viewname = nodes.Const("".join([b.value for b in bits]))
if not bits:
raise TemplateSyntaxError("'%s' requires path to view" %
tag.value, tag.lineno)
# get arguments
args = []
kwargs = []
while not stream.current.test_any('block_end', 'name:as'):
if args or kwargs:
stream.expect('comma')
if stream.current.test('name') and stream.look().test('assign'):
key = nodes.Const(stream.next().value)
stream.skip()
value = parser.parse_expression()
kwargs.append(nodes.Pair(key, value, lineno=key.lineno))
else:
args.append(parser.parse_expression())
make_call_node = lambda *kw: \
self.call_method('_reverse',
args=[viewname, nodes.List(args), nodes.Dict(kwargs)],
kwargs=kw)
# if an as-clause is specified, write the result to context...
if stream.next_if('name:as'):
var = nodes.Name(stream.expect('name').value, 'store')
call_node = make_call_node(nodes.Keyword('fail', nodes.Const(False)))
return nodes.Assign(var, call_node)
# ...otherwise print it out.
else:
return nodes.Output([make_call_node()]).set_lineno(tag.lineno)
@classmethod
def _reverse(self, viewname, args, kwargs, fail=True):
from django.core.urlresolvers import reverse, NoReverseMatch
# Try to look up the URL twice: once given the view name,
# and again relative to what we guess is the "main" app.
url = ''
try:
url = reverse(viewname, args=args, kwargs=kwargs)
except NoReverseMatch:
projectname = settings.SETTINGS_MODULE.split('.')[0]
try:
url = reverse(projectname + '.' + viewname,
args=args, kwargs=kwargs)
except NoReverseMatch:
if fail:
raise
else:
return ''
return url
class WithExtension(Extension):
"""Adds a value to the context (inside this block) for caching and
easy access, just like the Django-version does.
For example::
{% with person.some_sql_method as total %}
{{ total }} object{{ total|pluralize }}
{% endwith %}
TODO: The new Scope node introduced in Jinja2 6334c1eade73 (the 2.2
dev version) would help here, but we don't want to rely on that yet.
See also:
http://dev.pocoo.org/projects/jinja/browser/tests/test_ext.py
http://dev.pocoo.org/projects/jinja/ticket/331
http://dev.pocoo.org/projects/jinja/ticket/329
"""
tags = set(['with'])
def parse(self, parser):
lineno = parser.stream.next().lineno
value = parser.parse_expression()
parser.stream.expect('name:as')
name = parser.stream.expect('name')
body = parser.parse_statements(['name:endwith'], drop_needle=True)
return nodes.CallBlock(
self.call_method('_render_block', args=[value]),
[nodes.Name(name.value, 'store')], [], body).\
set_lineno(lineno)
def _render_block(self, value, caller=None):
return caller(value)
class CacheExtension(Extension):
"""Exactly like Django's own tag, but supports full Jinja2
expressiveness for all arguments.
{% cache gettimeout()*2 "foo"+options.cachename %}
...
{% endcache %}
This actually means that there is a considerable incompatibility
to Django: In Django, the second argument is simply a name, but
interpreted as a literal string. This tag, with Jinja2 stronger
emphasis on consistent syntax, requires you to actually specify the
quotes around the name to make it a string. Otherwise, allowing
Jinja2 expressions would be very hard to impossible (one could use
a lookahead to see if the name is followed by an operator, and
evaluate it as an expression if so, or read it as a string if not.
TODO: This may not be the right choice. Supporting expressions
here is probably not very important, so compatibility should maybe
prevail. Unfortunately, it is actually pretty hard to be compatibly
in all cases, simply because Django's per-character parser will
just eat everything until the next whitespace and consider it part
of the fragment name, while we have to work token-based: ``x*2``
would actually be considered ``"x*2"`` in Django, while Jinja2
would give us three tokens: ``x``, ``*``, ``2``.
General Syntax:
{% cache [expire_time] [fragment_name] [var1] [var2] .. %}
.. some expensive processing ..
{% endcache %}
Available by default (does not need to be loaded).
Partly based on the ``FragmentCacheExtension`` from the Jinja2 docs.
TODO: Should there be scoping issues with the internal dummy macro
limited access to certain outer variables in some cases, there is a
different way to write this. Generated code would look like this:
internal_name = environment.extensions['..']._get_cache_value():
if internal_name is not None:
yield internal_name
else:
internal_name = "" # or maybe use [] and append() for performance
internalname += "..."
internalname += "..."
internalname += "..."
environment.extensions['..']._set_cache_value(internalname):
yield internalname
In other words, instead of using a CallBlock which uses a local
function and calls into python, we have to separate calls into
python, but put the if-else logic itself into the compiled template.
"""
tags = set(['cache'])
def parse(self, parser):
lineno = parser.stream.next().lineno
expire_time = parser.parse_expression()
fragment_name = parser.parse_expression()
vary_on = []
while not parser.stream.current.test('block_end'):
vary_on.append(parser.parse_expression())
body = parser.parse_statements(['name:endcache'], drop_needle=True)
return nodes.CallBlock(
self.call_method('_cache_support',
[expire_time, fragment_name,
nodes.List(vary_on), nodes.Const(lineno)]),
[], [], body).set_lineno(lineno)
def _cache_support(self, expire_time, fragm_name, vary_on, lineno, caller):
from django.core.cache import cache # delay depending in settings
from django.utils.http import urlquote
try:
expire_time = int(expire_time)
except (ValueError, TypeError):
raise TemplateSyntaxError('"%s" tag got a non-integer '
'timeout value: %r' % (list(self.tags)[0], expire_time), lineno)
cache_key = u':'.join([fragm_name] + [urlquote(v) for v in vary_on])
value = cache.get(cache_key)
if value is None:
value = caller()
cache.set(cache_key, value, expire_time)
return value
class SpacelessExtension(Extension):
"""Removes whitespace between HTML tags, including tab and
newline characters.
Works exactly like Django's own tag.
"""
tags = ['spaceless']
def parse(self, parser):
lineno = parser.stream.next().lineno
body = parser.parse_statements(['name:endspaceless'], drop_needle=True)
return nodes.CallBlock(
self.call_method('_strip_spaces', [], [], None, None),
[], [], body
).set_lineno(lineno)
def _strip_spaces(self, caller=None):
from django.utils.html import strip_spaces_between_tags
return strip_spaces_between_tags(caller().strip())
# nicer import names
load = LoadExtension
url = URLExtension
with_ = WithExtension
cache = CacheExtension
spaceless = SpacelessExtension
register = Library()
register.tag(load)
register.tag(url)
register.tag(with_)
register.tag(cache)
register.tag(spaceless)
|