Mercurial > genshi > genshi-test
comparison genshi/core.py @ 230:24757b771651
Renamed Markup to Genshi in repository.
author | cmlenz |
---|---|
date | Mon, 11 Sep 2006 15:07:07 +0000 |
parents | markup/core.py@e4dad1145f84 |
children | 8de2620504b9 |
comparison
equal
deleted
inserted
replaced
229:110d69dbbda3 | 230:24757b771651 |
---|---|
1 # -*- coding: utf-8 -*- | |
2 # | |
3 # Copyright (C) 2006 Edgewall Software | |
4 # All rights reserved. | |
5 # | |
6 # This software is licensed as described in the file COPYING, which | |
7 # you should have received as part of this distribution. The terms | |
8 # are also available at http://genshi.edgewall.org/wiki/License. | |
9 # | |
10 # This software consists of voluntary contributions made by many | |
11 # individuals. For the exact contribution history, see the revision | |
12 # history and logs, available at http://genshi.edgewall.org/log/. | |
13 | |
14 """Core classes for markup processing.""" | |
15 | |
16 import htmlentitydefs | |
17 import operator | |
18 import re | |
19 | |
20 __all__ = ['Stream', 'Markup', 'escape', 'unescape', 'Namespace', 'QName'] | |
21 | |
22 | |
23 class StreamEventKind(str): | |
24 """A kind of event on an XML stream.""" | |
25 | |
26 | |
27 class Stream(object): | |
28 """Represents a stream of markup events. | |
29 | |
30 This class is basically an iterator over the events. | |
31 | |
32 Also provided are ways to serialize the stream to text. The `serialize()` | |
33 method will return an iterator over generated strings, while `render()` | |
34 returns the complete generated text at once. Both accept various parameters | |
35 that impact the way the stream is serialized. | |
36 | |
37 Stream events are tuples of the form: | |
38 | |
39 (kind, data, position) | |
40 | |
41 where `kind` is the event kind (such as `START`, `END`, `TEXT`, etc), `data` | |
42 depends on the kind of event, and `position` is a `(filename, line, offset)` | |
43 tuple that contains the location of the original element or text in the | |
44 input. If the original location is unknown, `position` is `(None, -1, -1)`. | |
45 """ | |
46 __slots__ = ['events'] | |
47 | |
48 START = StreamEventKind('START') # a start tag | |
49 END = StreamEventKind('END') # an end tag | |
50 TEXT = StreamEventKind('TEXT') # literal text | |
51 DOCTYPE = StreamEventKind('DOCTYPE') # doctype declaration | |
52 START_NS = StreamEventKind('START_NS') # start namespace mapping | |
53 END_NS = StreamEventKind('END_NS') # end namespace mapping | |
54 START_CDATA = StreamEventKind('START_CDATA') # start CDATA section | |
55 END_CDATA = StreamEventKind('END_CDATA') # end CDATA section | |
56 PI = StreamEventKind('PI') # processing instruction | |
57 COMMENT = StreamEventKind('COMMENT') # comment | |
58 | |
59 def __init__(self, events): | |
60 """Initialize the stream with a sequence of markup events. | |
61 | |
62 @param events: a sequence or iterable providing the events | |
63 """ | |
64 self.events = events | |
65 | |
66 def __iter__(self): | |
67 return iter(self.events) | |
68 | |
69 def __or__(self, function): | |
70 """Override the "bitwise or" operator to apply filters or serializers | |
71 to the stream, providing a syntax similar to pipes on Unix shells. | |
72 | |
73 Assume the following stream produced by the `HTML` function: | |
74 | |
75 >>> from genshi.input import HTML | |
76 >>> html = HTML('''<p onclick="alert('Whoa')">Hello, world!</p>''') | |
77 >>> print html | |
78 <p onclick="alert('Whoa')">Hello, world!</p> | |
79 | |
80 A filter such as the HTML sanitizer can be applied to that stream using | |
81 the pipe notation as follows: | |
82 | |
83 >>> from genshi.filters import HTMLSanitizer | |
84 >>> sanitizer = HTMLSanitizer() | |
85 >>> print html | sanitizer | |
86 <p>Hello, world!</p> | |
87 | |
88 Filters can be any function that accepts and produces a stream (where | |
89 a stream is anything that iterators over events): | |
90 | |
91 >>> def uppercase(stream): | |
92 ... for kind, data, pos in stream: | |
93 ... if kind is TEXT: | |
94 ... data = data.upper() | |
95 ... yield kind, data, pos | |
96 >>> print html | sanitizer | uppercase | |
97 <p>HELLO, WORLD!</p> | |
98 | |
99 Serializers can also be used with this notation: | |
100 | |
101 >>> from genshi.output import TextSerializer | |
102 >>> output = TextSerializer() | |
103 >>> print html | sanitizer | uppercase | output | |
104 HELLO, WORLD! | |
105 | |
106 Commonly, serializers should be used at the end of the "pipeline"; | |
107 using them somewhere in the middle may produce unexpected results. | |
108 """ | |
109 return Stream(_ensure(function(self))) | |
110 | |
111 def filter(self, *filters): | |
112 """Apply filters to the stream. | |
113 | |
114 This method returns a new stream with the given filters applied. The | |
115 filters must be callables that accept the stream object as parameter, | |
116 and return the filtered stream. | |
117 | |
118 The call: | |
119 | |
120 stream.filter(filter1, filter2) | |
121 | |
122 is equivalent to: | |
123 | |
124 stream | filter1 | filter2 | |
125 """ | |
126 return reduce(operator.or_, (self,) + filters) | |
127 | |
128 def render(self, method='xml', encoding='utf-8', **kwargs): | |
129 """Return a string representation of the stream. | |
130 | |
131 @param method: determines how the stream is serialized; can be either | |
132 "xml", "xhtml", "html", "text", or a custom serializer | |
133 class | |
134 @param encoding: how the output string should be encoded; if set to | |
135 `None`, this method returns a `unicode` object | |
136 | |
137 Any additional keyword arguments are passed to the serializer, and thus | |
138 depend on the `method` parameter value. | |
139 """ | |
140 generator = self.serialize(method=method, **kwargs) | |
141 output = u''.join(list(generator)) | |
142 if encoding is not None: | |
143 errors = 'replace' | |
144 if method != 'text': | |
145 errors = 'xmlcharrefreplace' | |
146 return output.encode(encoding, errors) | |
147 return output | |
148 | |
149 def select(self, path): | |
150 """Return a new stream that contains the events matching the given | |
151 XPath expression. | |
152 | |
153 @param path: a string containing the XPath expression | |
154 """ | |
155 from genshi.path import Path | |
156 return Path(path).select(self) | |
157 | |
158 def serialize(self, method='xml', **kwargs): | |
159 """Generate strings corresponding to a specific serialization of the | |
160 stream. | |
161 | |
162 Unlike the `render()` method, this method is a generator that returns | |
163 the serialized output incrementally, as opposed to returning a single | |
164 string. | |
165 | |
166 @param method: determines how the stream is serialized; can be either | |
167 "xml", "xhtml", "html", "text", or a custom serializer | |
168 class | |
169 | |
170 Any additional keyword arguments are passed to the serializer, and thus | |
171 depend on the `method` parameter value. | |
172 """ | |
173 from genshi import output | |
174 cls = method | |
175 if isinstance(method, basestring): | |
176 cls = {'xml': output.XMLSerializer, | |
177 'xhtml': output.XHTMLSerializer, | |
178 'html': output.HTMLSerializer, | |
179 'text': output.TextSerializer}[method] | |
180 return cls(**kwargs)(_ensure(self)) | |
181 | |
182 def __str__(self): | |
183 return self.render() | |
184 | |
185 def __unicode__(self): | |
186 return self.render(encoding=None) | |
187 | |
188 | |
189 START = Stream.START | |
190 END = Stream.END | |
191 TEXT = Stream.TEXT | |
192 DOCTYPE = Stream.DOCTYPE | |
193 START_NS = Stream.START_NS | |
194 END_NS = Stream.END_NS | |
195 START_CDATA = Stream.START_CDATA | |
196 END_CDATA = Stream.END_CDATA | |
197 PI = Stream.PI | |
198 COMMENT = Stream.COMMENT | |
199 | |
200 def _ensure(stream): | |
201 """Ensure that every item on the stream is actually a markup event.""" | |
202 for event in stream: | |
203 if type(event) is not tuple: | |
204 if hasattr(event, 'totuple'): | |
205 event = event.totuple() | |
206 else: | |
207 event = TEXT, unicode(event), (None, -1, -1) | |
208 yield event | |
209 | |
210 | |
211 class Attrs(list): | |
212 """Sequence type that stores the attributes of an element. | |
213 | |
214 The order of the attributes is preserved, while accessing and manipulating | |
215 attributes by name is also supported. | |
216 | |
217 >>> attrs = Attrs([('href', '#'), ('title', 'Foo')]) | |
218 >>> attrs | |
219 [(u'href', '#'), (u'title', 'Foo')] | |
220 | |
221 >>> 'href' in attrs | |
222 True | |
223 >>> 'tabindex' in attrs | |
224 False | |
225 | |
226 >>> attrs.get(u'title') | |
227 'Foo' | |
228 >>> attrs.set(u'title', 'Bar') | |
229 >>> attrs | |
230 [(u'href', '#'), (u'title', 'Bar')] | |
231 >>> attrs.remove(u'title') | |
232 >>> attrs | |
233 [(u'href', '#')] | |
234 | |
235 New attributes added using the `set()` method are appended to the end of | |
236 the list: | |
237 | |
238 >>> attrs.set(u'accesskey', 'k') | |
239 >>> attrs | |
240 [(u'href', '#'), (u'accesskey', 'k')] | |
241 | |
242 An `Attrs` instance can also be initialized with keyword arguments. | |
243 | |
244 >>> attrs = Attrs(class_='bar', href='#', title='Foo') | |
245 >>> attrs.get('class') | |
246 'bar' | |
247 >>> attrs.get('href') | |
248 '#' | |
249 >>> attrs.get('title') | |
250 'Foo' | |
251 | |
252 Reserved words can be used by appending a trailing underscore to the name, | |
253 and any other underscore is replaced by a dash: | |
254 | |
255 >>> attrs = Attrs(class_='bar', accept_charset='utf-8') | |
256 >>> attrs.get('class') | |
257 'bar' | |
258 >>> attrs.get('accept-charset') | |
259 'utf-8' | |
260 | |
261 Thus this shorthand can not be used if attribute names should contain | |
262 actual underscore characters. | |
263 """ | |
264 __slots__ = [] | |
265 | |
266 def __init__(self, attrib=None, **kwargs): | |
267 """Create the `Attrs` instance. | |
268 | |
269 If the `attrib` parameter is provided, it is expected to be a sequence | |
270 of `(name, value)` tuples. | |
271 """ | |
272 if attrib is None: | |
273 attrib = [] | |
274 list.__init__(self, [(QName(name), value) for name, value in attrib]) | |
275 for name, value in kwargs.items(): | |
276 self.set(name.rstrip('_').replace('_', '-'), value) | |
277 | |
278 def __contains__(self, name): | |
279 """Return whether the list includes an attribute with the specified | |
280 name. | |
281 """ | |
282 for attr, _ in self: | |
283 if attr == name: | |
284 return True | |
285 | |
286 def get(self, name, default=None): | |
287 """Return the value of the attribute with the specified name, or the | |
288 value of the `default` parameter if no such attribute is found. | |
289 """ | |
290 for attr, value in self: | |
291 if attr == name: | |
292 return value | |
293 return default | |
294 | |
295 def remove(self, name): | |
296 """Remove the attribute with the specified name. | |
297 | |
298 If no such attribute is found, this method does nothing. | |
299 """ | |
300 for idx, (attr, _) in enumerate(self): | |
301 if attr == name: | |
302 del self[idx] | |
303 break | |
304 | |
305 def set(self, name, value): | |
306 """Set the specified attribute to the given value. | |
307 | |
308 If an attribute with the specified name is already in the list, the | |
309 value of the existing entry is updated. Otherwise, a new attribute is | |
310 appended to the end of the list. | |
311 """ | |
312 for idx, (attr, _) in enumerate(self): | |
313 if attr == name: | |
314 self[idx] = (QName(attr), value) | |
315 break | |
316 else: | |
317 self.append((QName(name), value)) | |
318 | |
319 def totuple(self): | |
320 """Return the attributes as a markup event. | |
321 | |
322 The returned event is a TEXT event, the data is the value of all | |
323 attributes joined together. | |
324 """ | |
325 return TEXT, u''.join([x[1] for x in self]), (None, -1, -1) | |
326 | |
327 | |
328 def plaintext(text, keeplinebreaks=True): | |
329 """Returns the text as a `unicode` string with all entities and tags | |
330 removed. | |
331 """ | |
332 text = stripentities(striptags(text)) | |
333 if not keeplinebreaks: | |
334 text = text.replace(u'\n', u' ') | |
335 return text | |
336 | |
337 def stripentities(text, keepxmlentities=False): | |
338 """Return a copy of the given text with any character or numeric entities | |
339 replaced by the equivalent UTF-8 characters. | |
340 | |
341 If the `keepxmlentities` parameter is provided and evaluates to `True`, | |
342 the core XML entities (&, ', >, < and ") are not | |
343 stripped. | |
344 """ | |
345 def _replace_entity(match): | |
346 if match.group(1): # numeric entity | |
347 ref = match.group(1) | |
348 if ref.startswith('x'): | |
349 ref = int(ref[1:], 16) | |
350 else: | |
351 ref = int(ref, 10) | |
352 return unichr(ref) | |
353 else: # character entity | |
354 ref = match.group(2) | |
355 if keepxmlentities and ref in ('amp', 'apos', 'gt', 'lt', 'quot'): | |
356 return '&%s;' % ref | |
357 try: | |
358 codepoint = htmlentitydefs.name2codepoint[ref] | |
359 return unichr(codepoint) | |
360 except KeyError: | |
361 if keepxmlentities: | |
362 return '&%s;' % ref | |
363 else: | |
364 return ref | |
365 return re.sub(r'&(?:#((?:\d+)|(?:[xX][0-9a-fA-F]+));?|(\w+);)', | |
366 _replace_entity, text) | |
367 | |
368 def striptags(text): | |
369 """Return a copy of the text with all XML/HTML tags removed.""" | |
370 return re.sub(r'<[^>]*?>', '', text) | |
371 | |
372 | |
373 class Markup(unicode): | |
374 """Marks a string as being safe for inclusion in HTML/XML output without | |
375 needing to be escaped. | |
376 """ | |
377 __slots__ = [] | |
378 | |
379 def __new__(cls, text='', *args): | |
380 if args: | |
381 text %= tuple(map(escape, args)) | |
382 return unicode.__new__(cls, text) | |
383 | |
384 def __add__(self, other): | |
385 return Markup(unicode(self) + unicode(escape(other))) | |
386 | |
387 def __radd__(self, other): | |
388 return Markup(unicode(escape(other)) + unicode(self)) | |
389 | |
390 def __mod__(self, args): | |
391 if not isinstance(args, (list, tuple)): | |
392 args = [args] | |
393 return Markup(unicode.__mod__(self, tuple(map(escape, args)))) | |
394 | |
395 def __mul__(self, num): | |
396 return Markup(unicode(self) * num) | |
397 | |
398 def __rmul__(self, num): | |
399 return Markup(num * unicode(self)) | |
400 | |
401 def __repr__(self): | |
402 return '<%s "%s">' % (self.__class__.__name__, self) | |
403 | |
404 def join(self, seq, escape_quotes=True): | |
405 return Markup(unicode(self).join([escape(item, quotes=escape_quotes) | |
406 for item in seq])) | |
407 | |
408 def escape(cls, text, quotes=True): | |
409 """Create a Markup instance from a string and escape special characters | |
410 it may contain (<, >, & and \"). | |
411 | |
412 If the `quotes` parameter is set to `False`, the \" character is left | |
413 as is. Escaping quotes is generally only required for strings that are | |
414 to be used in attribute values. | |
415 """ | |
416 if not text: | |
417 return cls() | |
418 if type(text) is cls: | |
419 return text | |
420 text = unicode(text).replace('&', '&') \ | |
421 .replace('<', '<') \ | |
422 .replace('>', '>') | |
423 if quotes: | |
424 text = text.replace('"', '"') | |
425 return cls(text) | |
426 escape = classmethod(escape) | |
427 | |
428 def unescape(self): | |
429 """Reverse-escapes &, <, > and \" and returns a `unicode` object.""" | |
430 if not self: | |
431 return u'' | |
432 return unicode(self).replace('"', '"') \ | |
433 .replace('>', '>') \ | |
434 .replace('<', '<') \ | |
435 .replace('&', '&') | |
436 | |
437 def stripentities(self, keepxmlentities=False): | |
438 """Return a copy of the text with any character or numeric entities | |
439 replaced by the equivalent UTF-8 characters. | |
440 | |
441 If the `keepxmlentities` parameter is provided and evaluates to `True`, | |
442 the core XML entities (&, ', >, < and ") are not | |
443 stripped. | |
444 """ | |
445 return Markup(stripentities(self, keepxmlentities=keepxmlentities)) | |
446 | |
447 def striptags(self): | |
448 """Return a copy of the text with all XML/HTML tags removed.""" | |
449 return Markup(striptags(self)) | |
450 | |
451 | |
452 escape = Markup.escape | |
453 | |
454 def unescape(text): | |
455 """Reverse-escapes &, <, > and \" and returns a `unicode` object.""" | |
456 if not isinstance(text, Markup): | |
457 return text | |
458 return text.unescape() | |
459 | |
460 | |
461 class Namespace(object): | |
462 """Utility class creating and testing elements with a namespace. | |
463 | |
464 Internally, namespace URIs are encoded in the `QName` of any element or | |
465 attribute, the namespace URI being enclosed in curly braces. This class | |
466 helps create and test these strings. | |
467 | |
468 A `Namespace` object is instantiated with the namespace URI. | |
469 | |
470 >>> html = Namespace('http://www.w3.org/1999/xhtml') | |
471 >>> html | |
472 <Namespace "http://www.w3.org/1999/xhtml"> | |
473 >>> html.uri | |
474 u'http://www.w3.org/1999/xhtml' | |
475 | |
476 The `Namespace` object can than be used to generate `QName` objects with | |
477 that namespace: | |
478 | |
479 >>> html.body | |
480 u'{http://www.w3.org/1999/xhtml}body' | |
481 >>> html.body.localname | |
482 u'body' | |
483 >>> html.body.namespace | |
484 u'http://www.w3.org/1999/xhtml' | |
485 | |
486 The same works using item access notation, which is useful for element or | |
487 attribute names that are not valid Python identifiers: | |
488 | |
489 >>> html['body'] | |
490 u'{http://www.w3.org/1999/xhtml}body' | |
491 | |
492 A `Namespace` object can also be used to test whether a specific `QName` | |
493 belongs to that namespace using the `in` operator: | |
494 | |
495 >>> qname = html.body | |
496 >>> qname in html | |
497 True | |
498 >>> qname in Namespace('http://www.w3.org/2002/06/xhtml2') | |
499 False | |
500 """ | |
501 def __new__(cls, uri): | |
502 if type(uri) is cls: | |
503 return uri | |
504 return object.__new__(cls, uri) | |
505 | |
506 def __init__(self, uri): | |
507 self.uri = unicode(uri) | |
508 | |
509 def __contains__(self, qname): | |
510 return qname.namespace == self.uri | |
511 | |
512 def __eq__(self, other): | |
513 if isinstance(other, Namespace): | |
514 return self.uri == other.uri | |
515 return self.uri == other | |
516 | |
517 def __getitem__(self, name): | |
518 return QName(self.uri + u'}' + name) | |
519 __getattr__ = __getitem__ | |
520 | |
521 def __repr__(self): | |
522 return '<Namespace "%s">' % self.uri | |
523 | |
524 def __str__(self): | |
525 return self.uri.encode('utf-8') | |
526 | |
527 def __unicode__(self): | |
528 return self.uri | |
529 | |
530 | |
531 # The namespace used by attributes such as xml:lang and xml:space | |
532 XML_NAMESPACE = Namespace('http://www.w3.org/XML/1998/namespace') | |
533 | |
534 | |
535 class QName(unicode): | |
536 """A qualified element or attribute name. | |
537 | |
538 The unicode value of instances of this class contains the qualified name of | |
539 the element or attribute, in the form `{namespace}localname`. The namespace | |
540 URI can be obtained through the additional `namespace` attribute, while the | |
541 local name can be accessed through the `localname` attribute. | |
542 | |
543 >>> qname = QName('foo') | |
544 >>> qname | |
545 u'foo' | |
546 >>> qname.localname | |
547 u'foo' | |
548 >>> qname.namespace | |
549 | |
550 >>> qname = QName('http://www.w3.org/1999/xhtml}body') | |
551 >>> qname | |
552 u'{http://www.w3.org/1999/xhtml}body' | |
553 >>> qname.localname | |
554 u'body' | |
555 >>> qname.namespace | |
556 u'http://www.w3.org/1999/xhtml' | |
557 """ | |
558 __slots__ = ['namespace', 'localname'] | |
559 | |
560 def __new__(cls, qname): | |
561 if type(qname) is cls: | |
562 return qname | |
563 | |
564 parts = qname.split(u'}', 1) | |
565 if len(parts) > 1: | |
566 self = unicode.__new__(cls, u'{%s' % qname) | |
567 self.namespace, self.localname = map(unicode, parts) | |
568 else: | |
569 self = unicode.__new__(cls, qname) | |
570 self.namespace, self.localname = None, unicode(qname) | |
571 return self |