A Python parser for MediaWiki wikicode https://mwparserfromhell.readthedocs.io/
Ви не можете вибрати більше 25 тем Теми мають розпочинатися з літери або цифри, можуть містити дефіси (-) і не повинні перевищувати 35 символів.

11 роки тому
11 роки тому
11 роки тому
11 роки тому
11 роки тому
11 роки тому
11 роки тому
11 роки тому
11 роки тому
11 роки тому
11 роки тому
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206
  1. # -*- coding: utf-8 -*-
  2. #
  3. # Copyright (C) 2012 Ben Kurtovic <ben.kurtovic@verizon.net>
  4. #
  5. # Permission is hereby granted, free of charge, to any person obtaining a copy
  6. # of this software and associated documentation files (the "Software"), to deal
  7. # in the Software without restriction, including without limitation the rights
  8. # to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
  9. # copies of the Software, and to permit persons to whom the Software is
  10. # furnished to do so, subject to the following conditions:
  11. #
  12. # The above copyright notice and this permission notice shall be included in
  13. # all copies or substantial portions of the Software.
  14. #
  15. # THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
  16. # IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
  17. # FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
  18. # AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
  19. # LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
  20. # OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
  21. # SOFTWARE.
  22. from __future__ import unicode_literals
  23. from . import tokens
  24. from ..compat import str
  25. from ..nodes import Heading, HTMLEntity, Tag, Template, Text
  26. from ..nodes.extras import Attribute, Parameter
  27. from ..smart_list import SmartList
  28. from ..wikicode import Wikicode
  29. __all__ = ["Builder"]
  30. class Builder(object):
  31. """Combines a sequence of tokens into a tree of ``Wikicode`` objects.
  32. To use, pass a list of :py:class:`~mwparserfromhell.parser.tokens.Token`\ s
  33. to the :py:meth:`build` method. The list will be exhausted as it is parsed
  34. and a :py:class:`~mwparserfromhell.wikicode.Wikicode` object will be
  35. returned.
  36. """
  37. def __init__(self):
  38. self._tokens = []
  39. self._stacks = []
  40. def _wrap(self, nodes):
  41. """Properly wrap a list of nodes in a ``Wikicode`` object."""
  42. return Wikicode(SmartList(nodes))
  43. def _push(self):
  44. """Push a new node list onto the stack."""
  45. self._stacks.append([])
  46. def _pop(self, wrap=True):
  47. """Pop the topmost node list off of the stack.
  48. If *wrap* is ``True``, we will call :py:meth:`_wrap` on the list.
  49. """
  50. if wrap:
  51. return self._wrap(self._stacks.pop())
  52. return self._stacks.pop()
  53. def _write(self, item):
  54. """Append a node to the topmost node list."""
  55. self._stacks[-1].append(item)
  56. def _handle_parameter(self, default):
  57. """Handle a case where a parameter is at the head of the tokens.
  58. *default* is the value to use if no parameter name is defined.
  59. """
  60. key = None
  61. showkey = False
  62. self._push()
  63. while self._tokens:
  64. token = self._tokens.pop()
  65. if isinstance(token, tokens.TemplateParamEquals):
  66. key = self._pop()
  67. showkey = True
  68. self._push()
  69. elif isinstance(token, (tokens.TemplateParamSeparator,
  70. tokens.TemplateClose)):
  71. self._tokens.append(token)
  72. value = self._pop()
  73. if not key:
  74. key = self._wrap([Text(str(default))])
  75. return Parameter(key, value, showkey)
  76. else:
  77. self._write(self._handle_token(token))
  78. def _handle_template(self):
  79. """Handle a case where a template is at the head of the tokens."""
  80. params = []
  81. default = 1
  82. self._push()
  83. while self._tokens:
  84. token = self._tokens.pop()
  85. if isinstance(token, tokens.TemplateParamSeparator):
  86. if not params:
  87. name = self._pop()
  88. param = self._handle_parameter(default)
  89. params.append(param)
  90. if not param.showkey:
  91. default += 1
  92. elif isinstance(token, tokens.TemplateClose):
  93. if not params:
  94. name = self._pop()
  95. return Template(name, params)
  96. else:
  97. self._write(self._handle_token(token))
  98. def _handle_entity(self):
  99. """Handle a case where a HTML entity is at the head of the tokens."""
  100. token = self._tokens.pop()
  101. if isinstance(token, tokens.HTMLEntityNumeric):
  102. token = self._tokens.pop()
  103. if isinstance(token, tokens.HTMLEntityHex):
  104. text = self._tokens.pop()
  105. self._tokens.pop() # Remove HTMLEntityEnd
  106. return HTMLEntity(text.text, named=False, hexadecimal=True,
  107. hex_char=token.char)
  108. self._tokens.pop() # Remove HTMLEntityEnd
  109. return HTMLEntity(token.text, named=False, hexadecimal=False)
  110. self._tokens.pop() # Remove HTMLEntityEnd
  111. return HTMLEntity(token.text, named=True, hexadecimal=False)
  112. def _handle_heading(self, token):
  113. """Handle a case where a heading is at the head of the tokens."""
  114. level = token.level
  115. self._push()
  116. while self._tokens:
  117. token = self._tokens.pop()
  118. if isinstance(token, tokens.HeadingEnd):
  119. title = self._pop()
  120. return Heading(title, level)
  121. else:
  122. self._write(self._handle_token(token))
  123. def _handle_attribute(self):
  124. """Handle a case where a tag attribute is at the head of the tokens."""
  125. name, quoted = None, False
  126. self._push()
  127. while self._tokens:
  128. token = self._tokens.pop()
  129. if isinstance(token, tokens.TagAttrEquals):
  130. name = self._pop()
  131. self._push()
  132. elif isinstance(token, tokens.TagAttrQuote):
  133. quoted = True
  134. elif isinstance(token, (tokens.TagAttrStart,
  135. tokens.TagCloseOpen)):
  136. self._tokens.append(token)
  137. if name is not None:
  138. return Attribute(name, self._pop(), quoted)
  139. return Attribute(self._pop(), quoted=quoted)
  140. else:
  141. self._write(self._handle_token(token))
  142. def _handle_tag(self, token):
  143. """Handle a case where a tag is at the head of the tokens."""
  144. type_, showtag = token.type, token.showtag
  145. attrs = []
  146. self._push()
  147. while self._tokens:
  148. token = self._tokens.pop()
  149. if isinstance(token, tokens.TagAttrStart):
  150. attrs.append(self._handle_attribute())
  151. elif isinstance(token, tokens.TagCloseOpen):
  152. open_pad = token.padding
  153. tag = self._pop()
  154. self._push()
  155. elif isinstance(token, tokens.TagCloseSelfclose):
  156. tag = self._pop()
  157. return Tag(type_, tag, attrs=attrs, showtag=showtag,
  158. self_closing=True, open_padding=token.padding)
  159. elif isinstance(token, tokens.TagOpenClose):
  160. contents = self._pop()
  161. elif isinstance(token, tokens.TagCloseClose):
  162. return Tag(type_, tag, contents, attrs, showtag, False,
  163. open_pad, token.padding)
  164. else:
  165. self._write(self._handle_token(token))
  166. def _handle_token(self, token):
  167. """Handle a single token."""
  168. if isinstance(token, tokens.Text):
  169. return Text(token.text)
  170. elif isinstance(token, tokens.TemplateOpen):
  171. return self._handle_template()
  172. elif isinstance(token, tokens.HTMLEntityStart):
  173. return self._handle_entity()
  174. elif isinstance(token, tokens.HeadingStart):
  175. return self._handle_heading(token)
  176. elif isinstance(token, tokens.TagOpenOpen):
  177. return self._handle_tag(token)
  178. def build(self, tokenlist):
  179. """Build a Wikicode object from a list tokens and return it."""
  180. self._tokens = tokenlist
  181. self._tokens.reverse()
  182. self._push()
  183. while self._tokens:
  184. node = self._handle_token(self._tokens.pop())
  185. self._write(node)
  186. return self._pop()