A Python parser for MediaWiki wikicode https://mwparserfromhell.readthedocs.io/
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 

176 lines
6.7 KiB

  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 . import tokens
  23. from ..nodes import Heading, HTMLEntity, Tag, Template, Text
  24. from ..nodes.extras import Attribute, Parameter
  25. from ..smart_list import SmartList
  26. from ..wikicode import Wikicode
  27. __all__ = ["Builder"]
  28. class Builder(object):
  29. def __init__(self):
  30. self._tokens = []
  31. self._stacks = []
  32. def _wrap(self, nodes):
  33. return Wikicode(SmartList(nodes))
  34. def _push(self):
  35. self._stacks.append([])
  36. def _pop(self, wrap=True):
  37. if wrap:
  38. return self._wrap(self._stacks.pop())
  39. return self._stacks.pop()
  40. def _write(self, item):
  41. self._stacks[-1].append(item)
  42. def _handle_parameter(self, key):
  43. showkey = False
  44. self._push()
  45. while self._tokens:
  46. token = self._tokens.pop(0)
  47. if isinstance(token, tokens.TemplateParamEquals):
  48. key = self._pop()
  49. showkey = True
  50. self._push()
  51. elif isinstance(token, (tokens.TemplateParamSeparator,
  52. tokens.TemplateClose)):
  53. self._tokens.insert(0, token)
  54. value = self._pop()
  55. return Parameter(key, value, showkey)
  56. else:
  57. self._write(self._handle_token(token))
  58. def _handle_template(self):
  59. params = []
  60. int_keys = set()
  61. int_key_range = {1}
  62. self._push()
  63. while self._tokens:
  64. token = self._tokens.pop(0)
  65. if isinstance(token, tokens.TemplateParamSeparator):
  66. if not params:
  67. name = self._pop()
  68. default = self._wrap(unicode(min(int_key_range - int_keys)))
  69. param = self._handle_parameter(default)
  70. try:
  71. int_keys.add(int(unicode(param.name)))
  72. int_key_range.add(len(int_keys) + 1)
  73. except ValueError:
  74. pass
  75. params.append(param)
  76. elif isinstance(token, tokens.TemplateClose):
  77. if not params:
  78. name = self._pop()
  79. return Template(name, params)
  80. else:
  81. self._write(self._handle_token(token))
  82. def _handle_entity(self):
  83. token = self._tokens.pop(0)
  84. if isinstance(token, tokens.HTMLEntityNumeric):
  85. token = self._tokens.pop(0)
  86. if isinstance(token, tokens.HTMLEntityHex):
  87. text = self._tokens.pop(0)
  88. return HTMLEntity(text.text, named=False, hexadecimal=True,
  89. hex_char=token.char)
  90. return HTMLEntity(token.text, named=False, hexadecimal=False)
  91. return HTMLEntity(token.text, named=True, hexadecimal=False)
  92. def _handle_heading(self, token):
  93. level = token.level
  94. self._push()
  95. while self._tokens:
  96. token = self._tokens.pop(0)
  97. if isinstance(token, tokens.HeadingBlock):
  98. title = self._pop()
  99. return Heading(title, level)
  100. else:
  101. self._write(self._handle_token(token))
  102. def _handle_attribute(self):
  103. name, quoted = None, False
  104. self._push()
  105. while self._tokens:
  106. token = self._tokens.pop(0)
  107. if isinstance(token, tokens.TagAttrEquals):
  108. name = self._pop()
  109. self._push()
  110. elif isinstance(token, tokens.TagAttrQuote):
  111. quoted = True
  112. elif isinstance(token, (tokens.TagAttrStart,
  113. tokens.TagCloseOpen)):
  114. self._tokens.insert(0, token)
  115. if name is not None:
  116. return Attribute(name, self._pop(), quoted)
  117. return Attribute(self._pop(), quoted=quoted)
  118. else:
  119. self._write(self._handle_token(token))
  120. def _handle_tag(self, token):
  121. type_, showtag = token.type, token.showtag
  122. attrs = []
  123. self._push()
  124. while self._tokens:
  125. token = self._tokens.pop(0)
  126. if isinstance(token, tokens.TagAttrStart):
  127. attrs.append(self._handle_attribute())
  128. elif isinstance(token, tokens.TagCloseOpen):
  129. open_pad = token.padding
  130. tag = self._pop()
  131. self._push()
  132. elif isinstance(token, tokens.TagCloseSelfclose):
  133. tag = self._pop()
  134. return Tag(type_, tag, attrs=attrs, showtag=showtag,
  135. self_closing=True, open_padding=token.padding)
  136. elif isinstance(token, tokens.TagOpenClose):
  137. contents = self._pop()
  138. elif isinstance(token, tokens.TagCloseClose):
  139. return Tag(type_, tag, contents, attrs, showtag, False,
  140. open_pad, token.padding)
  141. else:
  142. self._write(self._handle_token(token))
  143. def _handle_token(self, token):
  144. if isinstance(token, tokens.Text):
  145. return Text(token.text)
  146. elif isinstance(token, tokens.TemplateOpen):
  147. return self._handle_template()
  148. elif isinstance(token, tokens.HTMLEntityStart):
  149. return self._handle_entity()
  150. elif isinstance(token, tokens.HeadingBlock):
  151. return self._handle_heading(token)
  152. elif isinstance(token, tokens.TagOpenOpen):
  153. return self._handle_tag(token)
  154. def build(self, tokenlist):
  155. self._tokens = tokenlist
  156. self._push()
  157. while self._tokens:
  158. node = self._handle_token(self._tokens.pop(0))
  159. self._write(node)
  160. return self._pop()