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.
 
 
 
 

178 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, default):
  43. key = None
  44. showkey = False
  45. self._push()
  46. while self._tokens:
  47. token = self._tokens.pop()
  48. if isinstance(token, tokens.TemplateParamEquals):
  49. key = self._pop()
  50. showkey = True
  51. self._push()
  52. elif isinstance(token, (tokens.TemplateParamSeparator,
  53. tokens.TemplateClose)):
  54. self._tokens.append(token)
  55. value = self._pop()
  56. if not key:
  57. key = self._wrap([Text(unicode(default))])
  58. return Parameter(key, value, showkey)
  59. else:
  60. self._write(self._handle_token(token))
  61. def _handle_template(self):
  62. params = []
  63. default = 1
  64. self._push()
  65. while self._tokens:
  66. token = self._tokens.pop()
  67. if isinstance(token, tokens.TemplateParamSeparator):
  68. if not params:
  69. name = self._pop()
  70. param = self._handle_parameter(default)
  71. params.append(param)
  72. if not param.showkey:
  73. default += 1
  74. elif isinstance(token, tokens.TemplateClose):
  75. if not params:
  76. name = self._pop()
  77. return Template(name, params)
  78. else:
  79. self._write(self._handle_token(token))
  80. def _handle_entity(self):
  81. token = self._tokens.pop()
  82. if isinstance(token, tokens.HTMLEntityNumeric):
  83. token = self._tokens.pop()
  84. if isinstance(token, tokens.HTMLEntityHex):
  85. text = self._tokens.pop()
  86. self._tokens.pop() # Remove HTMLEntityEnd
  87. return HTMLEntity(text.text, named=False, hexadecimal=True,
  88. hex_char=token.char)
  89. self._tokens.pop() # Remove HTMLEntityEnd
  90. return HTMLEntity(token.text, named=False, hexadecimal=False)
  91. self._tokens.pop() # Remove HTMLEntityEnd
  92. return HTMLEntity(token.text, named=True, hexadecimal=False)
  93. def _handle_heading(self, token):
  94. level = token.level
  95. self._push()
  96. while self._tokens:
  97. token = self._tokens.pop()
  98. if isinstance(token, tokens.HeadingEnd):
  99. title = self._pop()
  100. return Heading(title, level)
  101. else:
  102. self._write(self._handle_token(token))
  103. def _handle_attribute(self):
  104. name, quoted = None, False
  105. self._push()
  106. while self._tokens:
  107. token = self._tokens.pop()
  108. if isinstance(token, tokens.TagAttrEquals):
  109. name = self._pop()
  110. self._push()
  111. elif isinstance(token, tokens.TagAttrQuote):
  112. quoted = True
  113. elif isinstance(token, (tokens.TagAttrStart,
  114. tokens.TagCloseOpen)):
  115. self._tokens.append(token)
  116. if name is not None:
  117. return Attribute(name, self._pop(), quoted)
  118. return Attribute(self._pop(), quoted=quoted)
  119. else:
  120. self._write(self._handle_token(token))
  121. def _handle_tag(self, token):
  122. type_, showtag = token.type, token.showtag
  123. attrs = []
  124. self._push()
  125. while self._tokens:
  126. token = self._tokens.pop()
  127. if isinstance(token, tokens.TagAttrStart):
  128. attrs.append(self._handle_attribute())
  129. elif isinstance(token, tokens.TagCloseOpen):
  130. open_pad = token.padding
  131. tag = self._pop()
  132. self._push()
  133. elif isinstance(token, tokens.TagCloseSelfclose):
  134. tag = self._pop()
  135. return Tag(type_, tag, attrs=attrs, showtag=showtag,
  136. self_closing=True, open_padding=token.padding)
  137. elif isinstance(token, tokens.TagOpenClose):
  138. contents = self._pop()
  139. elif isinstance(token, tokens.TagCloseClose):
  140. return Tag(type_, tag, contents, attrs, showtag, False,
  141. open_pad, token.padding)
  142. else:
  143. self._write(self._handle_token(token))
  144. def _handle_token(self, token):
  145. if isinstance(token, tokens.Text):
  146. return Text(token.text)
  147. elif isinstance(token, tokens.TemplateOpen):
  148. return self._handle_template()
  149. elif isinstance(token, tokens.HTMLEntityStart):
  150. return self._handle_entity()
  151. elif isinstance(token, tokens.HeadingStart):
  152. return self._handle_heading(token)
  153. elif isinstance(token, tokens.TagOpenOpen):
  154. return self._handle_tag(token)
  155. def build(self, tokenlist):
  156. self._tokens = tokenlist
  157. self._tokens.reverse()
  158. self._push()
  159. while self._tokens:
  160. node = self._handle_token(self._tokens.pop())
  161. self._write(node)
  162. return self._pop()