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.
 
 
 
 

179 lines
6.9 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. import re
  23. from . import tokens
  24. from ..nodes import Heading, HTMLEntity, Tag, Template, Text
  25. from ..nodes.extras import Attribute, Parameter
  26. from ..smart_list import SmartList
  27. from ..wikicode import Wikicode
  28. __all__ = ["Builder"]
  29. class Builder(object):
  30. def __init__(self):
  31. self._tokens = []
  32. self._stacks = []
  33. def _wrap(self, nodes):
  34. return Wikicode(SmartList(nodes))
  35. def _push(self):
  36. self._stacks.append([])
  37. def _pop(self, wrap=True):
  38. if wrap:
  39. return self._wrap(self._stacks.pop())
  40. return self._stacks.pop()
  41. def _write(self, item):
  42. self._stacks[-1].append(item)
  43. def _handle_parameter(self, key):
  44. showkey = False
  45. self._push()
  46. while self._tokens:
  47. token = self._tokens.pop(0)
  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.insert(0, token)
  55. value = self._pop()
  56. return Parameter(key, value, showkey)
  57. else:
  58. self._write(self._handle_token(token))
  59. def _handle_template(self):
  60. params = []
  61. int_keys = set()
  62. int_key_range = {1}
  63. self._push()
  64. while self._tokens:
  65. token = self._tokens.pop(0)
  66. if isinstance(token, tokens.TemplateParamSeparator):
  67. if not params:
  68. name = self._pop()
  69. default = unicode(min(int_key_range - int_keys))
  70. param = self._handle_parameter(self._wrap([Text(default)]))
  71. if re.match(r"[1-9][0-9]*$", param.name.strip()):
  72. # We try a more restrictive test for integers than
  73. # try: int(), because "01" as a key will pass through int()
  74. # correctly but is not a valid integer key in wikicode:
  75. int_keys.add(int(unicode(param.name)))
  76. int_key_range.add(len(int_keys) + 1)
  77. params.append(param)
  78. elif isinstance(token, tokens.TemplateClose):
  79. if not params:
  80. name = self._pop()
  81. return Template(name, params)
  82. else:
  83. self._write(self._handle_token(token))
  84. def _handle_entity(self):
  85. token = self._tokens.pop(0)
  86. if isinstance(token, tokens.HTMLEntityNumeric):
  87. token = self._tokens.pop(0)
  88. if isinstance(token, tokens.HTMLEntityHex):
  89. text = self._tokens.pop(0)
  90. return HTMLEntity(text.text, named=False, hexadecimal=True,
  91. hex_char=token.char)
  92. return HTMLEntity(token.text, named=False, hexadecimal=False)
  93. return HTMLEntity(token.text, named=True, hexadecimal=False)
  94. def _handle_heading(self, token):
  95. level = token.level
  96. self._push()
  97. while self._tokens:
  98. token = self._tokens.pop(0)
  99. if isinstance(token, tokens.HeadingBlock):
  100. title = self._pop()
  101. return Heading(title, level)
  102. else:
  103. self._write(self._handle_token(token))
  104. def _handle_attribute(self):
  105. name, quoted = None, False
  106. self._push()
  107. while self._tokens:
  108. token = self._tokens.pop(0)
  109. if isinstance(token, tokens.TagAttrEquals):
  110. name = self._pop()
  111. self._push()
  112. elif isinstance(token, tokens.TagAttrQuote):
  113. quoted = True
  114. elif isinstance(token, (tokens.TagAttrStart,
  115. tokens.TagCloseOpen)):
  116. self._tokens.insert(0, token)
  117. if name is not None:
  118. return Attribute(name, self._pop(), quoted)
  119. return Attribute(self._pop(), quoted=quoted)
  120. else:
  121. self._write(self._handle_token(token))
  122. def _handle_tag(self, token):
  123. type_, showtag = token.type, token.showtag
  124. attrs = []
  125. self._push()
  126. while self._tokens:
  127. token = self._tokens.pop(0)
  128. if isinstance(token, tokens.TagAttrStart):
  129. attrs.append(self._handle_attribute())
  130. elif isinstance(token, tokens.TagCloseOpen):
  131. open_pad = token.padding
  132. tag = self._pop()
  133. self._push()
  134. elif isinstance(token, tokens.TagCloseSelfclose):
  135. tag = self._pop()
  136. return Tag(type_, tag, attrs=attrs, showtag=showtag,
  137. self_closing=True, open_padding=token.padding)
  138. elif isinstance(token, tokens.TagOpenClose):
  139. contents = self._pop()
  140. elif isinstance(token, tokens.TagCloseClose):
  141. return Tag(type_, tag, contents, attrs, showtag, False,
  142. open_pad, token.padding)
  143. else:
  144. self._write(self._handle_token(token))
  145. def _handle_token(self, token):
  146. if isinstance(token, tokens.Text):
  147. return Text(token.text)
  148. elif isinstance(token, tokens.TemplateOpen):
  149. return self._handle_template()
  150. elif isinstance(token, tokens.HTMLEntityStart):
  151. return self._handle_entity()
  152. elif isinstance(token, tokens.HeadingBlock):
  153. return self._handle_heading(token)
  154. elif isinstance(token, tokens.TagOpenOpen):
  155. return self._handle_tag(token)
  156. def build(self, tokenlist):
  157. self._tokens = tokenlist
  158. self._push()
  159. while self._tokens:
  160. node = self._handle_token(self._tokens.pop(0))
  161. self._write(node)
  162. return self._pop()