A Python parser for MediaWiki wikicode https://mwparserfromhell.readthedocs.io/
25'ten fazla konu seçemezsiniz Konular bir harf veya rakamla başlamalı, kısa çizgiler ('-') içerebilir ve en fazla 35 karakter uzunluğunda olabilir.
 
 
 
 

170 satır
6.4 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 _push(self):
  34. self._stacks.append([])
  35. def _pop(self):
  36. return Wikicode(SmartList(self._stacks.pop()))
  37. def _write(self, item):
  38. self._stacks[-1].append(item)
  39. def _handle_parameter(self, key):
  40. showkey = False
  41. self._push()
  42. while self._tokens:
  43. token = self._tokens.pop(0)
  44. if isinstance(token, tokens.TemplateParamEquals):
  45. key = self._pop()
  46. showkey = True
  47. self._push()
  48. elif isinstance(token, (tokens.TemplateParamSeparator,
  49. tokens.TemplateClose)):
  50. self._tokens.insert(0, token)
  51. value = self._pop()
  52. return Parameter(key, value, showkey)
  53. else:
  54. self._write(self._handle_token())
  55. def _handle_template(self):
  56. params = []
  57. int_keys = set()
  58. int_key_range = {1}
  59. self._push()
  60. while self._tokens:
  61. token = self._tokens.pop(0)
  62. if isinstance(token, tokens.TemplateParamSeparator):
  63. if not params:
  64. name = self._pop()
  65. param = self._handle_parameter(min(int_key_range - int_keys))
  66. if re.match(r"[1-9][0-9]*$", param.name.strip()):
  67. int_keys.add(int(param.name))
  68. int_key_range.add(len(int_keys) + 1)
  69. params.append(param)
  70. elif isinstance(token, tokens.TemplateClose):
  71. if not params:
  72. name = self._pop()
  73. return Template(name, params)
  74. else:
  75. self._write(self._handle_token())
  76. def _handle_entity(self):
  77. token = self._tokens.pop(0)
  78. if isinstance(token, tokens.HTMLEntityNumeric):
  79. token = self._tokens.pop(0)
  80. if isinstance(token, tokens.HTMLEntityHex):
  81. text = self._tokens.pop(0)
  82. return HTMLEntity(text.text, named=False, hexadecimal=True,
  83. hex_char=token.char)
  84. return HTMLEntity(token.text, named=False, hexadecimal=False)
  85. return HTMLEntity(token.text, named=True, hexadecimal=False)
  86. def _handle_heading(self, token):
  87. level = token.level
  88. self._push()
  89. while self._tokens:
  90. token = self._tokens.pop(0)
  91. if isinstance(token, tokens.HeadingBlock):
  92. title = self._pop()
  93. return Heading(title, level)
  94. else:
  95. self._write(self._handle_token())
  96. def _handle_attribute(self):
  97. name, quoted = None, False
  98. self._push()
  99. while self._tokens:
  100. token = self._tokens.pop(0)
  101. if isinstance(token, tokens.TagAttrEquals):
  102. name = self._pop()
  103. self._push()
  104. elif isinstance(token, tokens.TagAttrQuote):
  105. quoted = True
  106. elif isinstance(token, (tokens.TagAttrStart,
  107. tokens.TagCloseOpen)):
  108. self._tokens.insert(0, token)
  109. if name is not None:
  110. return Attribute(name, self._pop(), quoted)
  111. return Attribute(self._pop(), quoted=quoted)
  112. else:
  113. self._write(self._handle_token())
  114. def _handle_tag(self, token):
  115. type_, showtag = token.type, token.showtag
  116. attrs = []
  117. self._push()
  118. while self._tokens:
  119. token = self._tokens.pop(0)
  120. if isinstance(token, tokens.TagAttrStart):
  121. attrs.append(self._handle_attribute())
  122. elif isinstance(token, tokens.TagCloseOpen):
  123. open_pad = token.padding
  124. tag = self._pop()
  125. self._push()
  126. elif isinstance(token, tokens.TagCloseSelfclose):
  127. tag = self._pop()
  128. return Tag(type_, tag, attrs=attrs, showtag=showtag,
  129. self_closing=True, open_padding=token.padding)
  130. elif isinstance(token, tokens.TagOpenClose):
  131. contents = self._pop()
  132. elif isinstance(token, tokens.TagCloseClose):
  133. return Tag(type_, tag, contents, attrs, showtag, False,
  134. open_pad, token.padding)
  135. else:
  136. self._write(self._handle_token())
  137. def _handle_token(self):
  138. token = self._tokens.pop(0)
  139. if isinstance(token, tokens.Text):
  140. return Text(token.text)
  141. elif isinstance(token, tokens.TemplateOpen):
  142. return self._handle_template()
  143. elif isinstance(token, tokens.HTMLEntityStart):
  144. return self._handle_entity()
  145. elif isinstance(token, tokens.HeadingBlock):
  146. return self._handle_heading(token)
  147. elif isinstance(token, tokens.TagOpenOpen):
  148. return self._handle_tag(token)
  149. def build(self, tokenlist):
  150. self._tokens = tokenlist
  151. self._push()
  152. while self._tokens:
  153. self._write(self._handle_token())
  154. return self._pop()