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.
 
 
 
 

176 рядки
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. from collections import defaultdict
  23. import re
  24. from . import HTMLEntity, Node, Text
  25. from .extras import Parameter
  26. from ..utils import parse_anything
  27. __all__ = ["Template"]
  28. FLAGS = re.DOTALL | re.UNICODE
  29. class Template(Node):
  30. def __init__(self, name, params=None):
  31. super(Template, self).__init__(self)
  32. self._name = name
  33. if params:
  34. self._params = params
  35. else:
  36. self._params = []
  37. def __unicode__(self):
  38. if self.params:
  39. params = u"|".join([unicode(param) for param in self.params])
  40. return "{{" + unicode(self.name) + "|" + params + "}}"
  41. else:
  42. return "{{" + unicode(self.name) + "}}"
  43. def __iternodes__(self, getter):
  44. yield None, self
  45. for child in getter(self.name):
  46. yield self.name, child
  47. for param in self.params:
  48. if param.showkey:
  49. for child in getter(param.name):
  50. yield param.name, child
  51. for child in getter(param.value):
  52. yield param.value, child
  53. def __showtree__(self, write, get, mark):
  54. write("{{")
  55. get(self.name)
  56. for param in self.params:
  57. write(" | ")
  58. mark()
  59. get(param.name)
  60. write(" = ")
  61. mark()
  62. get(param.value)
  63. write("}}")
  64. def _surface_escape(self, code, char):
  65. replacement = HTMLEntity(value=ord(char))
  66. for node in code.filter_text(recursive=False):
  67. if char in node:
  68. code.replace(node, node.replace(char, replacement))
  69. def _blank_param_value(self, value):
  70. match = re.search("^(\s*).*?(\s*)$", unicode(value), FLAGS)
  71. value.nodes = [Text(match.group(1)), Text(match.group(2))]
  72. def _select_theory(self, theories):
  73. if theories:
  74. best = max(theories.values())
  75. confidence = float(best) / sum(theories.values())
  76. if confidence > 0.75:
  77. return theories.keys()[theories.values().index(best)]
  78. def _get_spacing_conventions(self):
  79. before_theories = defaultdict(lambda: 0)
  80. after_theories = defaultdict(lambda: 0)
  81. for param in self.params:
  82. match = re.search("^(\s*).*?(\s*)$", unicode(param.value), FLAGS)
  83. before, after = match.group(1), match.group(2)
  84. before_theories[before] += 1
  85. after_theories[after] += 1
  86. before = self._select_theory(before_theories)
  87. after = self._select_theory(after_theories)
  88. return before, after
  89. @property
  90. def name(self):
  91. return self._name
  92. @property
  93. def params(self):
  94. return self._params
  95. def has_param(self, name, ignore_empty=True):
  96. name = name.strip() if isinstance(name, basestring) else unicode(name)
  97. for param in self.params:
  98. if param.name.strip() == name:
  99. if ignore_empty and not param.value.strip():
  100. continue
  101. return True
  102. return False
  103. def get(self, name):
  104. name = name.strip() if isinstance(name, basestring) else unicode(name)
  105. for param in self.params:
  106. if param.name.strip() == name:
  107. return param
  108. raise ValueError(name)
  109. def add(self, name, value, showkey=None, force_nonconformity=False):
  110. name, value = parse_anything(name), parse_anything(value)
  111. self._surface_escape(value, "|")
  112. if self.has_param(name):
  113. self.remove(name, keep_field=True)
  114. existing = self.get(name)
  115. if showkey is None: # Infer showkey from current value
  116. showkey = existing.showkey
  117. if not showkey:
  118. self._surface_escape(value, "=")
  119. nodes = existing.value.nodes
  120. if force_nonconformity:
  121. existing.value = value
  122. else:
  123. existing.value = parse_anything([nodes[0], value, nodes[1]])
  124. return existing
  125. if showkey is None:
  126. try:
  127. int(name)
  128. showkey = True
  129. except ValueError:
  130. showkey = False
  131. if not showkey:
  132. self._surface_escape(value, "=")
  133. if not force_nonconformity:
  134. before, after = self._get_spacing_conventions()
  135. if before and after:
  136. value = parse_anything([before, value, after])
  137. elif before:
  138. value = parse_anything([before, value])
  139. elif after:
  140. value = parse_anything([value, after])
  141. param = Parameter(name, value, showkey)
  142. self.params.append(param)
  143. return param
  144. def remove(self, name, keep_field=False, force_no_field=False):
  145. name = name.strip() if isinstance(name, basestring) else unicode(name)
  146. for i, param in enumerate(self.params):
  147. if param.name.strip() == name:
  148. if keep_field:
  149. return self._blank_param_value(param.value)
  150. dependent = [not after.showkey for after in self.params[i+1:]]
  151. if any(dependent) and not param.showkey and not force_no_field:
  152. return self._blank_param_value(param.value)
  153. return self.params.remove(param)
  154. raise ValueError(name)