html5lib-python/src/html5lib/serializer/htmlserializer.py at python3-old · moben/html5lib-python · GitHub

History

221 lines (194 loc) · 8.41 KB

Raw

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

try:

frozenset

except NameError:

# Import from the sets module for python 2.3

from sets import ImmutableSet as frozenset

import gettext

_ = gettext.gettext

from html5lib.constants import voidElements, booleanAttributes, spaceCharacters

from html5lib.constants import rcdataElements

from xml.sax.saxutils import escape

spaceCharacters = "".join(spaceCharacters)

try:

from codecs import register_error, xmlcharrefreplace_errors

except ImportError:

unicode_encode_errors = "strict"

else:

unicode_encode_errors = "htmlentityreplace"

from html5lib.constants import entities

encode_entity_map = {}

for k, v in list(entities.items()):

if v != "&" and encode_entity_map.get(v) != k.lower():

# prefer < over &LT; and similarly for &, >, etc.

encode_entity_map[v] = k

def htmlentityreplace_errors(exc):

if isinstance(exc, (UnicodeEncodeError, UnicodeTranslateError)):

res = []

for c in exc.object[exc.start:exc.end]:

e = encode_entity_map.get(c)

if e:

res.append("&")

res.append(e)

if not e.endswith(";"):

res.append(";")

else:

res.append(c.encode(exc.encoding, "xmlcharrefreplace"))

return ("".join(res), exc.end)

else:

return xmlcharrefreplace_errors(exc)

register_error(unicode_encode_errors, htmlentityreplace_errors)

del register_error

def encode(text, encoding):

return text.encode(encoding, unicode_encode_errors)

class HTMLSerializer(object):

quote_attr_values = False

quote_char = '"'

use_best_quote_char = True

minimize_boolean_attributes = True

use_trailing_solidus = False

space_before_trailing_solidus = True

escape_lt_in_attrs = False

escape_rcdata = False

inject_meta_charset = True

strip_whitespace = False

sanitize = False

omit_optional_tags = True

options = ("quote_attr_values", "quote_char", "use_best_quote_char",

"minimize_boolean_attributes", "use_trailing_solidus",

"space_before_trailing_solidus", "omit_optional_tags",

"strip_whitespace", "inject_meta_charset", "escape_lt_in_attrs",

"escape_rcdata", 'use_trailing_solidus', "sanitize")

def __init__(self, **kwargs):

if 'quote_char' in kwargs:

self.use_best_quote_char = False

for attr in self.options:

setattr(self, attr, kwargs.get(attr, getattr(self, attr)))

self.errors = []

self.strict = False

def serialize(self, treewalker, encoding=None):

in_cdata = False

self.errors = []

if encoding and self.inject_meta_charset:

from html5lib.filters.inject_meta_charset import Filter

treewalker = Filter(treewalker, encoding)

# XXX: WhitespaceFilter should be used before OptionalTagFilter

# for maximum efficiently of this latter filter

if self.strip_whitespace:

from html5lib.filters.whitespace import Filter

treewalker = Filter(treewalker)

if self.sanitize:

from html5lib.filters.sanitizer import Filter

treewalker = Filter(treewalker)

if self.omit_optional_tags:

from html5lib.filters.optionaltags import Filter

treewalker = Filter(treewalker)

for token in treewalker:

type = token["type"]

if type == "Doctype":

doctype = "<!DOCTYPE %s>" % token["name"]

if encoding:

yield doctype.encode(encoding)

else:

yield doctype

elif type in ("Characters", "SpaceCharacters"):

if type == "SpaceCharacters" or in_cdata:

if in_cdata and token["data"].find("</") >= 0:

self.serializeError(_("Unexpected </ in CDATA"))

if encoding:

yield token["data"].encode(encoding, "strict")

else:

yield token["data"]

elif encoding:

yield encode(escape(token["data"]), encoding)

else:

yield escape(token["data"])

elif type in ("StartTag", "EmptyTag"):

name = token["name"]

if name in rcdataElements and not self.escape_rcdata:

in_cdata = True

elif in_cdata:

self.serializeError(_("Unexpected child element of a CDATA element"))

attrs = token["data"]

if hasattr(attrs, "items"):

attrs = list(attrs.items())

attrs.sort()

attributes = []

for k,v in attrs:

if encoding:

k = k.encode(encoding, "strict")

attributes.append(' ')

attributes.append(k)

if (not self.minimize_boolean_attributes or

(k not in booleanAttributes.get(name, tuple())

and k not in booleanAttributes.get("", tuple()))):

attributes.append("=")

if self.quote_attr_values or not v:

quote_attr = True

else:

quote_attr = False

for char in spaceCharacters + ">\"'=":

if char in v:

quote_attr = True

break

v = v.replace("&", "&")

if self.escape_lt_in_attrs: v = v.replace("<", "<")

if encoding:

v = encode(v, encoding)

if quote_attr:

quote_char = self.quote_char

if self.use_best_quote_char:

if "'" in v and '"' not in v:

quote_char = '"'

elif '"' in v and "'" not in v:

quote_char = "'"

if quote_char == "'":

v = v.replace("'", "'")

else:

v = v.replace('"', """)

attributes.append(quote_char)

attributes.append(v)

attributes.append(quote_char)

else:

attributes.append(v)

if name in voidElements and self.use_trailing_solidus:

if self.space_before_trailing_solidus:

attributes.append(" /")

else:

attributes.append("/")

if encoding:

yield "<%s%s>" % (name.encode(encoding, "strict"), "".join(attributes))

else:

yield "<%s%s>" % (name, "".join(attributes))

elif type == "EndTag":

name = token["name"]

if name in rcdataElements:

in_cdata = False

elif in_cdata:

self.serializeError(_("Unexpected child element of a CDATA element"))

end_tag = "</%s>" % name

if encoding:

end_tag = end_tag.encode(encoding, "strict")

yield end_tag

elif type == "Comment":

data = token["data"]

if data.find("--") >= 0:

self.serializeError(_("Comment contains --"))

comment = "" % token["data"]

if encoding:

comment = comment.encode(encoding, unicode_encode_errors)

yield comment

else:

self.serializeError(token["data"])

def render(self, treewalker, encoding=None):

if encoding:

return "".join(list(self.serialize(treewalker, encoding)))

else:

return "".join(list(self.serialize(treewalker)))

def serializeError(self, data="XXX ERROR MESSAGE NEEDED"):

# XXX The idea is to make data mandatory.

self.errors.append(data)

if self.strict:

raise SerializeError

def SerializeError(Exception):

"""Error in serialized tree"""

pass

html5lib-python/src/html5lib/serializer/htmlserializer.py at python3-old · moben/html5lib-python · GitHub

Expand file tree

Breadcrumbs

htmlserializer.py

Latest commit

History

Breadcrumbs

htmlserializer.py

File metadata and controls