Hide keyboard shortcuts

Hot-keys on this page

r m x p   toggle line displays

j k   next/prev highlighted chunk

0   (zero) top of page

1   (one) first highlighted chunk

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

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

222

223

224

225

226

227

228

229

230

from __future__ import absolute_import 

from collections import namedtuple 

 

from ..exceptions import LocationParseError 

 

 

url_attrs = ['scheme', 'auth', 'host', 'port', 'path', 'query', 'fragment'] 

 

# We only want to normalize urls with an HTTP(S) scheme. 

# urllib3 infers URLs without a scheme (None) to be http. 

NORMALIZABLE_SCHEMES = ('http', 'https', None) 

 

 

class Url(namedtuple('Url', url_attrs)): 

""" 

Datastructure for representing an HTTP URL. Used as a return value for 

:func:`parse_url`. Both the scheme and host are normalized as they are 

both case-insensitive according to RFC 3986. 

""" 

__slots__ = () 

 

def __new__(cls, scheme=None, auth=None, host=None, port=None, path=None, 

query=None, fragment=None): 

if path and not path.startswith('/'): 

path = '/' + path 

if scheme: 

scheme = scheme.lower() 

if host and scheme in NORMALIZABLE_SCHEMES: 

host = host.lower() 

return super(Url, cls).__new__(cls, scheme, auth, host, port, path, 

query, fragment) 

 

@property 

def hostname(self): 

"""For backwards-compatibility with urlparse. We're nice like that.""" 

return self.host 

 

@property 

def request_uri(self): 

"""Absolute path including the query string.""" 

uri = self.path or '/' 

 

if self.query is not None: 

uri += '?' + self.query 

 

return uri 

 

@property 

def netloc(self): 

"""Network location including host and port""" 

if self.port: 

return '%s:%d' % (self.host, self.port) 

return self.host 

 

@property 

def url(self): 

""" 

Convert self into a url 

 

This function should more or less round-trip with :func:`.parse_url`. The 

returned url may not be exactly the same as the url inputted to 

:func:`.parse_url`, but it should be equivalent by the RFC (e.g., urls 

with a blank port will have : removed). 

 

Example: :: 

 

>>> U = parse_url('http://google.com/mail/') 

>>> U.url 

'http://google.com/mail/' 

>>> Url('http', 'username:password', 'host.com', 80, 

... '/path', 'query', 'fragment').url 

'http://username:password@host.com:80/path?query#fragment' 

""" 

scheme, auth, host, port, path, query, fragment = self 

url = '' 

 

# We use "is not None" we want things to happen with empty strings (or 0 port) 

if scheme is not None: 

url += scheme + '://' 

if auth is not None: 

url += auth + '@' 

if host is not None: 

url += host 

if port is not None: 

url += ':' + str(port) 

if path is not None: 

url += path 

if query is not None: 

url += '?' + query 

if fragment is not None: 

url += '#' + fragment 

 

return url 

 

def __str__(self): 

return self.url 

 

 

def split_first(s, delims): 

""" 

Given a string and an iterable of delimiters, split on the first found 

delimiter. Return two split parts and the matched delimiter. 

 

If not found, then the first part is the full input string. 

 

Example:: 

 

>>> split_first('foo/bar?baz', '?/=') 

('foo', 'bar?baz', '/') 

>>> split_first('foo/bar?baz', '123') 

('foo/bar?baz', '', None) 

 

Scales linearly with number of delims. Not ideal for large number of delims. 

""" 

min_idx = None 

min_delim = None 

for d in delims: 

idx = s.find(d) 

if idx < 0: 

continue 

 

if min_idx is None or idx < min_idx: 

min_idx = idx 

min_delim = d 

 

if min_idx is None or min_idx < 0: 

return s, '', None 

 

return s[:min_idx], s[min_idx + 1:], min_delim 

 

 

def parse_url(url): 

""" 

Given a url, return a parsed :class:`.Url` namedtuple. Best-effort is 

performed to parse incomplete urls. Fields not provided will be None. 

 

Partly backwards-compatible with :mod:`urlparse`. 

 

Example:: 

 

>>> parse_url('http://google.com/mail/') 

Url(scheme='http', host='google.com', port=None, path='/mail/', ...) 

>>> parse_url('google.com:80') 

Url(scheme=None, host='google.com', port=80, path=None, ...) 

>>> parse_url('/foo?bar') 

Url(scheme=None, host=None, port=None, path='/foo', query='bar', ...) 

""" 

 

# While this code has overlap with stdlib's urlparse, it is much 

# simplified for our needs and less annoying. 

# Additionally, this implementations does silly things to be optimal 

# on CPython. 

 

if not url: 

# Empty 

return Url() 

 

scheme = None 

auth = None 

host = None 

port = None 

path = None 

fragment = None 

query = None 

 

# Scheme 

if '://' in url: 

scheme, url = url.split('://', 1) 

 

# Find the earliest Authority Terminator 

# (http://tools.ietf.org/html/rfc3986#section-3.2) 

url, path_, delim = split_first(url, ['/', '?', '#']) 

 

if delim: 

# Reassemble the path 

path = delim + path_ 

 

# Auth 

if '@' in url: 

# Last '@' denotes end of auth part 

auth, url = url.rsplit('@', 1) 

 

# IPv6 

if url and url[0] == '[': 

host, url = url.split(']', 1) 

host += ']' 

 

# Port 

if ':' in url: 

_host, port = url.split(':', 1) 

 

if not host: 

host = _host 

 

if port: 

# If given, ports must be integers. No whitespace, no plus or 

# minus prefixes, no non-integer digits such as ^2 (superscript). 

if not port.isdigit(): 

raise LocationParseError(url) 

try: 

port = int(port) 

except ValueError: 

raise LocationParseError(url) 

else: 

# Blank ports are cool, too. (rfc3986#section-3.2.3) 

port = None 

 

elif not host and url: 

host = url 

 

if not path: 

return Url(scheme, auth, host, port, path, query, fragment) 

 

# Fragment 

if '#' in path: 

path, fragment = path.split('#', 1) 

 

# Query 

if '?' in path: 

path, query = path.split('?', 1) 

 

return Url(scheme, auth, host, port, path, query, fragment) 

 

 

def get_host(url): 

""" 

Deprecated. Use :func:`parse_url` instead. 

""" 

p = parse_url(url) 

return p.scheme or 'http', p.hostname, p.port