DEFINITIONS
This source file includes following functions.
1 #
2 # $Id: common.rb,v 1.3 2002/06/12 09:31:41 akira Exp $
3 #
4 # Copyright (c) 2001 akira yamada <akira@ruby-lang.org>
5 # You can redistribute it and/or modify it under the same term as Ruby.
6 #
7
8 =begin
9
10 == URI
11
12 =end
13
14 module URI
15 module REGEXP
16 module PATTERN
17 # RFC 2396 (URI Generic Syntax)
18 # RFC 2732 (IPv6 Literal Addresses in URL's)
19 # RFC 2373 (IPv6 Addressing Architecture)
20
21 # alpha = lowalpha | upalpha
22 ALPHA = "a-zA-Z"
23 # alphanum = alpha | digit
24 ALNUM = "#{ALPHA}\\d"
25
26 # hex = digit | "A" | "B" | "C" | "D" | "E" | "F" |
27 # "a" | "b" | "c" | "d" | "e" | "f"
28 HEX = "a-fA-F\\d"
29 # escaped = "%" hex hex
30 ESCAPED = "%[#{HEX}]{2}"
31 # mark = "-" | "_" | "." | "!" | "~" | "*" | "'" |
32 # "(" | ")"
33 # unreserved = alphanum | mark
34 UNRESERVED = "-_.!~*'()#{ALNUM}"
35 # reserved = ";" | "/" | "?" | ":" | "@" | "&" | "=" | "+" |
36 # "$" | ","
37 # reserved = ";" | "/" | "?" | ":" | "@" | "&" | "=" | "+" |
38 # "$" | "," | "[" | "]" (RFC 2732)
39 RESERVED = ";/?:@&=+$,\\[\\]"
40
41 # uric = reserved | unreserved | escaped
42 URIC = "(?:[#{UNRESERVED}#{RESERVED}]|#{ESCAPED})"
43 # uric_no_slash = unreserved | escaped | ";" | "?" | ":" | "@" |
44 # "&" | "=" | "+" | "$" | ","
45 URIC_NO_SLASH = "(?:[#{UNRESERVED};?:@&=+$,]|#{ESCAPED})"
46 # query = *uric
47 QUERY = "#{URIC}*"
48 # fragment = *uric
49 FRAGMENT = "#{URIC}*"
50
51 # domainlabel = alphanum | alphanum *( alphanum | "-" ) alphanum
52 DOMLABEL = "(?:[#{ALNUM}](?:[-#{ALNUM}]*[#{ALNUM}])?)"
53 # toplabel = alpha | alpha *( alphanum | "-" ) alphanum
54 TOPLABEL = "(?:[#{ALPHA}](?:[-#{ALNUM}]*[#{ALNUM}])?)"
55 # hostname = *( domainlabel "." ) toplabel [ "." ]
56 HOSTNAME = "(?:#{DOMLABEL}\\.)*#{TOPLABEL}\\.?"
57
58 # RFC 2373, APPENDIX B:
59 # IPv6address = hexpart [ ":" IPv4address ]
60 # IPv4address = 1*3DIGIT "." 1*3DIGIT "." 1*3DIGIT "." 1*3DIGIT
61 # hexpart = hexseq | hexseq "::" [ hexseq ] | "::" [ hexseq ]
62 # hexseq = hex4 *( ":" hex4)
63 # hex4 = 1*4HEXDIG
64 #
65 # XXX: This definition has a flaw. "::" + IPv4address must be
66 # allowed too. Here is a replacement.
67 #
68 # IPv4address = 1*3DIGIT "." 1*3DIGIT "." 1*3DIGIT "." 1*3DIGIT
69 IPV4ADDR = "\\d{1,3}\\.\\d{1,3}\\.\\d{1,3}\\.\\d{1,3}"
70 # hex4 = 1*4HEXDIG
71 HEX4 = "[#{HEX}]{1,4}"
72 # lastpart = hex4 | IPv4address
73 LASTPART = "(?:#{HEX4}|#{IPV4ADDR})"
74 # hexseq1 = *( hex4 ":" ) hex4
75 HEXSEQ1 = "(?:#{HEX4}:)*#{HEX4}"
76 # hexseq2 = *( hex4 ":" ) lastpart
77 HEXSEQ2 = "(?:#{HEX4}:)*#{LASTPART}"
78 # IPv6address = hexseq2 | [ hexseq1 ] "::" [ hexseq2 ]
79 IPV6ADDR = "(?:#{HEXSEQ2}|(?:#{HEXSEQ1})?::(?:#{HEXSEQ2})?)"
80
81 # IPv6prefix = ( hexseq1 | [ hexseq1 ] "::" [ hexseq1 ] ) "/" 1*2DIGIT
82 # unused
83
84 # ipv6reference = "[" IPv6address "]" (RFC 2732)
85 IPV6REF = "\\[#{IPV6ADDR}\\]"
86
87 # host = hostname | IPv4address
88 # host = hostname | IPv4address | IPv6reference (RFC 2732)
89 HOST = "(?:#{HOSTNAME}|#{IPV4ADDR}|#{IPV6REF})"
90 # port = *digit
91 PORT = "\d*"
92 # hostport = host [ ":" port ]
93 HOSTPORT = "#{HOST}(?:#{PORT})?"
94
95 # userinfo = *( unreserved | escaped |
96 # ";" | ":" | "&" | "=" | "+" | "$" | "," )
97 USERINFO = "(?:[#{UNRESERVED};:&=+$,]|#{ESCAPED})*"
98
99 # pchar = unreserved | escaped |
100 # ":" | "@" | "&" | "=" | "+" | "$" | ","
101 PCHAR = "(?:[#{UNRESERVED}:@&=+$,]|#{ESCAPED})"
102 # param = *pchar
103 PARAM = "#{PCHAR}*"
104 # segment = *pchar *( ";" param )
105 SEGMENT = "#{PCHAR}*(?:;#{PARAM})*"
106 # path_segments = segment *( "/" segment )
107 PATH_SEGMENTS = "#{SEGMENT}(?:/#{SEGMENT})*"
108
109 # server = [ [ userinfo "@" ] hostport ]
110 SERVER = "(?:#{USERINFO}@)?#{HOSTPORT}"
111 # reg_name = 1*( unreserved | escaped | "$" | "," |
112 # ";" | ":" | "@" | "&" | "=" | "+" )
113 REG_NAME = "(?:[#{UNRESERVED}$,;+@&=+]|#{ESCAPED})+"
114 # authority = server | reg_name
115 AUTHORITY = "(?:#{SERVER}|#{REG_NAME})"
116
117 # rel_segment = 1*( unreserved | escaped |
118 # ";" | "@" | "&" | "=" | "+" | "$" | "," )
119 REL_SEGMENT = "(?:[#{UNRESERVED};@&=+$,]|#{ESCAPED})+"
120
121 # scheme = alpha *( alpha | digit | "+" | "-" | "." )
122 SCHEME = "[#{ALPHA}][-+.#{ALPHA}\\d]*"
123
124 # abs_path = "/" path_segments
125 ABS_PATH = "/#{PATH_SEGMENTS}"
126 # rel_path = rel_segment [ abs_path ]
127 REL_PATH = "#{REL_SEGMENT}(?:#{ABS_PATH})?"
128 # net_path = "//" authority [ abs_path ]
129 NET_PATH = "//#{AUTHORITY}(?:#{ABS_PATH})?"
130
131 # hier_part = ( net_path | abs_path ) [ "?" query ]
132 HIER_PART = "(?:#{NET_PATH}|#{ABS_PATH})(?:\\?(?:#{QUERY}))?"
133 # opaque_part = uric_no_slash *uric
134 OPAQUE_PART = "#{URIC_NO_SLASH}#{URIC}*"
135
136 # absoluteURI = scheme ":" ( hier_part | opaque_part )
137 ABS_URI = "#{SCHEME}:(?:#{HIER_PART}|#{OPAQUE_PART})"
138 # relativeURI = ( net_path | abs_path | rel_path ) [ "?" query ]
139 REL_URI = "(?:#{NET_PATH}|#{ABS_PATH}|#{REL_PATH})(?:\\?#{QUERY})?"
140
141 # URI-reference = [ absoluteURI | relativeURI ] [ "#" fragment ]
142 URI_REF = "(?:#{ABS_URI}|#{REL_URI})?(?:##{FRAGMENT})?"
143
144 # XXX:
145 X_ABS_URI = "
146 (#{PATTERN::SCHEME}): (?# 1: scheme)
147 (?:
148 (?:(?:
149
150 (?:(?:(#{PATTERN::USERINFO})@)? (?# 2: userinfo)
151 (?:(#{PATTERN::HOST})(?::(\\d*))?))?(?# 3: host, 4: port)
152 |
153 (#{PATTERN::REG_NAME}) (?# 5: registry)
154 ))?
155 ((?!
156 )(?:\\?(#{PATTERN::QUERY}))? (?# 7: query)
157 |
158 (#{PATTERN::OPAQUE_PART}) (?# 8: opaque)
159 )
160 (?:\\#(#{PATTERN::FRAGMENT}))? (?# 9: fragment)
161 "
162 X_REL_URI = "
163 (?:
164 (?:
165
166 (?:
167 (?:(#{PATTERN::USERINFO})@)? (?# 1: userinfo)
168 (#{PATTERN::HOST})?(?::(\\d*))? (?# 2: host, 3: port)
169 |
170 (#{PATTERN::REG_NAME}) (?# 4: registry)
171 )
172 )
173 |
174 (#{PATTERN::REL_SEGMENT}) (?# 5: rel_segment)
175 )?
176 (#{PATTERN::ABS_PATH})? (?# 6: abs_path)
177 (?:\\?(#{PATTERN::QUERY}))? (?# 7: query)
178 (?:\\#(#{PATTERN::FRAGMENT}))? (?# 8: fragment)
179 "
180 end # PATTERN
181
182 # for URI::split
183 ABS_URI = Regexp.new('^' + PATTERN::X_ABS_URI + '$', #'
184 Regexp::EXTENDED, 'N').freeze
185 REL_URI = Regexp.new('^' + PATTERN::X_REL_URI + '$', #'
186 Regexp::EXTENDED, 'N').freeze
187
188 # for URI::extract
189 URI_REF = Regexp.new(PATTERN::URI_REF, false, 'N').freeze
190 ABS_URI_REF = Regexp.new(PATTERN::X_ABS_URI, Regexp::EXTENDED, 'N').freeze
191 REL_URI_REF = Regexp.new(PATTERN::X_REL_URI, Regexp::EXTENDED, 'N').freeze
192
193 # for URI::escape/unescape
194 ESCAPED = Regexp.new(PATTERN::ESCAPED, false, 'N').freeze
195 UNSAFE = Regexp.new("[^#{PATTERN::UNRESERVED}#{PATTERN::RESERVED}]",
196 false, 'N').freeze
197
198 # for Generic#initialize
199 SCHEME = Regexp.new("^#{PATTERN::SCHEME}$", false, 'N').freeze #"
200 USERINFO = Regexp.new("^#{PATTERN::USERINFO}$", false, 'N').freeze #"
201 HOST = Regexp.new("^#{PATTERN::HOST}$", false, 'N').freeze #"
202 PORT = Regexp.new("^#{PATTERN::PORT}$", false, 'N').freeze #"
203 OPAQUE = Regexp.new("^#{PATTERN::OPAQUE_PART}$", false, 'N').freeze #"
204 REGISTRY = Regexp.new("^#{PATTERN::REG_NAME}$", false, 'N').freeze #"
205 ABS_PATH = Regexp.new("^#{PATTERN::ABS_PATH}$", false, 'N').freeze #"
206 REL_PATH = Regexp.new("^#{PATTERN::REL_PATH}$", false, 'N').freeze #"
207 QUERY = Regexp.new("^#{PATTERN::QUERY}$", false, 'N').freeze #"
208 FRAGMENT = Regexp.new("^#{PATTERN::FRAGMENT}$", false, 'N').freeze #"
209 end # REGEXP
210
211 module Util
212 def make_components_hash(klass, array_hash)
213 tmp = {}
214 if array_hash.kind_of?(Array) &&
215 array_hash.size == klass.component.size - 1
216 klass.component[1..-1].each_index do |i|
217 begin
218 tmp[klass.component[i + 1]] = array_hash[i].clone
219 rescue TypeError
220 tmp[klass.component[i + 1]] = array_hash[i]
221 end
222 end
223
224 elsif array_hash.kind_of?(Hash)
225 array_hash.each do |key, value|
226 begin
227 tmp[key] = value.clone
228 rescue TypeError
229 tmp[key] = value
230 end
231 end
232 else
233 raise ArgumentError,
234 "expected Array of or Hash of components of #{klass.to_s} (#{klass.component[1..-1].join(', ')})"
235 end
236 tmp[:scheme] = klass.to_s.sub(/\A.*::/, '').downcase
237
238 return tmp
239 end
240 module_function :make_components_hash
241 end
242
243 module Escape
244 include REGEXP
245
246 def escape(str, unsafe = UNSAFE)
247 unless unsafe.kind_of?(Regexp)
248 # perhaps unsafe is String object
249 unsafe = Regexp.new(Regexp.quote(unsafe), false, 'N')
250 end
251 str.gsub(unsafe) do |us|
252 tmp = ''
253 us.each_byte do |uc|
254 tmp << sprintf('%%%02X', uc)
255 end
256 tmp
257 end
258 end
259 alias encode escape
260
261 def unescape(str)
262 str.gsub(ESCAPED) do
263 $&[1,2].hex.chr
264 end
265 end
266 alias decode unescape
267 end
268
269 include REGEXP
270 extend Escape
271
272 @@schemes = {}
273
274 class Error < StandardError; end
275 class InvalidURIError < Error; end # it is not URI.
276 class InvalidComponentError < Error; end # it is not component of URI.
277 class BadURIError < Error; end # the URI is valid but it is bad for the position.
278
279 =begin
280
281 === Methods
282
283 --- URI::split(uri)
284
285 =end
286
287 def self.split(uri)
288 case uri
289 when ''
290 # null uri
291
292 when ABS_URI
293 scheme, userinfo, host, port,
294 registry, path, query, opaque, fragment = $~[1..-1]
295
296 # URI-reference = [ absoluteURI | relativeURI ] [ "#" fragment ]
297
298 # absoluteURI = scheme ":" ( hier_part | opaque_part )
299 # hier_part = ( net_path | abs_path ) [ "?" query ]
300 # opaque_part = uric_no_slash *uric
301
302 # abs_path = "/" path_segments
303 # net_path = "//" authority [ abs_path ]
304
305 # authority = server | reg_name
306 # server = [ [ userinfo "@" ] hostport ]
307
308 if !scheme
309 raise InvalidURIError,
310 "bad URI(absolute but no scheme): #{uri}"
311 end
312 if !opaque && (!path && (!host && !registry))
313 raise InvalidURIError,
314 "bad URI(absolute but no path): #{uri}"
315 end
316
317 when REL_URI
318 scheme = nil
319 opaque = nil
320
321 userinfo, host, port, registry,
322 rel_segment, abs_path, query, fragment = $~[1..-1]
323 if rel_segment && abs_path
324 path = rel_segment + abs_path
325 elsif rel_segment
326 path = rel_segment
327 elsif abs_path
328 path = abs_path
329 end
330
331 # URI-reference = [ absoluteURI | relativeURI ] [ "#" fragment ]
332
333 # relativeURI = ( net_path | abs_path | rel_path ) [ "?" query ]
334
335 # net_path = "//" authority [ abs_path ]
336 # abs_path = "/" path_segments
337 # rel_path = rel_segment [ abs_path ]
338
339 # authority = server | reg_name
340 # server = [ [ userinfo "@" ] hostport ]
341
342 else
343 raise InvalidURIError, "bad URI(is not URI?): #{uri}"
344 end
345
346 path = '' if !path && !opaque # (see RFC2396 Section 5.2)
347 ret = [
348 scheme,
349 userinfo, host, port, # X
350 registry, # X
351 path, # Y
352 opaque, # Y
353 query,
354 fragment
355 ]
356 return ret
357 end
358
359 =begin
360
361 --- URI::parse(uri_str)
362
363 =end
364 def self.parse(uri)
365 scheme, userinfo, host, port,
366 registry, path, opaque, query, fragment = self.split(uri)
367
368 if scheme && @@schemes.include?(scheme.upcase)
369 @@schemes[scheme.upcase].new(scheme, userinfo, host, port,
370 registry, path, opaque, query,
371 fragment)
372 else
373 Generic.new(scheme, userinfo, host, port,
374 registry, path, opaque, query,
375 fragment)
376 end
377 end
378
379 =begin
380
381 --- URI::join(str[, str, ...])
382
383 =end
384 def self.join(*str)
385 u = self.parse(str[0])
386 str[1 .. -1].each do |x|
387 u = u.merge(x)
388 end
389 u
390 end
391
392 =begin
393
394 --- URI::extract(str[, schemes])
395
396 =end
397 def self.extract(str, schemes = [])
398 urls = []
399 if schemes.size > 0
400 tmp = Regexp.new('(?:' + schemes.collect{|s|
401 Regexp.quote(s + ':')
402 }.join('|') + ')',
403 Regexp::IGNORECASE, 'N')
404 str.scan(tmp) {
405 tmp_str = $& + $'
406 if ABS_URI_REF =~ tmp_str
407 if block_given?
408 yield($&)
409 else
410 urls << $&
411 end
412 end
413 }
414
415 else
416 str.scan(ABS_URI_REF) {
417 if block_given?
418 yield($&)
419 else
420 urls << $&
421 end
422 }
423 end
424
425 if block_given?
426 return nil
427 else
428 return urls
429 end
430 end
431
432 end # URI