Debian Codegraph @

rfc1808.html

raw
1 <pre>Network Working Group R. Fielding
2 Request for Comments: 1808 UC Irvine
3 Category: Standards Track June 1995
4
5
6 <span class="h1">Relative Uniform Resource Locators</span>
7
8 Status of this Memo
9
10 This document specifies an Internet standards track protocol for the
11 Internet community, and requests discussion and suggestions for
12 improvements. Please refer to the current edition of the "Internet
13 Official Protocol Standards" (STD 1) for the standardization state
14 and status of this protocol. Distribution of this memo is unlimited.
15
16 Abstract
17
18 A Uniform Resource Locator (URL) is a compact representation of the
19 location and access method for a resource available via the Internet.
20 When embedded within a base document, a URL in its absolute form may
21 contain a great deal of information which is already known from the
22 context of that base document's retrieval, including the scheme,
23 network location, and parts of the url-path. In situations where the
24 base URL is well-defined and known to the parser (human or machine),
25 it is useful to be able to embed URL references which inherit that
26 context rather than re-specifying it in every instance. This
27 document defines the syntax and semantics for such Relative Uniform
28 Resource Locators.
29
30 <span class="h2"><a class="selflink" id="section-1" href="#section-1">1</a>. Introduction</span>
31
32 This document describes the syntax and semantics for "relative"
33 Uniform Resource Locators (relative URLs): a compact representation
34 of the location of a resource relative to an absolute base URL. It
35 is a companion to <a href="./rfc1738">RFC 1738</a>, "Uniform Resource Locators (URL)" [<a href="#ref-2" title="&quot;Uniform Resource Locators (URL)&quot;">2</a>],
36 which specifies the syntax and semantics of absolute URLs.
37
38 A common use for Uniform Resource Locators is to embed them within a
39 document (referred to as the "base" document) for the purpose of
40 identifying other Internet-accessible resources. For example, in
41 hypertext documents, URLs can be used as the identifiers for
42 hypertext link destinations.
43
44 Absolute URLs contain a great deal of information which may already
45 be known from the context of the base document's retrieval, including
46 the scheme, network location, and parts of the URL path. In
47 situations where the base URL is well-defined and known, it is useful
48 to be able to embed a URL reference which inherits that context
49
50
51
52 <span class="grey">Fielding Standards Track [Page 1]</span></pre>
53 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-2" ></span>
54 <span class="grey"><a href="./rfc1808">RFC 1808</a> Relative Uniform Resource Locators June 1995</span>
55
56
57 rather than re-specifying it within each instance. Relative URLs can
58 also be used within data-entry dialogs to decrease the number of
59 characters necessary to describe a location.
60
61 In addition, it is often the case that a group or "tree" of documents
62 has been constructed to serve a common purpose; the vast majority of
63 URLs in these documents point to locations within the tree rather
64 than outside of it. Similarly, documents located at a particular
65 Internet site are much more likely to refer to other resources at
66 that site than to resources at remote sites.
67
68 Relative addressing of URLs allows document trees to be partially
69 independent of their location and access scheme. For instance, it is
70 possible for a single set of hypertext documents to be simultaneously
71 accessible and traversable via each of the "file", "http", and "ftp"
72 schemes if the documents refer to each other using relative URLs.
73 Furthermore, document trees can be moved, as a whole, without
74 changing any of the embedded URLs. Experience within the World-Wide
75 Web has demonstrated that the ability to perform relative referencing
76 is necessary for the long-term usability of embedded URLs.
77
78 <span class="h2"><a class="selflink" id="section-2" href="#section-2">2</a>. Relative URL Syntax</span>
79
80 The syntax for relative URLs is a shortened form of that for absolute
81 URLs [<a href="#ref-2" title="&quot;Uniform Resource Locators (URL)&quot;">2</a>], where some prefix of the URL is missing and certain path
82 components ("." and "..") have a special meaning when interpreting a
83 relative path. Because a relative URL may appear in any context that
84 could hold an absolute URL, systems that support relative URLs must
85 be able to recognize them as part of the URL parsing process.
86
87 Although this document does not seek to define the overall URL
88 syntax, some discussion of it is necessary in order to describe the
89 parsing of relative URLs. In particular, base documents can only
90 make use of relative URLs when their base URL fits within the
91 generic-RL syntax described below. Although some URL schemes do not
92 require this generic-RL syntax, it is assumed that any document which
93 contains a relative reference does have a base URL that obeys the
94 syntax. In other words, relative URLs cannot be used within
95 documents that have unsuitable base URLs.
96
97 <span class="h3"><a class="selflink" id="section-2.1" href="#section-2.1">2.1</a>. URL Syntactic Components</span>
98
99 The URL syntax is dependent upon the scheme. Some schemes use
100 reserved characters like "?" and ";" to indicate special components,
101 while others just consider them to be part of the path. However,
102 there is enough uniformity in the use of URLs to allow a parser to
103 resolve relative URLs based upon a single, generic-RL syntax. This
104 generic-RL syntax consists of six components:
105
106
107
108 <span class="grey">Fielding Standards Track [Page 2]</span></pre>
109 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-3" ></span>
110 <span class="grey"><a href="./rfc1808">RFC 1808</a> Relative Uniform Resource Locators June 1995</span>
111
112
113 &lt;scheme&gt;://&lt;net_loc&gt;/&lt;path&gt;;&lt;params&gt;?&lt;query&gt;#&lt;fragment&gt;
114
115 each of which, except &lt;scheme&gt;, may be absent from a particular URL.
116 These components are defined as follows (a complete BNF is provided
117 in <a href="#section-2.2">Section 2.2</a>):
118
119 scheme ":" ::= scheme name, as per <a href="./rfc1738#section-2.1">Section&nbsp;2.1 of RFC 1738</a> [<a href="#ref-2" title="&quot;Uniform Resource Locators (URL)&quot;">2</a>].
120
121 "//" net_loc ::= network location and login information, as per
122 <a href="./rfc1738#section-3.1">Section&nbsp;3.1 of RFC 1738</a> [<a href="#ref-2" title="&quot;Uniform Resource Locators (URL)&quot;">2</a>].
123
124 "/" path ::= URL path, as per <a href="./rfc1738#section-3.1">Section&nbsp;3.1 of RFC 1738</a> [<a href="#ref-2" title="&quot;Uniform Resource Locators (URL)&quot;">2</a>].
125
126 ";" params ::= object parameters (e.g., ";type=a" as in
127 <a href="./rfc1738#section-3.2.2">Section&nbsp;3.2.2 of RFC 1738</a> [<a href="#ref-2" title="&quot;Uniform Resource Locators (URL)&quot;">2</a>]).
128
129 "?" query ::= query information, as per <a href="./rfc1738#section-3.3">Section&nbsp;3.3 of
130 RFC 1738</a> [<a href="#ref-2" title="&quot;Uniform Resource Locators (URL)&quot;">2</a>].
131
132 "#" fragment ::= fragment identifier.
133
134 Note that the fragment identifier (and the "#" that precedes it) is
135 not considered part of the URL. However, since it is commonly used
136 within the same string context as a URL, a parser must be able to
137 recognize the fragment when it is present and set it aside as part of
138 the parsing process.
139
140 The order of the components is important. If both &lt;params&gt; and
141 &lt;query&gt; are present, the &lt;query&gt; information must occur after the
142 &lt;params&gt;.
143
144 <span class="h3"><a class="selflink" id="section-2.2" href="#section-2.2">2.2</a>. BNF for Relative URLs</span>
145
146 This is a BNF-like description of the Relative Uniform Resource
147 Locator syntax, using the conventions of <a href="./rfc822">RFC 822</a> [<a href="#ref-5" title="&quot;Standard for the Format of ARPA Internet Text Messages&quot;">5</a>], except that "|"
148 is used to designate alternatives. Briefly, literals are quoted with
149 "", parentheses "(" and ")" are used to group elements, optional
150 elements are enclosed in [brackets], and elements may be preceded
151 with &lt;n&gt;* to designate n or more repetitions of the following
152 element; n defaults to 0.
153
154 This BNF also describes the generic-RL syntax for valid base URLs.
155 Note that this differs from the URL syntax defined in <a href="./rfc1738">RFC 1738</a> [<a href="#ref-2" title="&quot;Uniform Resource Locators (URL)&quot;">2</a>] in
156 that all schemes are required to use a single set of reserved
157 characters and use them consistently within the major URL components.
158
159
160
161
162
163
164 <span class="grey">Fielding Standards Track [Page 3]</span></pre>
165 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-4" ></span>
166 <span class="grey"><a href="./rfc1808">RFC 1808</a> Relative Uniform Resource Locators June 1995</span>
167
168
169 URL = ( absoluteURL | relativeURL ) [ "#" fragment ]
170
171 absoluteURL = generic-RL | ( scheme ":" *( uchar | reserved ) )
172
173 generic-RL = scheme ":" relativeURL
174
175 relativeURL = net_path | abs_path | rel_path
176
177 net_path = "//" net_loc [ abs_path ]
178 abs_path = "/" rel_path
179 rel_path = [ path ] [ ";" params ] [ "?" query ]
180
181 path = fsegment *( "/" segment )
182 fsegment = 1*pchar
183 segment = *pchar
184
185 params = param *( ";" param )
186 param = *( pchar | "/" )
187
188 scheme = 1*( alpha | digit | "+" | "-" | "." )
189 net_loc = *( pchar | ";" | "?" )
190 query = *( uchar | reserved )
191 fragment = *( uchar | reserved )
192
193 pchar = uchar | ":" | "@" | "&amp;" | "="
194 uchar = unreserved | escape
195 unreserved = alpha | digit | safe | extra
196
197 escape = "%" hex hex
198 hex = digit | "A" | "B" | "C" | "D" | "E" | "F" |
199 "a" | "b" | "c" | "d" | "e" | "f"
200
201 alpha = lowalpha | hialpha
202 lowalpha = "a" | "b" | "c" | "d" | "e" | "f" | "g" | "h" | "i" |
203 "j" | "k" | "l" | "m" | "n" | "o" | "p" | "q" | "r" |
204 "s" | "t" | "u" | "v" | "w" | "x" | "y" | "z"
205 hialpha = "A" | "B" | "C" | "D" | "E" | "F" | "G" | "H" | "I" |
206 "J" | "K" | "L" | "M" | "N" | "O" | "P" | "Q" | "R" |
207 "S" | "T" | "U" | "V" | "W" | "X" | "Y" | "Z"
208
209 digit = "0" | "1" | "2" | "3" | "4" | "5" | "6" | "7" |
210 "8" | "9"
211
212 safe = "$" | "-" | "_" | "." | "+"
213 extra = "!" | "*" | "'" | "(" | ")" | ","
214 national = "{" | "}" | "|" | "\" | "^" | "~" | "[" | "]" | "`"
215 reserved = ";" | "/" | "?" | ":" | "@" | "&amp;" | "="
216 punctuation = "&lt;" | "&gt;" | "#" | "%" | &lt;"&gt;
217
218
219
220 <span class="grey">Fielding Standards Track [Page 4]</span></pre>
221 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-5" ></span>
222 <span class="grey"><a href="./rfc1808">RFC 1808</a> Relative Uniform Resource Locators June 1995</span>
223
224
225 <span class="h3"><a class="selflink" id="section-2.3" href="#section-2.3">2.3</a>. Specific Schemes and their Syntactic Categories</span>
226
227 Each URL scheme has its own rules regarding the presence or absence
228 of the syntactic components described in Sections <a href="#section-2.1">2.1</a> and <a href="#section-2.2">2.2</a>. In
229 addition, some schemes are never appropriate for use with relative
230 URLs. However, since relative URLs will only be used within contexts
231 in which they are useful, these scheme-specific differences can be
232 ignored by the resolution process.
233
234 Within this section, we include as examples only those schemes that
235 have a defined URL syntax in <a href="./rfc1738">RFC 1738</a> [<a href="#ref-2" title="&quot;Uniform Resource Locators (URL)&quot;">2</a>]. The following schemes are
236 never used with relative URLs:
237
238 mailto Electronic Mail
239 news USENET news
240 telnet TELNET Protocol for Interactive Sessions
241
242 Some URL schemes allow the use of reserved characters for purposes
243 outside the generic-RL syntax given above. However, such use is
244 rare. Relative URLs can be used with these schemes whenever the
245 applicable base URL follows the generic-RL syntax.
246
247 gopher Gopher and Gopher+ Protocols
248 prospero Prospero Directory Service
249 wais Wide Area Information Servers Protocol
250
251 Users of gopher URLs should note that gopher-type information is
252 almost always included at the beginning of what would be the
253 generic-RL path. If present, this type information prevents
254 relative-path references to documents with differing gopher-types.
255
256 Finally, the following schemes can always be parsed using the
257 generic-RL syntax. This does not necessarily imply that relative
258 URLs will be useful with these schemes -- that decision is left to
259 the system implementation and the author of the base document.
260
261 file Host-specific Files
262 ftp File Transfer Protocol
263 http Hypertext Transfer Protocol
264 nntp USENET news using NNTP access
265
266 NOTE: <a href="./rfc1738#section-5">Section&nbsp;5 of RFC 1738</a> specifies that the question-mark
267 character ("?") is allowed in an ftp or file path segment.
268 However, this is not true in practice and is believed to be an
269 error in the RFC. Similarly, <a href="./rfc1738">RFC 1738</a> allows the reserved
270 character semicolon (";") within an http path segment, but does
271 not define its semantics; the correct semantics are as defined
272 by this document for &lt;params&gt;.
273
274
275
276 <span class="grey">Fielding Standards Track [Page 5]</span></pre>
277 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-6" ></span>
278 <span class="grey"><a href="./rfc1808">RFC 1808</a> Relative Uniform Resource Locators June 1995</span>
279
280
281 We recommend that new schemes be designed to be parsable via the
282 generic-RL syntax if they are intended to be used with relative URLs.
283 A description of the allowed relative forms should be included when a
284 new scheme is registered, as per <a href="./rfc1738#section-4">Section&nbsp;4 of RFC 1738</a> [<a href="#ref-2" title="&quot;Uniform Resource Locators (URL)&quot;">2</a>].
285
286 <span class="h3"><a class="selflink" id="section-2.4" href="#section-2.4">2.4</a>. Parsing a URL</span>
287
288 An accepted method for parsing URLs is useful to clarify the
289 generic-RL syntax of <a href="#section-2.2">Section 2.2</a> and to describe the algorithm for
290 resolving relative URLs presented in <a href="#section-4">Section 4</a>. This section
291 describes the parsing rules for breaking down a URL (relative or
292 absolute) into the component parts described in <a href="#section-2.1">Section 2.1</a>. The
293 rules assume that the URL has already been separated from any
294 surrounding text and copied to a "parse string". The rules are
295 listed in the order in which they would be applied by the parser.
296
297 <span class="h4"><a class="selflink" id="section-2.4.1" href="#section-2.4.1">2.4.1</a>. Parsing the Fragment Identifier</span>
298
299 If the parse string contains a crosshatch "#" character, then the
300 substring after the first (left-most) crosshatch "#" and up to the
301 end of the parse string is the &lt;fragment&gt; identifier. If the
302 crosshatch is the last character, or no crosshatch is present, then
303 the fragment identifier is empty. The matched substring, including
304 the crosshatch character, is removed from the parse string before
305 continuing.
306
307 Note that the fragment identifier is not considered part of the URL.
308 However, since it is often attached to the URL, parsers must be able
309 to recognize and set aside fragment identifiers as part of the
310 process.
311
312 <span class="h4"><a class="selflink" id="section-2.4.2" href="#section-2.4.2">2.4.2</a>. Parsing the Scheme</span>
313
314 If the parse string contains a colon ":" after the first character
315 and before any characters not allowed as part of a scheme name (i.e.,
316 any not an alphanumeric, plus "+", period ".", or hyphen "-"), the
317 &lt;scheme&gt; of the URL is the substring of characters up to but not
318 including the first colon. These characters and the colon are then
319 removed from the parse string before continuing.
320
321 <span class="h4"><a class="selflink" id="section-2.4.3" href="#section-2.4.3">2.4.3</a>. Parsing the Network Location/Login</span>
322
323 If the parse string begins with a double-slash "//", then the
324 substring of characters after the double-slash and up to, but not
325 including, the next slash "/" character is the network location/login
326 (&lt;net_loc&gt;) of the URL. If no trailing slash "/" is present, the
327 entire remaining parse string is assigned to &lt;net_loc&gt;. The double-
328 slash and &lt;net_loc&gt; are removed from the parse string before
329
330
331
332 <span class="grey">Fielding Standards Track [Page 6]</span></pre>
333 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-7" ></span>
334 <span class="grey"><a href="./rfc1808">RFC 1808</a> Relative Uniform Resource Locators June 1995</span>
335
336
337 continuing.
338
339 <span class="h4"><a class="selflink" id="section-2.4.4" href="#section-2.4.4">2.4.4</a>. Parsing the Query Information</span>
340
341 If the parse string contains a question mark "?" character, then the
342 substring after the first (left-most) question mark "?" and up to the
343 end of the parse string is the &lt;query&gt; information. If the question
344 mark is the last character, or no question mark is present, then the
345 query information is empty. The matched substring, including the
346 question mark character, is removed from the parse string before
347 continuing.
348
349 <span class="h4"><a class="selflink" id="section-2.4.5" href="#section-2.4.5">2.4.5</a>. Parsing the Parameters</span>
350
351 If the parse string contains a semicolon ";" character, then the
352 substring after the first (left-most) semicolon ";" and up to the end
353 of the parse string is the parameters (&lt;params&gt;). If the semicolon
354 is the last character, or no semicolon is present, then &lt;params&gt; is
355 empty. The matched substring, including the semicolon character, is
356 removed from the parse string before continuing.
357
358 <span class="h4"><a class="selflink" id="section-2.4.6" href="#section-2.4.6">2.4.6</a>. Parsing the Path</span>
359
360 After the above steps, all that is left of the parse string is the
361 URL &lt;path&gt; and the slash "/" that may precede it. Even though the
362 initial slash is not part of the URL path, the parser must remember
363 whether or not it was present so that later processes can
364 differentiate between relative and absolute paths. Often this is
365 done by simply storing the preceding slash along with the path.
366
367 <span class="h2"><a class="selflink" id="section-3" href="#section-3">3</a>. Establishing a Base URL</span>
368
369 The term "relative URL" implies that there exists some absolute "base
370 URL" against which the relative reference is applied. Indeed, the
371 base URL is necessary to define the semantics of any embedded
372 relative URLs; without it, a relative reference is meaningless. In
373 order for relative URLs to be usable within a document, the base URL
374 of that document must be known to the parser.
375
376
377
378
379
380
381
382
383
384
385
386
387
388 <span class="grey">Fielding Standards Track [Page 7]</span></pre>
389 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-8" ></span>
390 <span class="grey"><a href="./rfc1808">RFC 1808</a> Relative Uniform Resource Locators June 1995</span>
391
392
393 The base URL of a document can be established in one of four ways,
394 listed below in order of precedence. The order of precedence can be
395 thought of in terms of layers, where the innermost defined base URL
396 has the highest precedence. This can be visualized graphically as:
397
398 .----------------------------------------------------------.
399 | .----------------------------------------------------. |
400 | | .----------------------------------------------. | |
401 | | | .----------------------------------------. | | |
402 | | | | (3.1) Base URL embedded in the | | | |
403 | | | | document's content | | | |
404 | | | `----------------------------------------' | | |
405 | | | (3.2) Base URL of the encapsulating entity | | |
406 | | | (message, document, or none). | | |
407 | | `----------------------------------------------' | |
408 | | (3.3) URL used to retrieve the entity | |
409 | `----------------------------------------------------' |
410 | (3.4) Base URL = "" (undefined) |
411 `----------------------------------------------------------'
412
413 <span class="h3"><a class="selflink" id="section-3.1" href="#section-3.1">3.1</a>. Base URL within Document Content</span>
414
415 Within certain document media types, the base URL of the document can
416 be embedded within the content itself such that it can be readily
417 obtained by a parser. This can be useful for descriptive documents,
418 such as tables of content, which may be transmitted to others through
419 protocols other than their usual retrieval context (e.g., E-Mail or
420 USENET news).
421
422 It is beyond the scope of this document to specify how, for each
423 media type, the base URL can be embedded. It is assumed that user
424 agents manipulating such media types will be able to obtain the
425 appropriate syntax from that media type's specification. An example
426 of how the base URL can be embedded in the Hypertext Markup Language
427 (HTML) [<a href="#ref-3" title="&quot;HyperText Markup Language Specification -- 2.0&quot;">3</a>] is provided in an Appendix (<a href="#section-10">Section 10</a>).
428
429 Messages are considered to be composite documents. The base URL of a
430 message can be specified within the message headers (or equivalent
431 tagged metainformation) of the message. For protocols that make use
432 of message headers like those described in <a href="./rfc822">RFC 822</a> [<a href="#ref-5" title="&quot;Standard for the Format of ARPA Internet Text Messages&quot;">5</a>], we recommend
433 that the format of this header be:
434
435 base-header = "Base" ":" "&lt;URL:" absoluteURL "&gt;"
436
437 where "Base" is case-insensitive and any whitespace (including that
438 used for line folding) inside the angle brackets is ignored. For
439 example, the header field
440
441
442
443
444 <span class="grey">Fielding Standards Track [Page 8]</span></pre>
445 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-9" ></span>
446 <span class="grey"><a href="./rfc1808">RFC 1808</a> Relative Uniform Resource Locators June 1995</span>
447
448
449 Base: &lt;URL:http://www.ics.uci.edu/Test/a/b/c&gt;
450
451 would indicate that the base URL for that message is the string
452 "<a href="http://www.ics.uci.edu/Test/a/b/c">http://www.ics.uci.edu/Test/a/b/c</a>". The base URL for a message
453 serves as both the base for any relative URLs within the message
454 headers and the default base URL for documents enclosed within the
455 message, as described in the next section.
456
457 Protocols which do not use the <a href="./rfc822">RFC 822</a> message header syntax, but
458 which do allow some form of tagged metainformation to be included
459 within messages, may define their own syntax for defining the base
460 URL as part of a message.
461
462 <span class="h3"><a class="selflink" id="section-3.2" href="#section-3.2">3.2</a>. Base URL from the Encapsulating Entity</span>
463
464 If no base URL is embedded, the base URL of a document is defined by
465 the document's retrieval context. For a document that is enclosed
466 within another entity (such as a message or another document), the
467 retrieval context is that entity; thus, the default base URL of the
468 document is the base URL of the entity in which the document is
469 encapsulated.
470
471 Composite media types, such as the "multipart/*" and "message/*"
472 media types defined by MIME (<a href="./rfc1521">RFC 1521</a>, [<a href="#ref-4" title="&quot;MIME (Multipurpose Internet Mail Extensions): Mechanisms for Specifying and Describing the Format of Internet Message Bodies&quot;">4</a>]), define a hierarchy of
473 retrieval context for their enclosed documents. In other words, the
474 retrieval context of a component part is the base URL of the
475 composite entity of which it is a part. Thus, a composite entity can
476 redefine the retrieval context of its component parts via the
477 inclusion of a base-header, and this redefinition applies recursively
478 for a hierarchy of composite parts. Note that this might not change
479 the base URL of the components, since each component may include an
480 embedded base URL or base-header that takes precedence over the
481 retrieval context.
482
483 <span class="h3"><a class="selflink" id="section-3.3" href="#section-3.3">3.3</a>. Base URL from the Retrieval URL</span>
484
485 If no base URL is embedded and the document is not encapsulated
486 within some other entity (e.g., the top level of a composite entity),
487 then, if a URL was used to retrieve the base document, that URL shall
488 be considered the base URL. Note that if the retrieval was the
489 result of a redirected request, the last URL used (i.e., that which
490 resulted in the actual retrieval of the document) is the base URL.
491
492 <span class="h3"><a class="selflink" id="section-3.4" href="#section-3.4">3.4</a>. Default Base URL</span>
493
494 If none of the conditions described in Sections <a href="#section-3.1">3.1</a> -- 3.3 apply,
495 then the base URL is considered to be the empty string and all
496 embedded URLs within that document are assumed to be absolute URLs.
497
498
499
500 <span class="grey">Fielding Standards Track [Page 9]</span></pre>
501 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-10" ></span>
502 <span class="grey"><a href="./rfc1808">RFC 1808</a> Relative Uniform Resource Locators June 1995</span>
503
504
505 It is the responsibility of the distributor(s) of a document
506 containing relative URLs to ensure that the base URL for that
507 document can be established. It must be emphasized that relative
508 URLs cannot be used reliably in situations where the document's base
509 URL is not well-defined.
510
511 <span class="h2"><a class="selflink" id="section-4" href="#section-4">4</a>. Resolving Relative URLs</span>
512
513 This section describes an example algorithm for resolving URLs within
514 a context in which the URLs may be relative, such that the result is
515 always a URL in absolute form. Although this algorithm cannot
516 guarantee that the resulting URL will equal that intended by the
517 original author, it does guarantee that any valid URL (relative or
518 absolute) can be consistently transformed to an absolute form given a
519 valid base URL.
520
521 The following steps are performed in order:
522
523 Step 1: The base URL is established according to the rules of
524 <a href="#section-3">Section 3</a>. If the base URL is the empty string (unknown),
525 the embedded URL is interpreted as an absolute URL and
526 we are done.
527
528 Step 2: Both the base and embedded URLs are parsed into their
529 component parts as described in <a href="#section-2.4">Section 2.4</a>.
530
531 a) If the embedded URL is entirely empty, it inherits the
532 entire base URL (i.e., is set equal to the base URL)
533 and we are done.
534
535 b) If the embedded URL starts with a scheme name, it is
536 interpreted as an absolute URL and we are done.
537
538 c) Otherwise, the embedded URL inherits the scheme of
539 the base URL.
540
541 Step 3: If the embedded URL's &lt;net_loc&gt; is non-empty, we skip to
542 Step 7. Otherwise, the embedded URL inherits the &lt;net_loc&gt;
543 (if any) of the base URL.
544
545 Step 4: If the embedded URL path is preceded by a slash "/", the
546 path is not relative and we skip to Step 7.
547
548
549
550
551
552
553
554
555
556 <span class="grey">Fielding Standards Track [Page 10]</span></pre>
557 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-11" ></span>
558 <span class="grey"><a href="./rfc1808">RFC 1808</a> Relative Uniform Resource Locators June 1995</span>
559
560
561 Step 5: If the embedded URL path is empty (and not preceded by a
562 slash), then the embedded URL inherits the base URL path,
563 and
564
565 a) if the embedded URL's &lt;params&gt; is non-empty, we skip to
566 step 7; otherwise, it inherits the &lt;params&gt; of the base
567 URL (if any) and
568
569 b) if the embedded URL's &lt;query&gt; is non-empty, we skip to
570 step 7; otherwise, it inherits the &lt;query&gt; of the base
571 URL (if any) and we skip to step 7.
572
573 Step 6: The last segment of the base URL's path (anything
574 following the rightmost slash "/", or the entire path if no
575 slash is present) is removed and the embedded URL's path is
576 appended in its place. The following operations are
577 then applied, in order, to the new path:
578
579 a) All occurrences of "./", where "." is a complete path
580 segment, are removed.
581
582 b) If the path ends with "." as a complete path segment,
583 that "." is removed.
584
585 c) All occurrences of "&lt;segment&gt;/../", where &lt;segment&gt; is a
586 complete path segment not equal to "..", are removed.
587 Removal of these path segments is performed iteratively,
588 removing the leftmost matching pattern on each iteration,
589 until no matching pattern remains.
590
591 d) If the path ends with "&lt;segment&gt;/..", where &lt;segment&gt; is a
592 complete path segment not equal to "..", that
593 "&lt;segment&gt;/.." is removed.
594
595 Step 7: The resulting URL components, including any inherited from
596 the base URL, are recombined to give the absolute form of
597 the embedded URL.
598
599 Parameters, regardless of their purpose, do not form a part of the
600 URL path and thus do not affect the resolving of relative paths. In
601 particular, the presence or absence of the ";type=d" parameter on an
602 ftp URL does not affect the interpretation of paths relative to that
603 URL. Fragment identifiers are only inherited from the base URL when
604 the entire embedded URL is empty.
605
606
607
608
609
610
611
612 <span class="grey">Fielding Standards Track [Page 11]</span></pre>
613 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-12" ></span>
614 <span class="grey"><a href="./rfc1808">RFC 1808</a> Relative Uniform Resource Locators June 1995</span>
615
616
617 The above algorithm is intended to provide an example by which the
618 output of implementations can be tested -- implementation of the
619 algorithm itself is not required. For example, some systems may find
620 it more efficient to implement Step 6 as a pair of segment stacks
621 being merged, rather than as a series of string pattern matches.
622
623 <span class="h2"><a class="selflink" id="section-5" href="#section-5">5</a>. Examples and Recommended Practice</span>
624
625 Within an object with a well-defined base URL of
626
627 Base: &lt;URL:http://a/b/c/d;p?q#f&gt;
628
629 the relative URLs would be resolved as follows:
630
631 <span class="h3"><a class="selflink" id="section-5.1" href="#section-5.1">5.1</a>. Normal Examples</span>
632
633 g:h = &lt;URL:g:h&gt;
634 g = &lt;URL:http://a/b/c/g&gt;
635 ./g = &lt;URL:http://a/b/c/g&gt;
636 g/ = &lt;URL:http://a/b/c/g/&gt;
637 /g = &lt;URL:http://a/g&gt;
638 //g = &lt;URL:http://g&gt;
639 ?y = &lt;URL:http://a/b/c/d;p?y&gt;
640 g?y = &lt;URL:http://a/b/c/g?y&gt;
641 g?y/./x = &lt;URL:http://a/b/c/g?y/./x&gt;
642 #s = &lt;URL:http://a/b/c/d;p?q#s&gt;
643 g#s = &lt;URL:http://a/b/c/g#s&gt;
644 g#s/./x = &lt;URL:http://a/b/c/g#s/./x&gt;
645 g?y#s = &lt;URL:http://a/b/c/g?y#s&gt;
646 ;x = &lt;URL:http://a/b/c/d;x&gt;
647 g;x = &lt;URL:http://a/b/c/g;x&gt;
648 g;x?y#s = &lt;URL:http://a/b/c/g;x?y#s&gt;
649 . = &lt;URL:http://a/b/c/&gt;
650 ./ = &lt;URL:http://a/b/c/&gt;
651 .. = &lt;URL:http://a/b/&gt;
652 ../ = &lt;URL:http://a/b/&gt;
653 ../g = &lt;URL:http://a/b/g&gt;
654 ../.. = &lt;URL:http://a/&gt;
655 ../../ = &lt;URL:http://a/&gt;
656 ../../g = &lt;URL:http://a/g&gt;
657
658 <span class="h3"><a class="selflink" id="section-5.2" href="#section-5.2">5.2</a>. Abnormal Examples</span>
659
660 Although the following abnormal examples are unlikely to occur in
661 normal practice, all URL parsers should be capable of resolving them
662 consistently. Each example uses the same base as above.
663
664
665
666
667
668 <span class="grey">Fielding Standards Track [Page 12]</span></pre>
669 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-13" ></span>
670 <span class="grey"><a href="./rfc1808">RFC 1808</a> Relative Uniform Resource Locators June 1995</span>
671
672
673 An empty reference resolves to the complete base URL:
674
675 &lt;&gt; = &lt;URL:http://a/b/c/d;p?q#f&gt;
676
677 Parsers must be careful in handling the case where there are more
678 relative path ".." segments than there are hierarchical levels in the
679 base URL's path. Note that the ".." syntax cannot be used to change
680 the &lt;net_loc&gt; of a URL.
681
682 ../../../g = &lt;URL:http://a/../g&gt;
683 ../../../../g = &lt;URL:http://a/../../g&gt;
684
685 Similarly, parsers must avoid treating "." and ".." as special when
686 they are not complete components of a relative path.
687
688 /./g = &lt;URL:http://a/./g&gt;
689 /../g = &lt;URL:http://a/../g&gt;
690 g. = &lt;URL:http://a/b/c/g.&gt;
691 .g = &lt;URL:http://a/b/c/.g&gt;
692 g.. = &lt;URL:http://a/b/c/g..&gt;
693 ..g = &lt;URL:http://a/b/c/..g&gt;
694
695 Less likely are cases where the relative URL uses unnecessary or
696 nonsensical forms of the "." and ".." complete path segments.
697
698 ./../g = &lt;URL:http://a/b/g&gt;
699 ./g/. = &lt;URL:http://a/b/c/g/&gt;
700 g/./h = &lt;URL:http://a/b/c/g/h&gt;
701 g/../h = &lt;URL:http://a/b/c/h&gt;
702
703 Finally, some older parsers allow the scheme name to be present in a
704 relative URL if it is the same as the base URL scheme. This is
705 considered to be a loophole in prior specifications of partial URLs
706 [<a href="#ref-1" title="&quot;Universal Resource Identifiers in WWW: A Unifying Syntax for the Expression of Names and Addresses of Objects on the Network as used in the World-Wide Web&quot;">1</a>] and should be avoided by future parsers.
707
708 http:g = &lt;URL:http:g&gt;
709 http: = &lt;URL:http:&gt;
710
711 <span class="h3"><a class="selflink" id="section-5.3" href="#section-5.3">5.3</a>. Recommended Practice</span>
712
713 Authors should be aware that path names which contain a colon ":"
714 character cannot be used as the first component of a relative URL
715 path (e.g., "this:that") because they will likely be mistaken for a
716 scheme name. It is therefore necessary to precede such cases with
717 other components (e.g., "./this:that"), or to escape the colon
718 character (e.g., "this%3Athat"), in order for them to be correctly
719 parsed. The former solution is preferred because it does not affect
720 the absolute form of the URL.
721
722
723
724 <span class="grey">Fielding Standards Track [Page 13]</span></pre>
725 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-14" ></span>
726 <span class="grey"><a href="./rfc1808">RFC 1808</a> Relative Uniform Resource Locators June 1995</span>
727
728
729 There is an ambiguity in the semantics for the ftp URL scheme
730 regarding the use of a trailing slash ("/") character and/or a
731 parameter ";type=d" to indicate a resource that is an ftp directory.
732 If the result of retrieving that directory includes embedded relative
733 URLs, it is necessary that the base URL path for that result include
734 a trailing slash. For this reason, we recommend that the ";type=d"
735 parameter value not be used within contexts that allow relative URLs.
736
737 <span class="h2"><a class="selflink" id="section-6" href="#section-6">6</a>. Security Considerations</span>
738
739 There are no security considerations in the use or parsing of
740 relative URLs. However, once a relative URL has been resolved to its
741 absolute form, the same security considerations apply as those
742 described in <a href="./rfc1738">RFC 1738</a> [<a href="#ref-2" title="&quot;Uniform Resource Locators (URL)&quot;">2</a>].
743
744 <span class="h2"><a class="selflink" id="section-7" href="#section-7">7</a>. Acknowledgements</span>
745
746 This work is derived from concepts introduced by Tim Berners-Lee and
747 the World-Wide Web global information initiative. Relative URLs are
748 described as "Partial URLs" in <a href="./rfc1630">RFC 1630</a> [<a href="#ref-1" title="&quot;Universal Resource Identifiers in WWW: A Unifying Syntax for the Expression of Names and Addresses of Objects on the Network as used in the World-Wide Web&quot;">1</a>]. That description was
749 expanded for inclusion as an appendix for an early draft of <a href="./rfc1738">RFC 1738</a>,
750 "Uniform Resource Locators (URL)" [<a href="#ref-2" title="&quot;Uniform Resource Locators (URL)&quot;">2</a>]. However, after further
751 discussion, the URI-WG decided to specify Relative URLs separately
752 from the primary URL draft.
753
754 This document is intended to fulfill the recommendations for Internet
755 Resource Locators as stated in [<a href="#ref-6" title="&quot;Functional Recommendations for Internet Resource Locators&quot;">6</a>]. It has benefited greatly from
756 the comments of all those participating in the URI-WG. Particular
757 thanks go to Larry Masinter, Michael A. Dolan, Guido van Rossum, Dave
758 Kristol, David Robinson, and Brad Barber for identifying
759 problems/deficiencies in earlier drafts.
760
761 <span class="h2"><a class="selflink" id="section-8" href="#section-8">8</a>. References</span>
762
763 [<a id="ref-1">1</a>] Berners-Lee, T., "Universal Resource Identifiers in WWW: A
764 Unifying Syntax for the Expression of Names and Addresses of
765 Objects on the Network as used in the World-Wide Web", <a href="./rfc1630">RFC 1630</a>,
766 CERN, June 1994.
767
768 [<a id="ref-2">2</a>] Berners-Lee, T., Masinter, L., and M. McCahill, Editors, "Uniform
769 Resource Locators (URL)", <a href="./rfc1738">RFC 1738</a>, CERN, Xerox Corporation,
770 University of Minnesota, December 1994.
771
772 [<a id="ref-3">3</a>] Berners-Lee T., and D. Connolly, "HyperText Markup Language
773 Specification -- 2.0", Work in Progress, MIT, HaL Computer
774 Systems, February 1995.
775 &lt;URL:http://www.ics.uci.edu/pub/ietf/html/&gt;
776
777
778
779
780 <span class="grey">Fielding Standards Track [Page 14]</span></pre>
781 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-15" ></span>
782 <span class="grey"><a href="./rfc1808">RFC 1808</a> Relative Uniform Resource Locators June 1995</span>
783
784
785 [<a id="ref-4">4</a>] Borenstein, N., and N. Freed, "MIME (Multipurpose Internet Mail
786 Extensions): Mechanisms for Specifying and Describing the Format
787 of Internet Message Bodies", <a href="./rfc1521">RFC 1521</a>, Bellcore, Innosoft,
788 September 1993.
789
790 [<a id="ref-5">5</a>] Crocker, D., "Standard for the Format of ARPA Internet Text
791 Messages", STD 11, <a href="./rfc822">RFC 822</a>, UDEL, August 1982.
792
793 [<a id="ref-6">6</a>] Kunze, J., "Functional Recommendations for Internet Resource
794 Locators", <a href="./rfc1736">RFC 1736</a>, IS&amp;T, UC Berkeley, February 1995.
795
796 <span class="h2"><a class="selflink" id="section-9" href="#section-9">9</a>. Author's Address</span>
797
798 Roy T. Fielding
799 Department of Information and Computer Science
800 University of California
801 Irvine, CA 92717-3425
802 U.S.A.
803
804 Tel: +1 (714) 824-4049
805 Fax: +1 (714) 824-4056
806 EMail: [email protected]
807
808 <span class="h2"><a class="selflink" id="section-10" href="#section-10">10</a>. Appendix - Embedding the Base URL in HTML documents</span>
809
810 It is useful to consider an example of how the base URL of a document
811 can be embedded within the document's content. In this appendix, we
812 describe how documents written in the Hypertext Markup Language
813 (HTML) [<a href="#ref-3" title="&quot;HyperText Markup Language Specification -- 2.0&quot;">3</a>] can include an embedded base URL. This appendix does not
814 form a part of the relative URL specification and should not be
815 considered as anything more than a descriptive example.
816
817 HTML defines a special element "BASE" which, when present in the
818 "HEAD" portion of a document, signals that the parser should use the
819 BASE element's "HREF" attribute as the base URL for resolving any
820 relative URLs. The "HREF" attribute must be an absolute URL. Note
821 that, in HTML, element and attribute names are case-insensitive. For
822 example:
823
824 &lt;!doctype html public "-//IETF//DTD HTML//EN"&gt;
825 &lt;HTML&gt;&lt;HEAD&gt;
826 &lt;TITLE&gt;An example HTML document&lt;/TITLE&gt;
827 &lt;BASE href="http://www.ics.uci.edu/Test/a/b/c"&gt;
828 &lt;/HEAD&gt;&lt;BODY&gt;
829 ... &lt;A href="../x"&gt;a hypertext anchor&lt;/A&gt; ...
830 &lt;/BODY&gt;&lt;/HTML&gt;
831
832
833
834
835
836 <span class="grey">Fielding Standards Track [Page 15]</span></pre>
837 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-16" ></span>
838 <span class="grey"><a href="./rfc1808">RFC 1808</a> Relative Uniform Resource Locators June 1995</span>
839
840
841 A parser reading the example document should interpret the given
842 relative URL "../x" as representing the absolute URL
843
844 &lt;URL:http://www.ics.uci.edu/Test/a/x&gt;
845
846 regardless of the context in which the example document was obtained.
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892 Fielding Standards Track [Page 16]
893 </pre>