source: project/wiki/eggref/4/uri-generic @ 15481

[[tags: eggs]]
[[toc:]]

== uri-generic

=== Description

The {{uri-generic}} library contains procedures for parsing and
manipulation of Uniform Resource Identifiers
([[http://tools.ietf.org/html/rfc3986|RFC 3986]]). It is intended to
conform more closely to the RFC, and uses combinator parsing and
character classes rather than regular expressions.

This library should be considered to be a ''basis'' for creating
scheme-specific URI parser libraries. This library only parses
the generic components from a URI.  Any specific library can
further parse subcomponents. For this reason, encoding and decoding
of percent-encoded characters is not done automatically.
This should be handled by specific URI scheme implementations.

=== Library Procedures

==== Constructors and predicates

As specified in section 2.3 of RFC 3986, URI constructors
automatically decode percent-encoded octets in the range of unreserved
characters. This means that the following holds true:

 (equal? (uri-reference "http://example.com/foo-bar")
         (uri-reference "http://example.com/foo%2Dbar"))  => #t

<procedure>(uri-reference STRING) => URI</procedure>

A URI reference is either a URI or a relative reference (RFC 3986,
Section 4.1).  If the given string's prefix does not match the syntax
of a scheme followed by a colon separator, then the given string is
parsed as a relative reference.

<procedure>(uri-reference? URI) => BOOL</procedure>

Is the given object a URI reference?  '''All objects created by
URI-generic constructors are URI references'''; they are either URIs
or relative references.  The constructors below are just more strict
checking versions of {{uri-reference}}.  They all create
URI references.

<procedure>(absolute-uri STRING) => URI</procedure>

Parses the given string as an absolute URI, in which no fragments are
allowed.  If no URI scheme is found, or a fragment is detected, this
raises an error.

Absolute URIs are defined by RFC 3986 as non-relative URI references
without a fragment (RFC 3986, Section 4.2).  Absolute URIs can be used
as a base URI to resolve a relative-ref against, using
{{uri-relative-to}} (see below).

<procedure>(absolute-uri? URI) => BOOL</procedure>

Is the given object an absolute URI?

<procedure>(uri? URI) => BOOL</procedure>

Is the given object a URI?  URIs are all URI references that include
a scheme part.  The other type of URI references are relative
references.

<procedure>(relative-ref? URI) => BOOL</procedure>

Is the given object a relative reference?  Relative references are
defined by RFC 3986 as URI references which are not URIs; they contain
no URI scheme and can be resolved against an absolute URI to obtain
a complete URI using {{uri-relative-to}}.

<procedure>(uri-path-absolute? URI) => BOOL</procedure>

Is the {{URI}}'s path component an absolute path?

<procedure>(uri-path-relative? URI) => BOOL</procedure>

Is the {{URI}}'s path component a relative path?

==== Attribute accessors

* <procedure>(uri-authority URI) => URI-AUTH</procedure>
* <procedure>(uri-scheme URI) => SYMBOL</procedure>
* <procedure>(uri-path URI) => LIST</procedure>
* <procedure>(uri-query URI) => STRING</procedure>
* <procedure>(uri-fragment) URI => STRING</procedure>
* <procedure>(uri-host URI) => STRING</procedure>
* <procedure>(uri-port URI) => INTEGER</procedure>
* <procedure>(uri-username URI) => STRING</procedure>
* <procedure>(uri-password URI) => STRING</procedure>
* <procedure>(authority? URI-AUTH) => BOOL</procedure>
* <procedure>(authority-host URI-AUTH) => STRING</procedure>
* <procedure>(authority-port URI-AUTH) => INTEGER</procedure>
* <procedure>(authority-username URI-AUTH) => STRING</procedure>
* <procedure>(authority-password URI-AUTH) => STRING</procedure>

If a component is not defined in the given URI, then the corresponding
accessor returns {{#f}}.

* <procedure>(update-uri URI #!key authority scheme path query fragment host port username password) => URI</procedure>
* <procedure>(update-authority URI-AUTH #!key host port username password) => URI</procedure>

Update the specified keys in the URI or URI-AUTH object in a
functional way (ie, it creates a new copy with the modifications).

==== String and List Representations

<procedure>(uri->string URI USERINFO) => STRING</procedure>

Reconstructs the given URI into a string; uses a supplied function
{{LAMBDA USERNAME PASSWORD -> STRING}} to map the userinfo part of the
URI

<procedure>(uri->list URI USERINFO) => LIST</procedure>

Returns a list of the form {{(SCHEME SPECIFIC FRAGMENT)}};
{{SPECIFIC}} is of the form {{(AUTHORITY PATH QUERY)}}.

==== Reference Resolution

<procedure>(uri-relative-to URI URI) => URI</procedure>

Resolve the first URI as a reference relative to the second URI,
returning a new URI (RFC 3986, Section 5.2.2).

<procedure>(uri-relative-from URI URI) => URI</procedure>

Constructs a new, possibly relative, URI which represents the location
of the first URI with respect to the second URI.

<examples>
<example>
<init>(use uri-generic)</init>
<expr>(uri->string (uri-relative-to (uri-reference "../qux") (uri-reference "http://example.com/foo/bar/")))</expr>
<result>"http://example.com/foo/qux"</result>
</example>
<example>
<init>(use uri-generic)</init>
<expr>(uri->string (uri-relative-from (uri-reference "http://example.com/foo/qux") (uri-reference "http://example.com/foo/bar/")))</expr>
<result>"../qux"</result>
</example>
</examples>

==== String encoding and decoding

<procedure>(uri-encode-string STRING [CHAR-SET]) => STRING</procedure>

Returns the percent-encoded form of the given string.  The optional
char-set argument controls which characters should be encoded.
It defaults to the complement of {{char-set:uri-unreserved}}. This is
always safe, but often overly careful; it is allowed to leave certain
characters unquoted depending on the context.

<procedure>(uri-decode-string STRING [CHAR-SET]) => STRING</procedure>

Returns the decoded form of the given string.  The optional char-set
argument controls which characters should be decoded.  It defaults to
{{char-set:full}}.


==== Normalization  

<procedure>(uri-normalize-case URI) => URI</procedure>

URI case normalization (RFC 3986 section 6.2.2.1)

<procedure>(uri-normalize-path-segments URI) => URI</procedure>

URI path segment normalization (RFC 3986 section 6.2.2.3)


==== Character sets

As a convenience for sub-parsers or other special-purpose URI handling
code, there are a couple of character sets exported by uri-generic.

<constant>char-set:gen-delims</constant>

Generic delimiters.
  gen-delims  =  ":" / "/" / "?" / "#" / "[" / "]" / "@"

<constant>char-set:sub-delims</constant>

Sub-delimiters.
  sub-delims  =  "!" / "$" / "&" / "'" / "(" / ")" / "*" / "+" / "," / ";" / "="

<constant>char-set:uri-reserved</constant>

The union of {{gen-delims}} and {{sub-delims}}; all reserved URI characters.
  reserved    =  gen-delims / sub-delims

<constant>char-set:uri-unreserved</constant>

All unreserved characters that are allowed in a URI.
  unreserved  =  ALPHA / DIGIT / "-" / "." / "_" / "~"

Note that this is _not_ the complement of {{char-set:uri-reserved}}!
There are several characters (even printable, noncontrol characters)
which are not allowed at all in a URI.


=== Requires

* [[matchable]]
* [[defstruct]]

=== Version History

* 2.32 Empty absolute path directly followed by query is now properly recognised as an URI reference.
* 2.31 Return {{#f}} in constructors if unconsumed input remains after parsing
* 2.3 Add predicates uri-path-relative? and uri-path-absolute?
* 2.2 Improvements to uri->string.
* 2.1 Add new predicates for URIs, absolute URIs and relative references. Fix absolute-uri so it raises a condition when passing in a non-absolute uri string, instead of returning a string with the error. Also throw an error if a fragment is detected in the string.
* 2.0 Export char-sets, add char-set arg to uri-encode/uri-decode,
       do not decode query args as x-www-form-urlencoded, change path
       representation.  Lots of bugfixes.
* 1.12 Fix relative path normalization when original path ends in a slash, remove consecutive slashes from paths in URIs
* 1.11 Added accessors for the authority components, functional update procedures. Fixed case-normalization.
* 1.10 Fixed edge case in {{uri-relative-to}} with empty path in base uri,
       fixed {{uri->string}} for URIs with query args, fixed {{uri->string}}
       to not add an extraneous slash after authority in case of empty path.
* 1.9 Fixed bug in uri-encode-string with reserved characters, added
      tests for decoding and encoding [Peter Bex]
* 1.8 Added uri-encode-string and uri-decode-string. 
      URI constructors now perform automatic normalization 
      of percent-encoded unreserved characters. [suggested by Peter Bex]
* 1.6 Added error message about missing scheme in absolute-uri. 
* trunk Small bugfix in absolute-uri. [Peter Bex]
* 1.5 Bug fixes in uri->string and absolute-uri. [reported by Peter Bex]
* 1.3 Ported to Hygienic Chicken and the [[test]] egg [Peter Bex]
* 1.2 Now using defstruct instead of define-record [suggested by Peter Bex]
* 1.1 Added utf8 compatibility
* 1.0 Initial Release

=== License

Based on the
[[http://www.ninebynine.org/Software/ReadMe-URI-Haskell.txt|Haskell
URI library]] by Graham Klyne <gk@ninebynine.org>.


  Copyright 2008-2009 Ivan Raikov, Peter Bex.
  All rights reserved.
  
  Redistribution and use in source and binary forms, with or without
  modification, are permitted provided that the following conditions are
  met:
  
  Redistributions of source code must retain the above copyright
  notice, this list of conditions and the following disclaimer.
  
  Redistributions in binary form must reproduce the above copyright
  notice, this list of conditions and the following disclaimer in the
  documentation and/or other materials provided with the distribution.
  
  Neither the name of the author nor the names of its contributors may
  be used to endorse or promote products derived from this software
  without specific prior written permission.
  
  THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
  "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
  LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
  FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
  COPYRIGHT HOLDERS OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT,
  INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
  (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
  SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
  STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
  ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
  OF THE POSSIBILITY OF SUCH DAMAGE.