Skip to content
Snippets Groups Projects
Commit ebf5927e authored by Kevin M. Rosenberg's avatar Kevin M. Rosenberg
Browse files

r5331: *** empty log message ***

parent 97ace431
No related branches found
No related tags found
No related merge requests found
PURI - Portable URI Library PURI - Portable URI Library
Kevin Rosenberg <kevin@rosenberg.net>
Franz, Inc <http://www.franz.com> Franz, Inc <http://www.franz.com>
Kevin Rosenberg <kevin@rosenberg.net>
This is portable Universal Resource Identifier library for Common Lisp This is portable Universal Resource Identifier library for Common Lisp
programs. It parses URI according to the RFC 2396 specification. It's programs. It parses URI according to the RFC 2396 specification. It's
is based on Franz, Inc's opensource URI package and hash been is based on Franz, Inc's opensource URI package and has been ported to
ported to work other CL implementations. It is licensed with the work other CL implementations. It is licensed under the LLGPL which
LLGPL as include in this distribution. is included in this distribution.
A regression suite is included which uses Franz's open-source tester
library. I've ported that library for use on other CL
implementations. Puri completes 126/126 regression tests successfully.
A regression package is include which uses Franz's open-source Franz's unmodified documentation file is included in the file
tester library. I've also ported that library for use on other uri.html. The only divergence in usage between Puri and Franz's
CL implementations. Puri completes 126/126 regression tests package is that Puri's symbols are located in the package PURI while
successfully. Franz's original uses the package NET.URI.
Puri home: http://files.b9.com/puri/ Puri home: http://files.b9.com/puri/
Portable tester home: http://files.b9.com/tester/ Portable tester home: http://files.b9.com/tester/
......
cl-puri (1.2.3-1) unstable; urgency=low
* Include uri.html documentation file.
-- Kevin M. Rosenberg <kmr@debian.org> Fri, 18 Jul 2003 19:57:03 -0600
cl-puri (1.2.2-1) unstable; urgency=low cl-puri (1.2.2-1) unstable; urgency=low
* Improve tests.lisp * Improve tests.lisp
......
...@@ -51,7 +51,7 @@ binary-indep: build install ...@@ -51,7 +51,7 @@ binary-indep: build install
binary-arch: build install binary-arch: build install
dh_testdir dh_testdir
dh_testroot dh_testroot
dh_installdocs README dh_installdocs README uri.html
dh_installchangelogs dh_installchangelogs
dh_strip dh_strip
dh_compress dh_compress
......
uri.html 0 → 100644
<html>
<head>
<title>URI support in Allegro CL</title>
</head>
<body>
<h1>URI support in Allegro CL</h1>
<p>This document contains the following sections:</p>
<a href="#uri-intro-1">
<p>1.0 Introduction</a><br>
<a href="#uri-api-1">2.0 The URI API definition</a><br>
<a href="#parsing-decoding-1">3.0 Parsing, escape decoding/encoding and the path</a><br>
<a href="#interning-uris-1">4.0 Interning URIs</a><br>
<a href="#acl-implementation-1">5.0 Allegro CL implementation notes</a><br>
<a href="#examples-1">6.0 Examples</a><br>
</p>
<p>This version of the Allegro CL URI support documentation is for distribution with the
Open Source version of the URI code. Links to Allegro CL documentation other than
URI-specific files have been supressed. To see Allegro CL documentation, see <a
href="http://www.franz.com/support/documentation/">http://www.franz.com/support/documentation/</a>,
which is the Allegro CL documentation page of the franz inc. website. Links to Allegro CL
documentation can be found on that page. </p>
<hr>
<hr>
<h2><a name="uri-intro-1">1.0 Introduction</a></h2>
<p><em>URI</em> stands for <em>Universal Resource Identifier</em>. For a description of
URIs, see RFC2396, which can be found in several places, including the IETF web site (<a
href="http://www.ietf.org/rfc/rfc2396.txt">http://www.ietf.org/rfc/rfc2396.txt</a>) and
the UCI/ICS web site (<a href="http://www.ics.uci.edu/pub/ietf/uri/rfc2396.txt">http://www.ics.uci.edu/pub/ietf/uri/rfc2396.txt</a>).
We prefer the UCI/ICS one as it has more examples. </p>
<p>URIs are a superset in functionality and syntax to URLs (Universal Resource Locators)
and URNs (Universal Resource Names). That is, RFC2396 updates and merges RFC1738 and
RFC1808 into a single syntax, called the URI. It does exclude some portions of RFC1738
that define specific syntax of individual URL schemes. </p>
<p>In URL slang, the <em>scheme</em> is usually called the `protocol', but it is called
scheme in RFC1738. A URL `host' corresponds to the URI `authority.' The URL slang
`bookmark' or `anchor' is `fragment' in URI lingo. </p>
<p>The URI facility was available as a patch to Allegro CL 5.0.1 and is included with
release 6.0. the URI facility might not be in an Allegro CL image. Evaluate <code>(require
:uri)</code> to ensure the facility is loaded (that form returns <code>nil</code> if the
URI module is already loaded). </p>
<p>Broadly, the URI facility creates a Lisp object that represents a URI, and provides
setters and accessors to fields in the URI object. The URI object can also be interned,
much like symbols in CL are. This document describes the facility and the related
operators. </p>
<p>Aside from the obvious slots which are called out in the RFC, URIs also have a property
list. With interning, this is another similarity between URIs and CL symbols. </p>
<hr>
<hr>
<h2><a name="uri-api-1">2.0 The URI API definition</a></h2>
<p>Symbols naming objects (functions, variables, etc.) in the <em>uri</em> module are
exported from the <code>net.uri</code> package. </p>
<p>URIs are represented by CLOS objects. Their slots are: </p>
<pre>
scheme
host
port
path
query
fragment
plist
</pre>
<p>The <code>host</code> and <code>port</code> slots together correspond to the <code>authority</code>
(see RFC2396). There is an accessor-like function, <a href="operators/uri-authority.htm"><b>uri-authority</b></a>,
that can be used to extract the authority from a URI. See the RFC2396 specifications
pointed to at the beginning of the <a href="#uri-intro-1">1.0 Introduction</a> for details
of all the slots except <code>plist</code>. The <code>plist</code> slot contains a
standard Common Lisp property list. </p>
<p>All symbols are external in the <code>net.uri</code> package, unless otherwise noted.
Brief descriptions are given in this document, with complete descriptions in the
individual pages.
<ul>
<li><a href="classes/uri.htm"><code>uri</code></a>: the class of URI objects. </li>
<li><a href="classes/urn.htm"><code>urn</code></a>: the class of URN objects. </li>
<li><a href="operators/uri-p.htm"><b>uri-p</b></a> <p><b>Arguments: </b><i>object</i></p>
<p>Returns true if <i>object</i> is an instance of class <a href="classes/uri.htm"><code>uri</code></a>.
</p>
</li>
<li><a href="operators/copy-uri.htm"><b>copy-uri</b></a> <p><b>Arguments: </b><i>uri </i>&amp;key
<i>place scheme host port path query fragment plist </i></p>
<p>Copies the specified URI object. See the description page for information on the
keyword arguments. </p>
</li>
<li><a href="operators/uri-scheme.htm"><b>uri-scheme</b></a><br>
<a href="operators/uri-host.htm"><b>uri-host</b></a><br>
<a href="operators/uri-port.htm"><b>uri-port</b></a><br>
<a href="operators/uri-path.htm"><b>uri-path</b></a><br>
<a href="operators/uri-query.htm"><b>uri-query</b></a><br>
<a href="operators/uri-fragment.htm"><b>uri-fragment</b></a><br>
<a href="operators/uri-plist.htm"><b>uri-plist</b></a><br>
<p><b>Arguments: </b><i>uri-object </i></p>
<p>These accessors return the value of the associated slots of the <i>uri-object</i> </p>
</li>
<li><a href="operators/uri-authority.htm"><b>uri-authority</b></a> <p><b>Arguments: </b><i>uri-object
</i></p>
<p>Returns the authority of <i>uri-object</i>. The authority combines the host and port. </p>
</li>
<li><a href="operators/render-uri.htm"><b>render-uri</b></a> <p><b>Arguments: </b><i>uri
stream </i></p>
<p>Print to <i>stream</i> the printed representation of <i>uri</i>. </p>
</li>
<li><a href="operators/parse-uri.htm"><b>parse-uri</b></a> <p><b>Arguments: </b><i>string </i>&amp;key
(<i>class</i> 'uri)<i> </i></p>
<p>Parse <i>string</i> into a URI object. </p>
</li>
<li><a href="operators/merge-uris.htm"><b>merge-uris</b></a> <p><b>Arguments: </b><i>uri
base-uri </i>&amp;optional <i>place </i></p>
<p>Return an absolute URI, based on <i>uri</i>, which can be relative, and <i>base-uri</i>
which must be absolute. </p>
</li>
<li><a href="operators/enough-uri.htm"><b>enough-uri</b></a> <p><b>Arguments: </b><i>uri
base </i></p>
<p>Converts <i>uri</i> into a relative URI using <i>base</i> as the base URI. </p>
</li>
<li><a href="operators/uri-parsed-path.htm"><b>uri-parsed-path</b></a> <p><b>Arguments: </b><i>uri
</i></p>
<p>Return the parsed representation of the path. </p>
</li>
<li><a href="operators/uri.htm"><b>uri</b></a> <p><b>Arguments: </b><i>object </i></p>
<p>Defined methods: if argument is a uri object, return it; create a uri object if
possible and return it, or error if not possible. </p>
</li>
</ul>
<hr>
<hr>
<h2><a name="parsing-decoding-1">3.0 Parsing, escape decoding/encoding and the path</a></h2>
<p>The method <a href="operators/uri-path.htm"><b>uri-path</b></a> returns the path
portion of the URI, in string form. The method <a href="operators/uri-parsed-path.htm"><b>uri-parsed-path</b></a>
returns the path portion of the URI, in list form. This list form is discussed below,
after a discussion of decoding/encoding. </p>
<p>RFC2396 lays out a method for inserting into URIs <em>reserved characters</em>. You do
this by escaping the character. An <em>escaped</em> character is defined like this: </p>
<pre>
escaped = &quot;%&quot; hex hex
hex = digit | &quot;A&quot; | &quot;B&quot; | &quot;C&quot; | &quot;D&quot; | &quot;E&quot; | &quot;F&quot; | &quot;a&quot; | &quot;b&quot; | &quot;c&quot; | &quot;d&quot; | &quot;e&quot; | &quot;f&quot;
</pre>
<p>In addition, the RFC defines excluded characters: </p>
<pre>
&quot;&lt;&quot; | &quot;&gt;&quot; | &quot;#&quot; | &quot;%&quot; | &lt;&quot;&gt; | &quot;{&quot; | &quot;}&quot; | &quot;|&quot; | &quot;\&quot; | &quot;^&quot; | &quot;[&quot; | &quot;]&quot; | &quot;`&quot;
</pre>
<p>The set of reserved characters are: </p>
<pre>
&quot;;&quot; | &quot;/&quot; | &quot;?&quot; | &quot;:&quot; | &quot;@&quot; | &quot;&amp;&quot; | &quot;=&quot; | &quot;+&quot; | &quot;$&quot; | &quot;,&quot;
</pre>
<p>with the following exceptions:
<ul>
<li>within the authority component, the characters &quot;;&quot;, &quot;:&quot;,
&quot;@&quot;, &quot;?&quot;, and &quot;/&quot; are reserved. </li>
<li>within a path segment, the characters &quot;/&quot;, &quot;;&quot;, &quot;=&quot;, and
&quot;?&quot; are reserved. </li>
<li>within a query component, the characters &quot;;&quot;, &quot;/&quot;, &quot;?&quot;,
&quot;:&quot;, &quot;@&quot;, &quot;&amp;&quot;, &quot;=&quot;, &quot;+&quot;,
&quot;,&quot;, and &quot;$&quot; are reserved. </li>
</ul>
<p>From the RFC, there are two important rules about escaping and unescaping (encoding and
decoding):
<ul>
<li>decoding should only happen when the URI is parsed into component parts;</li>
<li>encoding can only occur when a URI is made from component parts (ie, rendered for
printing). </li>
</ul>
<p>The implication of this is that to decode the URI, it must be in a parsed state. That
is, you can't convert <font face="Courier New">%2f</font> (the escaped form of
&quot;/&quot;) until the path has been parsed into its component parts. Another important
desire is for the application viewing the component parts to see the decoded values of the
components. For example, consider: </p>
<pre>
http://www.franz.com/calculator/3%2f2
</pre>
<p>This might be the implementation of a calculator, and how someone would execute 3/2.
Clearly, the application that implements this would want to see path components of
&quot;calculator&quot; and &quot;3/2&quot;. &quot;3%2f2&quot; would not be useful to the
calculator application. </p>
<p>For the reasons given above, a parsed version of the path is available and has the
following form: </p>
<pre>
([:absolute | :relative] component1 [component2...])
</pre>
<p>where components are: </p>
<pre>
element | (element param1 [param2 ...])
</pre>
<p>and <em>element</em> is a path element, and the param's are path element parameters.
For example, the result of </p>
<pre>
(uri-parsed-path (parse-uri &quot;foo;10/bar:x;y;z/baz.htm&quot;))
</pre>
<p>is </p>
<pre>
(:relative (&quot;foo&quot; &quot;10&quot;) (&quot;bar:x&quot; &quot;y&quot; &quot;z&quot;) &quot;baz.htm&quot;)
</pre>
<p>There is a certain amount of canonicalization that occurs when parsing:
<ul>
<li>A path of <code>(:absolute)</code> or <code>(:absolute &quot;&quot;)</code> is
equivalent to a <code>nil</code> path. That is, <code>http://a/</code> is parsed with a <code>nil</code>
path and printed as <code>http://a</code>. </li>
<li>Escaped characters that are not reserved are not escaped upon printing. For example, <code>&quot;foob%61r&quot;</code>
is parsed into <code>&quot;foobar&quot;</code> and appears as <code>&quot;foobar&quot;</code>
when the URI is printed. </li>
</ul>
<hr>
<hr>
<h2><a name="interning-uris-1">4.0 Interning URIs</a></h2>
<p>This section describes how to intern URIs. Interning is not mandatory. URIs can be used
perfectly well without interning them. </p>
<p>Interned URIs in Allegro are like symbols. That is, a string representing a URI, when
parsed and interned, will always yield an <strong>eq</strong> object. For example: </p>
<pre>
(eq (intern-uri &quot;http://www.franz.com&quot;)
(intern-uri &quot;http://www.franz.com&quot;))
</pre>
<p>is always true. (Two strings with identical contents may or may not be <strong>eq</strong>
in Common Lisp, note.) </p>
<p>The functions associated with interning are:
<ul>
<li><a href="operators/make-uri-space.htm"><b>make-uri-space</b></a> <p><b>Arguments: </b>&amp;key
<i>size </i></p>
<p>Make a new hash-table object to contain interned URIs. </p>
</li>
<li><a href="operators/uri-space.htm"><b>uri-space</b></a> <p><b>Arguments: </b></p>
<p>Return the object into which URIs are currently being interned. </p>
</li>
<li><a href="operators/uri_eq.htm"><b>uri=</b></a> <p><b>Arguments: </b><i>uri1 uri2 </i></p>
<p>Returns true if <i>uri1</i> and <i>uri2</i> are equivalent. </p>
</li>
<li><a href="operators/intern-uri.htm"><b>intern-uri</b></a> <p><b>Arguments: </b><i>uri-name
</i>&amp;optional <i>uri-space </i></p>
<p>Intern the uri object specified in the uri-space specified. Methods exist for strings
and uri objects. </p>
</li>
<li><a href="operators/unintern-uri.htm"><b>unintern-uri</b></a> <p><b>Arguments: </b><i>uri
</i>&amp;optional <i>uri-space </i></p>
<p>Unintern the uri object specified or all uri objects (in <i>uri-space</i> if specified)
if <i>uri</i> is <code>t</code>. </p>
</li>
<li><a href="operators/do-all-uris.htm"><b>do-all-uris</b></a> <p><b>Arguments: </b><i>(var </i>&amp;optional
<i>uri-space result) </i>&amp;body <i>body </i></p>
<p>Bind <i>var</i> to all currently defined uris (in <i>uri-space</i> if specified) and
evaluate <i>body</i>. </p>
</li>
</ul>
<hr>
<hr>
<h2><a name="acl-implementation-1">5.0 Allegro CL implementation notes</a></h2>
<ol>
<li>The following are true: <br>
<code>(uri= (parse-uri &quot;http://www.franz.com/&quot;)</code> <br>
&nbsp; &nbsp; <code>(parse-uri &quot;http://www.franz.com&quot;))</code> <br>
<code>(eq (intern-uri &quot;http://www.franz.com/&quot;)</code> <br>
&nbsp;&nbsp; <code>(intern-uri &quot;http://www.franz.com&quot;))</code><br>
</li>
<li>The following is true: <br>
<code>(eq (intern-uri &quot;http://www.franz.com:80/foo/bar.htm&quot;)</code> <br>
&nbsp; &nbsp; <code>(intern-uri &quot;http://www.franz.com/foo/bar.htm&quot;))</code><br>
(I.e. specifying the default port is the same as specifying no port at all. This is
specific in RFC2396.) </li>
<li>The <em>scheme</em> and <em>authority</em> are case-insensitive. In Allegro CL, the
scheme is a keyword that appears in the normal case for the Lisp in which you are
executing. </li>
<li><code>#u&quot;...&quot;</code> is shorthand for <code>(parse-uri &quot;...&quot;)</code>
but if an existing <code>#u</code> dispatch macro definition exists, it will not be
overridden. </li>
<li>The interaction between setting the scheme, host, port, path, query, and fragment slots
of URI objects, in conjunction with interning URIs will have very bad and unpredictable
results. </li>
<li>The printable representation of URIs is cached, for efficiency. This caching is undone
when the above slots are changed. That is, when you create a URI the printed
representation is cached. When you change one of the above mentioned slots, the printed
representation is cleared and calculated when the URI is next printed. For example: </li>
</ol>
<pre>
user(10): (setq u #u&quot;http://foo.bar.com/foo/bar&quot;)
#&lt;uri http://foo.bar.com/foo/bar&gt;
user(11): (setf (net.uri:uri-host u) &quot;foo.com&quot;)
&quot;foo.com&quot;
user(12): u
#&lt;uri http://foo.com/foo/bar&gt;
user(13):
</pre>
<p>This allows URIs behavior to follow the principle of least surprise. </p>
<hr>
<hr>
<h2><a name="examples-1">6.0 Examples</a></h2>
<pre>
uri(10): (use-package :net.uri)
t
uri(11): (parse-uri &quot;foo&quot;)
#&lt;uri foo&gt;
uri(12): #u&quot;foo&quot;
#&lt;uri foo&gt;
uri(13): (setq base (intern-uri &quot;http://www.franz.com/foo/bar/&quot;))
#&lt;uri http://www.franz.com/foo/bar/&gt;
uri(14): (merge-uris (parse-uri &quot;foo.htm&quot;) base)
#&lt;uri http://www.franz.com/foo/bar/foo.htm&gt;
uri(15): (merge-uris (parse-uri &quot;?foo&quot;) base)
#&lt;uri http://www.franz.com/foo/bar/?foo&gt;
uri(16): (setq base (intern-uri &quot;http://www.franz.com/foo/bar/baz.htm&quot;))
#&lt;uri http://www.franz.com/foo/bar/baz.htm&gt;
uri(17): (merge-uris (parse-uri &quot;foo.htm&quot;) base)
#&lt;uri http://www.franz.com/foo/bar/foo.htm&gt;
uri(18): (merge-uris #u&quot;?foo&quot; base)
#&lt;uri http://www.franz.com/foo/bar/?foo&gt;
uri(19): (describe #u&quot;http://www.franz.com&quot;)
#&lt;uri http://www.franz.com&gt; is an instance of #&lt;standard-class net.uri:uri&gt;:
The following slots have :instance allocation:
scheme :http
host &quot;www.franz.com&quot;
port nil
path nil
query nil
fragment nil
plist nil
escaped nil
string &quot;http://www.franz.com&quot;
parsed-path nil
hashcode nil
uri(20): (describe #u&quot;http://www.franz.com/&quot;)
#&lt;uri http://www.franz.com&gt; is an instance of #&lt;standard-class net.uri:uri&gt;:
The following slots have :instance allocation:
scheme :http
host &quot;www.franz.com&quot;
port nil
path nil
query nil
fragment nil
plist nil
escaped nil
string &quot;http://www.franz.com&quot;
parsed-path nil
hashcode nil
uri(21): #u&quot;foobar#baz%23xxx&quot;
#&lt;uri foobar#baz#xxx&gt;
</pre>
<p><small>Copyright (c) 1998-2001, Franz Inc. Berkeley, CA., USA. All rights reserved.
Created 2001.8.16.</small></p>
</body>
</html>
0% Loading or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment