Abstract

PHP is a server-based technology, but it can also act as a client. Using additional packages, a PHP script can request information from other servers. Snoopy is such an example, it implements HTTP and allows users to request resources from Web servers. In addition, Snoopy supports a methods to access these resources. The goal of this lecture is to introduce Snoopy so that it can serve as a foundation for further experiments when implementing Web-based services.

PHP Form Processing

• PHP accepts and processes the form parameters
• The result of PHP processing is sent to the browser
• How the results are computed is invisible
  • if configured correctly, PHP should be invisible, too

Server-Side Implementation

• Bibliography data has to be stored in some structured form
  • two plausible candidates for this are RDBMS and XML
• The bibliography service is implemented as XSLT
  • XML is read from a file, the XSLT generates HTML

$command = $saxon . ' dret.xml bibquery.xsl author=' . pass_string($formval_author) . ' format=html' ;
system ($command) ;

Interface vs. Implementation

• The interface is defined on the URI level
  • parameters are given in the URI query string
  • the result is delivered as an HTML fragment
• The implementation is hidden behind two technologies
  • the Web server finds a PHP file (through content negotiation)
  • the PHP file transforms an XML using XSLT
• Both technologies could be changed without affecting the service
  • PHP might need to be replaced with a Java implementation
  • XML/XSLT might need to be replaced with some DBMS (RDBMS/XDBMS)

Reaching Out

• Most PHP scripts run internally or locally
  • PHP code is executed as part of the PHP execution environment
  • it also provides access to local resources such as databases
  • PHP often is used as glue, not as application platform
• PHP may also access remote resources
  • HTTP can be used to access any Web-based resource
  • SOAP provides access to Web Services
  • specialized modules provide customized access (Amazon access)
• HTTP requires additional PHP modules
  • HTTPClient provides pure HTTP access
  • Snoopy includes minimal HTML handling facilities
  • both modules only implement subsets of HTTP

Fetching Resources

• Basic HTTP requests and features
  • only HTTP 1.0 is supported
  • depending on the server, Snoopy might have problems

<?php 
include "../a/3/Snoopy.class.php";
$snoopy = new Snoopy;

$snoopy->fetch("http://www.ischool.berkeley.edu/");
print $snoopy->results;
?>

<?php
include "../a/3/Snoopy.class.php";
$snoopy = new Snoopy;

$snoopy->fetchtext("http://www.ischool.berkeley.edu/");
print $snoopy->results;
?>

HTTP Headers

<?php
include "../a/3/Snoopy.class.php";
$snoopy = new Snoopy;

if($snoopy->fetch("http://www.ischool.berkeley.edu/"))
{
    echo "response code: ".$snoopy->response_code."<br>\n";
    while(list($key,$val) = each($snoopy->headers))
        echo $key.": ".$val."<br>\n";
    echo "<p>\n";

    echo "<PRE>".htmlspecialchars($snoopy->results)."</PRE>\n";
}
else
    echo "error fetching document: ".$snoopy->error."\n";
?>

HTML Links

<?php
include "../a/3/Snoopy.class.php";
$snoopy = new Snoopy;

$snoopy->fetchlinks("http://www.ischool.berkeley.edu/");

while(list($key,$val) = each($snoopy->results))
 echo $key.": ".$val."<br>\n";
?>

Using Parameters

• All parameters are passed as URI query strings
  • some characters must be escaped to conform to URI naming rules
  • when using forms, the browser automatically escapes characters
  • when composing URIs out of strings, escaping functions must be used
• Parameters may be omitted
  • parameters which are nor specified do not perform any filtering
• All parameters are combined by a logical and
  • each parameter narrows the search
  • logical or is not implemented

Author Names

• Searches all author fields for the name

query?author=Hoare

query?author=Harry

Entry Titles

• Searching through all titles of an entry
  • entries may have more than one title, these cannot be separated

query?title=XSLT

query?title=Markup

Full-Text Search

• Searching through all fields of an entry
  • this searches through the XML, not the HTML (surprises are possible)
• Search strings are case insensitive and whitespace is normalized
  • relaxed spacing and casing is appropriate for full-text searches

query?fulltext=XDBMS

query?fulltext=xml+modeling

Entry Types

• Bibliographic entries can have different types
  • article, book, incollection, inproceedings, manual, misc, proceedings, report, thesis
  • the special value all selects all entry types

query?type=book

query?type=thesis

Date Ranges

• Queries can be limited to a date range
  • start date for the lower bound of the queried entries
  • end date for the lower bound of the queried entries
• Dates can use different formats for specifying bounds
  • 2006 specifies a year
  • 2006-09 specifies a year and a month
  • 2006-09-27 specifies a year and a month and a day

query?end=1950

query?start=2006-09&end=2006-09

Simple Server-Side Scripting

• PHP requires very little coding effort and environment
• Additional classes can be used to import new functionality
• Snoopy allows access to URIs
• PHP may be used as an intermediate in Web-based Services

Assignment 3

• Assignment 2 asked you to build an interface
  • some of the suggestions are now available in a bibliography query service
• Assignment 3 asks for making it work
  1. create a form for using the bibliography query service
  2. create an intermediary PHP for handling the form submission
  3. augment the intermediary PHP with additional functionality
• Assignment 4 will move to XML processing
  • use a Web-based services which returns XML instead of HTML