Web Architectures - Lecture 02 - Web Information Systems (4011474FNR)

2 December 2005
Web Information Systems
Web Architectures
Prof. Beat Signer
Department of Computer Science
Vrije Universiteit Brussel
http://www.beatsigner.com

October 3, 2014 Beat Signer - Department of Computer Science - bsigner@vub.ac.be 2
Web Information Systems
 A web information system uses web technologies
for information and service delivery
 Modern web information systems and web architectures
have to
 be extensible to cater for emerging technolgies and new forms of
interaction (e.g. multimodal interaction)
 manage heterogeneous information such as documents, structured
data, multimedia resources, semi-structured information, ...
 integrate various sources (e.g. DBs) via multi-tier architectures
 offer a notion of state to reflect the current application context
 deal with information about users and their environment (context)
 ...

Basic Client-Server Web Architecture
 Effect of typing http://www.vub.ac.be in the broswer bar
(1) use a Domain Name Service (DNS) to get the IP address for
www.vub.ac.be (answer 134.184.129.2)
(2) create a TCP connection to 134.184.129.2
(3) send an HTTP request message over the TCP connection
(4) visualise the received HTTP response message in the browser
Internet
Client Server
HTTP Request
HTTP Response

Web Server
 Tasks of a web server
(1) setup connection
(2) receive and process
HTTP request
(3) fetch resource
(4) create and send
HTTP response
(5) logging
 The most prominent web servers are the Apache HTTP
Server and Microsoft's Internet Information Services (IIS)
 A lot of devices have an embedded web server
 printers, WLAN routers, TVs, ...
Worldwide Web Servers, http://news.netcraft.com

Example HTTP Request Message
GET / HTTP/1.1
Host: www.vub.ac.be
User-Agent: Mozilla/5.0 (Windows NT 6.1; rv:24.0) Gecko/20100101
Firefox/24.0
Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8
Accept-Language: en-gb,en;q=0.5
Accept-Encoding: gzip, deflate
Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7
Connection: keep-alive

Example HTTP Response Message
HTTP/1.1 200 OK
Date: Thu, 03 Oct 2013 17:02:19 GMT
Server: Apache/2.2.14 (Ubuntu)
X-Powered-By: PHP/5.3.2-1ubuntu4.15
Content-Language: nl
Set-Cookie: lang=nl; path=/; domain=.vub.ac.be; expires=Mon, 18-Sep-2073
17:02:16 GMT
Content-Type: text/html; charset=utf-8
Keep-Alive: timeout=15, max=987
Connection: Keep-Alive
Transfer-Encoding: chunked
<!DOCTYPE html>
<html lang="nl" dir="ltr">
<head>
...
<title>Vrije Universiteit Brussel | Redelijk eigenzinnig</title>
<meta name="Description" content="Welkom aan de VUB" />
...
</html>

HTTP Protocol
 Request/response communication model
 HTTP Request
 HTTP Response
 Communication always has to be initiated by the client
 Stateless protocol
 HTTP can be used on top of various reliable protocols
 TCP is by far the most commonly used one
 runs on TCP port 80 by default
 Latest version: HTTP/1.1
 HTTPS scheme used for encrypted connections

Uniform Resource Identifier (URI)
 A Uniform Resource Identifier (URI) uniquely
identifies a resource
 There are two types of URIs
 Uniform Resource Locator (URL)
- contains information about the exact location of a resource
- consists of a scheme, a host and the path (resource name)
- e.g. http://wise.vub.ac.be/beat-signer/
- problem: the URL changes if resource is moved!
• idea of Persistent Uniform Resource Locators (PURLs) [https://purl.oclc.org]
 Uniform Resource Name (URN)
- unique and location independent name for a resource
- consists of a scheme name, a namespace identifier and a namespace-specific
string (separated by colons)
- e.g. urn:ISBN:3837027139

HTTP Message Format
 Request and response messages have the same format
<html>
...
</html>
HTTP/1.1 200 OK
Date: Thu, 03 Oct 2013 17:02:19 GMT
Server: Apache/2.2.14 (Ubuntu)
X-Powered-By: PHP/5.3.2-1ubuntu4.15
Transfer-Encoding: chunked
Content-Type: text/html
header field(s)
blank line (CRLF)
message body (optional)
start line
HTTP_message = start_line , {header} , "CRLF" , {body};

HTTP Request Message
 Request-specific start line
 Methods
 GET : get a resource from the server
 HEAD : get the header only (no body)
 POST : send data (in the body) to the server
 PUT : store request body on server
 TRACE : get the "final" request (after it has potentially been modified by proxies)
 OPTIONS : get a list of methods supported by the server
 DELETE: delete a resource on the server
start_line = method, " " , resource , " " , version;
method = "GET" , "HEAD" , "POST" , "PUT" , "TRACE" ,
"OPTIONS" , "DELETE";
resource = complete_URL | path;
version = "HTTP/" , major_version, "." , minor_version;

HTTP Response Message
 Response-specific start line
 Status codes
 100-199 : informational
 200-299 : success (e.g. 200 for 'OK')
 300-399 : redirection
 400-499 : client error (e.g. 404 for 'Not Found')
 500-599 : server error (e.g. 503 for 'Service Unavailable')
start_line = version , status_code , reason;
version = "HTTP/" , major_version, "." , minor_version;
status_code = digit , digit , digit;
reason = string_phrase;

HTTP Header Fields
 There exist general headers (for requests and
responses), request headers, response headers, entity
headers and extension headers
 Some important headers
 Accept
- request header definining the Multipurpose Internet Mail Extensions (MIME)
that the client will accept
 User-Agent
- request header specifying the type of client
 Keep-Alive (HTTP/1.0) and Persistent (HTTP/1.1)
- general header helping to improve the performance since otherwise a new
HTTP connection has to be established for every single webpage element
 Content-Type
- entity header specifing the body's MIME type

HTTP Header Fields ...
 Some important headers ...
 If-Modified-Since
- request header that is used in combination with a GET request (conditional
GET); the resource is only returned if it has been modified since the specified
date

MIME Types
 The MIME type defines the request or response body's
content and is used for the appropiate processing
 Standard MIME types are registered with the Internet
Assigned Numbers Authority (IANA) [RFC-2045]
mime = toplevel_type , "/" , subtype;
MIME Type Description
text/plain Human-readable text without formatting information
text/html HTML document
image/jpeg JPEG-encoded image
... ...

HTTP Message Information
 Various tools for HTTP message logging
 e.g. HttpFox add-on for Firefox browser
 Simple telnet connection
 Until 1999 the W3C has been working on the HTTP Next
Generation (HTTP-NG) protocol as a replacement for
HTTP/1.1
 never introduced
 recently some work on HTTP/2.0
telnet wise.vub.ac.be 80 (press Enter)
GET /beat-signer HTTP/1.1 (press Enter)
Host: wise.vub.ac.be (press Enter 2 times)

Proxies
 A web proxy is situated between the client and the server
 acts as a server to the client and as a client to the server
 can for example be specified in the browser settings; used for
- firewalls and content filters
- transcoding (on the fly transformation of HTTP message body)
- content router (e.g. select optimal server in content distribution networks)
- anonymous browsing, ...
Internet
Client Server
Proxy

Caches
 A proxy cache is a special type of proxy server
 can reduce server load if multiple clients share the same cache
 often multi-level hierarchies of caches (e.g. continent, country
and regional level) with communication between sibling and
parent caches as defined by the Internet Cache Protocol (ICP)
 passive or active (prefetching) caches
Internet
Client 1
Proxy Cache Server
Client 2
1
2
2 1

Caches ...
 Special HTTP cache control header fields
 Expires
- expiration date after which the cached resource has to be refetched
 Cache-Control: max-age
- maximum age of a document (in seconds) after it has been added to the cache
 Cache-Control: no-cache
- response cannot be directly served from the cache (has to be revalidated first)
 ...
 Validators
 Last-modified time as validator
- cache with resource that has been last modified at time t uses an
If-Modified-Since t request for updates
 Entity tags (ETag)
- changed by the publisher if content has changed; If-None-Match etag request

Caches ...
 Advantages
 reduces latency and used network bandwidth
 reduces server load (client and reverse proxy caches)
 transparent to client and server
 Disadvantages
 additional resources (hardware) required
 might get stale data out of the cache
 creates additional network traffic if we use an active caching
approach (prefetching) but achieve a low cache hit rate
 server loses control (e.g. access statistics) since no longer all
requests have to be sent to the server

Tunnels
 Implement one protocol on top of another protocol
 e.g. HTTP as a carrier for SSL connections
 Often used to "open" a firewall to protocols that would
otherwise be blocked
 e.g. tunneling of SSL connections through an open HTTP port
Internet
SSL Client SSL Server
SSL
HTTP
SSL
HTTP[SSL] HTTP[SSL]

Gateways
 A gateway can act as a kind of "glue" between
applications (client) and resources (server)
 translate between two protocols (e.g. from HTTP to FTP)
 security accelerator (e.g. HTTPS/HTTP on the server side)
 often the gateway and destination server are combined in a single
application server (HTTP to server application translator)
Internet
HTTP Client FTP Server
HTTP/FTP
Gateway
HTTP
FTP

Session Management
 HTTP is a stateless protocol
 Session (state) tracking solutions
 use of IP address
- problem: IP address is often not uniquely assigned to a single user
 browser login
- use of special HTTP authenticate headers
- after a login the browser sends the user information in each request
 URL rewriting
- add information to the URL in each request
 hidden form fields
- similar to URL rewriting but information can also be in body (POST request)
 cookies
- the server stores a piece of information on the client which is then sent back to
the server with each request

Cookies
 Introduced by Netscape in June 1994
 A cookie is a piece of information that is
assigned to a client on their first visit
 list of <key,value> pairs
 often just a unique identifier
 sent via Set-Cookie or Set-Cookie2 HTTP response headers
 Browser stores the information in a "cookie database" and
sends it back every time the same server is accessed
 Potential privacy issues
 third-party websites might use persistent cookies for user tracking
 Cookies can be disabled in the browser settings

Hypertext Markup Language (HTML)
 Dominant markup language for webpages
 If you never heard about HTML have a look at
 http://www.w3schools.com/html/
 More details in the exercise and in the next lecture
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
<head>
<title>Beat Signer: Interactive Paper, PaperWorks, Paper++, ...</title>
</head>
<body>
Beat Signer is Associate Professor of Computer Science at the VUB ...
</body>
</html>

Dynamic Web Content
 Often it is not enough to serve static web pages but
content should be changed on the client or server side
 Server-side processing
 Common Gateway Interface (CGI)
 Java Servlets
 JavaServer Pages (JSP)
 PHP: Hypertext Preprocessor (PHP)
 ...
 Client-side processing
 JavaScript
 Java Applets
 Adobe Flash
 ...

Common Gateway Interface (CGI)
 CGI was the first server-side processing solution
 transparent to the user
 certain requests (e.g. /account.pl) are forwarded via CGI to a
program by creating a new process
 program processes the request and creates an answer with
optional HTTP response headers
Internet
Client Server
HTTP Request
HTTP Response
Program in
Perl, Tcl, C,
C++, Java, ..
HTML Pages
CGI

Common Gateway Interface (CGI) ...
 CGI Problems
 a new process has to be started for each request
 if the CGI program for example acts as a gateway to a database,
a new DB connection has to be established for each request
which results in a very poor performance
 FastCGI solves some of the problems by introducing
persistent processes and process pools
 CGI/FastCGI becomes more and more replaced by other
technologies (e.g. Java Servlets)

Java Servlets
 A Java servlet is a Java class that has to extend the
abstract HTTPServlet class
 The Java servlet class is loaded by a servlet container
and relevant requests (based on a servlet binding) are
forwarded to the servlet instance for further processing
Internet
Client Server
HTTP Request
HTTP Response
HTML Pages
Servlet
Container
Servlets

Java Servlets ...
 Main HttpServlet methods
 Servlet life cycle
 a servlet is initialised once via the init() method
 the doGet(), doPost() methods may be executed multiple
times (by different HTTP requests)
 finally the servlet container may unload a servlet (upcall of the
destroy() method before that happens)
 Servlet container (e.g. Apache Tomcat) either integrated
with web server or as standalone component
doGet(HttpServletRequest req, HttpServletResponse resp)
doPost(HttpServletRequest req, HttpServletResponse resp)
init(ServletConfig config)
destroy()

Java Servlet Example
 In the exercise you will learn how to process parameters etc.
package org.vub.wise;
import java.io.*;
import java.util.Date;
import javax.servlet.http.*;
import javax.servlet.*;
public class HelloWorldServlet extends HttpServlet {
public void doGet (HttpServletRequest req, HttpServletResponse res)
throws ServletException, IOException {
PrintWriter out = res.getWriter();
out.println("<html>");
out.println("<head><title>Hello World</title></head>");
out.println("<body>The time is " + new Date().toString() + "</body>");
out.println("</html>");
out.close();
}
}

JavaServer Pages (JSP)
 A "drawback" of Java servlets is that the whole page
(e.g. HTML) has to be defined within the servlet
 not easy to share tasks between web designer and programmer
 Add program code through scriptlets and markup to
existing HTML pages
 These JSP documents are then either interpreted on the
fly (Apache Tomcat) or compiled into Java servlets
 The JSP approach is similar to PHP or Active Server
Pages (ASP)
 Note that Java servlets become more and more an
enabling technology (as with JSP)

JavaScript
 Interpreted scripting language for client-side processing
 JavaScript functionality often embedded in HTML
documents but can also be provided in separate files
 JavaScript often used to
 validate data (e.g. in a form)
 dynamically add content to a webpage
 process events (onLoad, onFocus, etc.)
 change parts of the original HTML document
 create cookies
 ...
 Note: Java and JavaScript are completely different
languages!

JavaScript Example
 Please have a look at the following JavaScript tutorial to
learn some of the basic constructs (operators, control
statements, etc.)
 http://www.w3schools.com/JS/
 In the exercise session you will use JavaScript to
implement a web application
<html>
<body>
<script type="text/javascript">
document.write("<h1>Hello World!</h1>");
</script>
</body>
</html>

Java Applets
 A Java applet is a program delivered to the client side in
the form of Java bytecode
 executed in the browser using a Java Virtual Machine (JVM)
 an applet has to extend the Applet or JApplet class
 runs in the sandbox
 Advantages
 the user automatically always has the most recent version
 high security for untrusted applets
 full Java API available
 Disadvantages
 requires a browser Java plug-in

Java Applets ...
 Disadvantages ...
 only signed applets can get more advanced functionality
- e.g. network connections to other machines than the source machine
 More recently Java Web Start (JavaWS) is replacing
Java Applets
 program no longer runs within the browser
- less problematic security restrictions
- less browser compatibility issues
 Java Chess Applet Example
 http://english.op.org/~peter/ChessApp/

Exercise 2
 Hands-on experience with the HTTP protocol

References
 David Gourley et al., HTTP: The Definitive
Guide, O'Reilly Media, September 2002
 R. Fielding et al., RFC2616 - Hypertext Transfer
Protocol - HTTP/1.1
 http://www.faqs.org/rfcs/rfc2616.html
 N. Freed et al., RFC2045 - Multipurpose Internet Mail
Extensions (MIME)
 http://www.faqs.org/rfcs/rfc2045.html
 HTML and JavaScript Tutorials
 http://www.w3schools.com

References ...
 Mick Knutson, HTTP: The Hypertext Transfer
Protocol (refcardz #172)
 http://refcardz.dzone.com/refcardz/http-hypertext-transfer-
0
 W. Jason Gilmore, PHP 5.4 (refcardz #23)
 http://refcardz.dzone.com/refcardz/php-54-scalable
 Java Servlet Tutorial
 http://www.tutorialspoint.com/servlets/

2 December 2005
Next Lecture
HTML5 and the Open Web Platform

Web Architectures - Lecture 02 - Web Information Systems (4011474FNR)

More Related Content

Similar to Web Architectures - Lecture 02 - Web Information Systems (4011474FNR) (20)

More from Beat Signer (20)

Recently uploaded (20)

Web Architectures - Lecture 02 - Web Information Systems (4011474FNR)