Description
Simple HTTP Client
When your simple HTTP client starts, it should ask the user to input an URL from the keyboard, for example,
“http://www.ece.ust.hk/~eetsang/index.html” (quotation marks excluded). Then the client extracts the
server hostname part from the URL, establishes a socket connection to the HTTP server, uses the GET method to
fetch the corresponding HTML base file from the server, prints the server-returned HTTP response on the screen,
and saves the HTML file locally if and only if the response status code is 200.
Detailed requirements:
1. You should manually implement the whole procedure, including socket establishment, sending GET requests
to the server, and receiving the raw data from the server. Every step above counts for some points.
a) You are not allowed to use or import any pre-implemented modules/packages for HTTP, such as the
http.client module in Python.
b) Only the external modules/packages that have been used in the first programming assignment can be
imported for this programming assignment (e.g., the socket module in Python).
c) You can still import external modules/packages not for socket programming purposes (e.g., to save files to
the operating system).
ELEC 4120: Computer Communication Networks
PA 2
Page 2 of 2
2. You only need to fetch the user-input HTML base file (not necessarily ending with “.html”, such as
“http://www.ece.ust.hk/ece.php”). Other objects embedded in the base file, such as images, are not
required to be downloaded. If the file name is not specified (e.g., “http://www.ece.ust.hk/”), the client
just reports this to the HTTP server (i.e., “GET /” for “http://www.ece.ust.hk/”).
3. The client should print the entire server-returned HTTP response on the screen, but not the content of the
HTML file. Here are two examples:
[Example 1] The client was requesting “http://www.ece.ust.hk/~eetsang/index.html” (quotation
marks excluded).
HTTP/1.1 200 OK
Date: Fri, 02 Nov 2018 09:36:21 GMT
Server: Apache/2.2.25 (Unix)
Last-Modified: Wed, 28 Mar 2018 07:38:52 GMT
ETag: “720e48-10c4-568741935bb00”
Accept-Ranges: bytes
Content-Length: 4292
Connection: close
Content-Type: text/html; charset=UTF-8
[Example 2] The client was requesting “http://www.ece.ust.hk/” (quotation marks excluded).
HTTP/1.1 302 Found
Date: Fri, 02 Nov 2018 09:30:45 GMT
Server: Apache/2.2.25 (Unix)
Location: http://www.ece.ust.hk/
Content-Length: 287
Connection: close
Content-Type: text/html; charset=iso-8859-1
4. The stored HTML file should be a complete HTML file, where the server-returned HTTP response has been
removed.
5. After fetching an HTML file, your client should remain active, waiting for the user to input the next URL. The
client exits when the user types “bye” (quotation marks excluded).
6. A user may request two HTML files with the same name but from different URLs. For example, a user may
sequentially request http://www.ece.ust.hk/~eetsang/index.html and
http://www.ece.ust.hk/~eeknlau/index.html. In this case, the client should not overwrite the first
HTML file when saving the second HTML file. (A simple solution is to attach a timestamp to or include the
entire URL in the name of your saved file.)
7. A user may omit the “http://” part when inputting the URL. In this case, the URL should still be considered
as legal.