ECE 4894 Project

A Simple HTML Parser for WindowsCE

Yannick Fortin
Murat Guler
Jian Liu

Project Description:

This project involved the programming of a simple HTML parser application with minimal functionality for WindowsCE. The parser was designed to extract and display only the text content of an HTML document. Theoretically, this would keep load times low for Web pages for those who need information quickly loaded and do not care about the visual or multimedia content of the page. However, Microsoft and Netscape have nothing to fear for the time being...

Member Contributions:

The application consists of three key components: the parser, the the GUI, and the network interface. The three components were developed separately and then integrated into the single parser application once each was complete.

Read about Yannick's contribution

Read about Murat's contribution

Read about Jian's contribution

Program Quirks:

Because we concentrated on successfully integrating the application, parser, and network code together, some aspects of the program had to be given minimal attention and therefore suffered a less-than-optimal implementation. Some of the more obvious ones are listed as follows:

The scroll bar behaves erratically after opening a URL. The problem is usually fixed when the application is closed and restarted.
The CANCEL button on the URL dialog box is not associated with a function and causes the InternetOpenUrl() function to return an error condition.
The maximum HTML file size is currently fixed, as there was not enough time to implement a dynamically allocated maximum file size.
The application actually writes the received HTML code to a file for the parser to read, so the original HTML document can be viewed with a REAL browser afterwards, so this application doubles as an HTML document copier.
Despite the driving purpose of speedy access to textual information by extracting only the text, this browser application is actually SLOWER than Internet Explorer, probably due to the parses a written file instead of directly parsing the incoming internet data stream.