Read the log file generated by Apache web server and reach to some human readable information. There are so many thing that you can conclude from the log file. But this will include following features.

Basic Features

• Provide Graphical User Interface and graphs of information
• Get page ranks and top hit pages
• Get increasing page popularity by timeline
• Get list of web sites which on which the website is popular
• Get list of keywords for which the host website is searched on search engine
• Get statistics by year, month, week, daily
• Get the type of resource downloaded eg. html file, image, audio, video
• Get broken links requested by user

Desirable Features

• Allows user to customize queries in texual format
• Provide a framework to do extension
• Predict possibility of request flooding
• Generation of html reports

The system will primarily consist of following layers

Layers of modules

Parsing and tabulation

   The raw log file generated by Apache Web Server will be parsed and then stored into a proper processable format. Natively crafted parser will be used to parse the log and will be stored in sqlite format

Extracting desired data from tabulated data

   Required data then extracted from sqlite format using sqlite interfacing APIs. User may allowed to customize at this level.

Request processing and and generation of statistical information

   Extracted data will undergo statistical processing, grouping and clssifications etc. Processes will be designed according to the features. User may allowed to customize at this level.

Ploting

   Well processed data has to be represented in proper visual format. Charting and graphing can be used for this. For larger interval histogram will be drawn and bar charts for classification. Charting and plotting will be done using library metaplotlib. User may allowed to customize at this level.

Report generation(optional)

   The report from the input log file will be generated in HTML format

Programming Language:	Python
Tools and editors:	kate, python interpreter
Predicted lines of code:	~1000
Pregress:	20%

The project will undergo tests like black box tests, performance test, acceptance test. Tests will not just be carried out at the end of the projects but during the development also. Emphasis of the test will be given on follwing parts:

• Proper classification
• GUI and component alignment
• Proper report generation

One special thing that will be tried in this project is customization. Conventional log analyzers allows only specific programmer defined inferences. But customized queries, data processing and plotting will be allowed in this project at experimental level.

Contexts:

• Measuring popularity and estimate current market for website through page hits and refferer site info
• Security : Tracing sneaky requests