 |  |  |
| Own website that collects, parses and documents signatures from the UserAgent string sent by browsers, web crawlers and other online tools. Tokens are recognized by specific patterns. Agents are grouped by categories. There are four top categories - Browsers, Platforms, Tools and Crawlers - each with subcategories. | ||
 | |  |
UserAgent at XtractPro started as a small personal project, implemented as a standalone C#/.NET console application with SQL-Server backend database.
I was looking at different UserAgent strings from the log files of my websites and a simple way to find out the meaning of some intriguing signature tokens. There are many tools that extract standard information about the client browser or his operating system from these string, but after spending several hours on the Internet I wasn't able to find any online catalog with better information. The idea was simple: given a UserAgent string, collected by the web server when a client gets access to a website through a request, split it in tokens and recognize through patterns documented Agents. The online catalog should have proper information about Agents and examples of UserAgent strings with their signatures within.
Here below is your own UserAgent. Click Parse to see what these tokens are:
The UserAgent string sent to any web server by my Firefox browser is "Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.2) Gecko/20100115 Firefox/3.6 GTB7.1 (.NET CLR 3.5.30729)". My browser tells the web server I'm using Firefox 3.6 on Windows Vista, and I have version 3.5 of Microsoft .NET installed on my computer. This is the information most log file analyzers will extract and will be satisfied with. But what about all other tokens? My online parser will also recognize U as USA 128-bit encryption strong security capabilities of my browser. It will tell me also where popular technical terms such as Mozilla and Gecko come from. It might probably tell me something about the misterious GTB signature too.
The full project took about 3-4 weeks to complete. I switched from a SQL-Server to a MySQL database, used an adapted CSS-based theme from xtracpro.com and built an ASP.NET website as subdomain of xtracpro.com. The database presently holds over 20,000 samples of UserAgent strings, collected from the logs of own websites, other sites on the net, and interactively, when a user parses a string. There are almost 2,000 Agents, most of them documented. Whenever a token doesn't map to an existing agent, it is marked as "unknown" pattern. Some patterns are deliberately ignored.
Looking for documented Agents, patterns and sample UserAgent strings can be done through a search by keyword, or by drill-down to a specific Agent, starting browsing with a top category and a subcategory. There are almost 30 categories and subcategories. For instance, the Platforms - Operating System subcategory contains Agents with information about the system installed on client's computer. The Linux agent is usually recognized through patterns such as "i686 linux" or the wildcard "linux*".
Whole project was built from scratch, in ASP.NET/C#, using CSS, HTML 4, JavaScript, and a MySQL database. Thousands of pages have been indexed by Google since 2007. It is still the only site of this kind on the Net, and the documented agents appear as useful to many web developers or administrators.
Subscribe and Share: 
