从不同性质的日志文件中提取数据

Question

I am not looking for any code in specific but an idea or a head start. I am trying to design an application with a User interface in which log files can be loaded.

These log files are of different formats and types. The data to be extracted also may vary with different files. I want to have one generic application to extract data from them. Thinking of Using some AI. Neural nets, Kalman, etc.

Sample1

A, B, C, D, E
1, 2, 3, 4, 5
6, 7, 8, 9,10
 ,11,12,13,14

in the above example I want to extract only column D and A. And column D should be extracted only when column A value exists.

Basically the log files are machine generated so there is no control on their formatting.

I have thought of creating an application in which a log file can be loaded. The required areas highlighted and then based on that the other similar types of log files are processed.

Any suggestions?

Answer 1

When it comes to doing "generic" things, your best bet is to use interfaces. You will also want to use a design principle like DDD. So, you want to create an ILogFile interface. You will have different implementers of the ILogFile interface (an implementer class per log type, format or file). You will have a service (an application or domain service) that can work on an ILogFile interface to extract data from it and apply rules to it.

Although you cant control the format, I am assuming the log file's format is predictable.