Skip to main content
  • EN

I have a workspace that outputs fewer features when run from Server than when it is run from Desktop. The log files are ~2700 lines long so finding out exactly where they differ from each other is quite hard.

 

I would like to feed the log files to a diff & compare program like WinMerge but the number of inconsequential differences between log files are too high. Even after removing timestamps I'm still getting hundreds of differences.

 

How to search/parse/transform/rewrite the log files so only the notable differences are apparent? For example it doesn't matter that the path of DLLs are in different locations (in this scenario, in others it will be). It also doesn't matter that hundreds of HttpCaller download requests differ by a few milliseconds and/or 1 byte. However it very much matters the total number of features written is 2654 instead of 2658.

 

Yeah the kinds of lines you're interested in really are the ones where you have

Deaggregator_4 DEAGGREGATED Splitter (TeeFactory): Cloned 130 input feature(s) into 520 output feature(s)

You can use FME if you like to read the log files with the CSV reader (using | as the separate chartacter). You can test for lines which contain the string "input feature". 

You can then try and sort the data alphabetically and write them out or even try to use a feature joiner or Matcher transfomers to identify unique lines. 

 

Another thing which might be easier is to just throw in a few extra logger transformers into parts of the workspace where you suspect the issues might be coming from. 

 

Yeah the kinds of lines you're interested in really are the ones where you have

Deaggregator_4 DEAGGREGATED Splitter (TeeFactory): Cloned 130 input feature(s) into 520 output feature(s)

You can use FME if you like to read the log files with the CSV reader (using | as the separate chartacter). You can test for lines which contain the string "input feature". 

You can then try and sort the data alphabetically and write them out or even try to use a feature joiner or Matcher transfomers to identify unique lines. 

 

Another thing which might be easier is to just throw in a few extra logger transformers into parts of the workspace where you suspect the issues might be coming from. 

 

Thanks! I've added that to the mental toolbox. I think using the same approach with Excel is probably faster and more interactive though (Open .log >> Data >> text to columns >> format as table >> filter).

 

Maybe someone from Safe tech support will chime in with what they do, as they must have to dig through log files all the time. ;-)

Reply


Community Stats

30,162
Posts
114,483
Replies
37,810
Members
Terms & ConditionsCookie settings
A product of
Safe Software respectfully acknowledges that we live, learn and work on the traditional and unceded territories of the Kwantlen, Katzie, and Semiahmoo First Nations.