» Forum

The jscoverage.json file should always have the source in it. You can see this in my code post:

http://siliconforks.com/forum/viewtopic.php?p=401#p401

... you can see it is reporting the source for each file in there.

Ok. Problem of not viewing source code is solved by including source code lines in json object. But when test uses large number of java script files(500-600 in mu case), it results in OOM memory while selenium.geteval call. In general my test used 500-600 js files.

So, do jscoverage have any recommendations for java heap size settings OR it be solved by changing jscoverage javascript function?

OK, that is a lot of files. Probably if there are that many files then using getEval is not the best way to get the source code.

You could do one of two things to view the source from the report:

1. Write the "source" to the jscoverage.json file from your Java code (not the JavaScript code). What you would have to do is read each .js file (the original, uninstrumented .js file) and write out the lines as a JSON array of strings. Keep in mind this will create a huge jscoverage.json file, though.

2. You could do what you were suggesting above, modify the jscoverage.js file (in the report directory) to get the source code for each .js file when it is viewed. Although if you do that, you may run into web browser security restrictions if you try to use a file:/// URL.

Ed,

Thanks a lot all the way.
I am able to get complete code coverage report for my application.

Regards,
Amit