jdt2famix - logging problems

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

jdt2famix - logging problems

Tudor Girba-2
Hi,

I added a new feature: explicit logging of problems encountered during import.

These problems do not stop the import, but they do lead to loss of information quality in the model. These problems are typically due to missing jars in the classpath. However, if you know which parts are problematic, you get a chance of fixing them.

To give you an example, entries might look like this:

unresolved type declaration - TestPlatform - /Users/girba/Work/Code/jdt2famix/tmp/guava/guava-tests/test/com/google/common/math/TestPlatform.java - line 21
unresolved method declaration - suite - /Users/girba/Work/Code/jdt2famix/tmp/guava/guava-tests/test/com/google/common/primitives/ByteArrayAsListTest.java - line 51

For example, in the case of guava, importing just the checked out code, gives us 419 problems, most of which were method declaration problems (due to unknown return types). However, after downloading all jars from the maven dependencies, we got only 47 problems (only type declaration problems, but no more method declaration problems). Interestingly, guava is made out of several sub-projects, and these problems appear if we import the overall project. But, if we import only a subproject at a time, the problems do not appear anymore, so likely the issue is related to conflicting jars, or duplicated sources.

The point is that having an insight on what problems are encountered can provide a hint of how to act to improve the quality of the resulting model.

The logging support is quite rudimentary right now, and I will likely build on this, but it already offers an added value.

Cheers,
Doru


--
www.tudorgirba.com
www.feenk.com

"Reasonable is what we are accustomed with."

_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.list.inf.unibe.ch/listinfo/moose-dev