search

nikgoodley-ibboost / cakupan

Automatically exported from code.google.com/p/cakupan
0 stars 0 forks source link

Performance improvement: Only dump stats after parsing top-level node, instead of after every node #16

Open GoogleCodeExporter opened 9 years ago

GoogleCodeExporter commented 9 years ago 
When running Cakupan over a large codebase, we found it took over half an hour 
to execute.  Profiling showed this was primarily due to dumping the database to 
disk after processing each element.

With this patch, Cakupan will only dump the statistics to disk after final 
processing of a top-level element.  That gives a speedup of about 20x, 
completing the coverage testing in 2-3 minutes.

The patch also has some @Override annotations which were automatically added by 
Eclipse.  I'd be happy to provide a new patch without them if necessary.

This patch is built on top of the patch included in: 
https://code.google.com/p/cakupan/issues/detail?id=11

Original issue reported on code.google.com by sgiff...@suspectclass.com on 8 Jan 2015 at 5:06

Attachments: