Community
Participate
Working Groups
The Photran indexer is extremely slow. I have a fortran 90 file that contains 14,115 lines and the indexer has been running on the file for over 30 minutes. This is one of 4,365 files in the code. While this is bad, what is even worse is that the entire UI is locked up while the indexer is running. I don't see any reason for this, and see it as a serious design flaw that makes Photran refactoring pretty useless for real applications.
Do you have the file open in a Fortran editor? There's a good chance it's deadlocking. Can you send/attach the file? If not, can you give a basic idea of its structure? Is it preprocessed? I agree that the indexer is slow -- this is exacerbated when there are many lines in a single file -- but it shouldn't be taking that long...
Greg, you said the underlying cause was an OutOfMemoryError, right?
Yes, that appears to be the case. At least when I reran the indexer with the heap size increased to 1GB, I didn't see the problem again.