Closed martinhpedersen closed 8 years ago
I am currently running a series of tests compressing random data with both implementation. The resulting compressed data are compared between the two implementations in an attempt to find any unintentional diffs.
The tests compresses millions of blobs of random data (max 500kB) with an estimated runtime of 3 hours. I will post the results here when the tests are done.
All random test blobs passed, except a zero length blob: The old implementation does not handle the empty data case correct, but the new one does. :+1:
Closes #41