bugs in r2301 test_uniq_seq

[congmao wang]

when testing many times, you will find some times fail:

Testing uniq_seq -I $BP_DIR/bp_test/in/uniq_seq.in -c | sort_records -k SEQ -O $BP_TMP/congmao.uniq_seq.out ... FAIL

The command can run well, but it seems sort_records not works well

[congmao wang]

PS: When I test 10 times, seems 4 times fail, 6 times ok

[Martin Asser Hansen]

Yes, this is due to unordered perl hashes. It used to be unordered in the same way, but now (maybe newer perl version) it causes the test to fail. I need to come up with a way to sort the records internally for testing.

Martin

--
You received this message because you are subscribed to the Google Groups "biopieces" group.
To unsubscribe from this group and stop receiving emails from it, send an email to biopieces+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[congmao wang]

I think it's not caused by newer perl version, I didn't update to a newer perl version for testing purpose

[Martin Asser Hansen]

On our server the result is stable and correct, but I can reproduce the issue on my Mac where I get this:

maasha@mel:~$ uniq_seq -I $BP_DIR/bp_test/in/uniq_seq.in -c

SEQ: ATGC

SEQ_LEN: 4

SEQ_COUNT: 3

---

maasha@mel:~$ uniq_seq -I $BP_DIR/bp_test/in/uniq_seq.in -c

SEQ: GCAT

SEQ_LEN: 4

SEQ_COUNT: 3

---

So it is not a problem with the testing suite (which is in fact handling unstable key order).

So now I have inserted a sort in the code of uniq_seq and the output is now stable

maasha@mel:~/biopieces/bp_test$ uniq_seq -I $BP_DIR/bp_test/in/uniq_seq.in -c

SEQ_LEN: 4

SEQ: GCAT

SEQ_COUNT: 3

---

maasha@mel:~/biopieces/bp_test$ uniq_seq -I $BP_DIR/bp_test/in/uniq_seq.in -c

SEQ_COUNT: 3

SEQ: GCAT

SEQ_LEN: 4

---

And with this the problem should be fixed, since the testing suite is handling the key order.

bp_update && bp_test

Nasty issue - thanks for pointing it out.

Cheers

Martin

[congmao wang]

Yes, it should be fixed now, 10 times test, all pass:)

cheers,

Congmao