Added test for splitted sequence

[r-mllr]

Added simple tests, which guarantees different md5 checksums for non-identical sequences, which are identical after concatination.

[sjanssen2]

Hi @r-mllr thanks for your PR. I am not sure if I understand exactly what you expect the method to do. Frank brought up the example of two files:

>Seq1
ACGT
>Seq2
ACGT

and

>Seq3
ACGTACGT

which would get the same md5 sum assigned. This is exactly what I want, since I think of fasta files as results of assemblies. The file above would represent an older assembly with two contigs / scaffolds, the lower file is a revision in which people found a way how to resolve the gap between Seq1 and Seq2. The genomic content itself remains unchanged - therefore it is intended that the md5sum also does not change.

[r-mllr]

Ah, then, I misunderstood the consept. I thought the checksums should not be equal.

Then, you should change

self.assertNotEqual
to
self.assertEqual

The tests should fail, then

[sjanssen2]

If this method is of general interest, we could add a parameter to control the behaviour.

[r-mllr]

not sure. I convert the pull request to a draft pull request for the case it comes up in the future