GitLab

Note that germlines.data is not used now. It will change soon.

Corrects bug introduced by 59a1bcf6.

Dd2 and Dd3 may play interesting roles.

096ab901 unleashes the power of 1a05fe6a and 422964cc.

core/fasta.{h,cpp}: default constructor also takes extract_field and extract_separator, factorize constructors

Even if we start from an empty Fasta, we should remember these values.

The constructor now uses the add() method which, itself, relies on the >> operator which has been overloaded.
Unit tests added for the add() method.

C++ code should prefer cstdlib over stdlib.h

One could use constructor delegation for an even better factorisation but it
is C++11 only.

(Now only the last option of each is really stored and indexed, see 1a05fe6a.)

One may want to use several V (or D, or J) files for a unique germline, for example with TRDV + TRDD(D2).
(Now only the last file is really stored and indexed, but this will change soon.)

The CMD_SEGMENT scenario now tries the germlines in the multigermline (until the read is segmented).

On the default file, Vidjil now returns some clones even with default options.

As KmerAffectAnalyser is only used with KmerAffect, we do not need to keep the templates.
The code should be easier to read for C++ newbies such as the author of this commit.

KmerStringAffect is no more used in the code.
If one day we would like to "store more things" in the affectations,
we could still extend KmerAffect with a pointer to something else.

Methods are tested in specific conditions:
* at first position for countBefore
* at last position for countAfter
* when the affectation doesn't exist in the sequence

A hard-coded sequence should not be given to the constructor, instead it must be stored into a variable.

Makes easier to compare two affect_infos structures (notably for unit tests)

Testing the function was not useless.

Sequence is shorter than the seed