EMBLmyGFF3-v2.3

• Remove python_version>="3.8.0" from install_requires section in setup.py that was not working properly. Use instead old fashion test e.g. if sys.version_info < (3,9): sys.exit('Sorry, Python >= 3.9 is required')
  This change fix Installation issue EMBLmyGFF3 & python version requirements #79 , fix if I can specify certain python source during python setup.py install #77
• Update test : Remove dates for comparison between expected output and result (DT and RL lines). Change taxonomy for prokka test that has recently changed in NCBI taxonomy DB.
• Move minimum python requirement from python 3.8 to python 3.9 due to solve some dependencies requirement issues
• fix #78 Not for ENA submission: Sequence too short add possibility to avoid to skip short sequences (< 100 nt)
• Change CI badge on README page

EMBLmyGFF3-v2.2

• fix #67 : --expose_translations option was not working
• Fix #64 : Update biopython to avoid Bio.Alphabet error
• Update numpy to fix security issues
• add a conda env yaml file for manual installation of dependencies via conda
• Fix #71 : legal dbxref check was not working properly due to regression (add test to avoid to meet again this problem)
• Update python to 3.8
• add test for translation
• Fix #14 again due to regression
• Fix #75 : avoid to crash when CDS does not have the phase set properly, inform the user and set it to 0 by default

EMBLmyGFF3-v2.1

• Faster (e.g. twice faster on huge files)
• Read json file only once (into EMBLmyGFF3.py) and store the info in an associative array (dict), and update the rest of the code to access the values from the dict. Previously the json file of the investigated feature was read and closed. So for 1000 exons the exon's json file was read 1000 times.
• biopython 1.77 max because after Bio.Alphabet has been removed
• require python 3.6 minimum otherwise not compatible with biopython 1.77
• #53. Fix translation line and translate other valid start codons by a Methionine
• upercase month in DT line
• Update test to reflect modifications

EMBLmyGFF3-v2

Now EMBLmyGFF3 use Python3

• update to get sorted qualifiers
• update the tests for reflecting sorted qualifiers
• version centralised in the version.py
• update shameless_plug message

EMBLmyGFF3-1.3

• Fix #24
• Fix #46
• Update of the Augustus test for travis (related to issue #46)
• Change error for NCBI taxonomy into warning (Life will be used and it is accepted by ENA)
• Now compatible with Biopython >=1.67 ((related to issue #24)
• Now compatible with bcbio-gff>=0.6.4

EMBLmyGFF3-1.2.8

Fix #42: raise an error and stop when we meet a CDS without strand. And the location is now in + strand by default
Location.py updates: code simplification + logging update
Cosmetic changes: few split long lines
Fix #41: add a warning when seqid from gff absent from fasta.
Fix #40: ZeroDivisionError (when --no_progress activated)

EMBLmyGFF3-1.2.7

Fix issue #38 where the tool crashes with a string index out of range when a sequence end by Ns.

EMBLmyGFF3-1.2.6

fix _splitWordsMax for edge cases. See issue #33 part 2.
change warn into info for qualifier having a value while none is expected

EMBLmyGFF3-1.2.5

• Better performance (Improvement when dealing with sequences: slicing + gap detection)
• Fix duplicated attributes
• Fix how to deal with non accepted dbxref database. Became a Warning not anymore an error. Fix #32, #34
• Skip sequence shorter than 100bp (not accepted by ENA)
• Update Documentation
• remove useless file (untitled.py)

EMBLmyGFF3-1.2.4

Added test-directory with test script and expected outputs
Added .travis.yml file for automatic testing
Updated test.sh to trigger travis
Synchronisation help and readme text
Documentation updates
Fix #28 translation error in minus strand
XXX locus_tag was not set automatically, while it was displayed it would.
Fix #27 accession becomes a boolean. The value will be filled by the sequence name if activated.
Fix #22 - duplicated db_xref are removed now