Bad line-breaks in long words, consider breaking at hyphenation about gff3toembl HOT 4 OPEN

According to https://docs.python.org/2/library/textwrap.html the TextWrapper class used from the Python library will by default break at hyphens.

Adding break_long_words=False might be helpful here? That gives:

FT                   /product="2-amino-4-hydroxy-6-hydroxymethyldihydropteridine
FT                   pyrophosphokinase"

This passes the ENA validation tool, although it isn't quite as strict as the Prokka Genbank wrapping. Pull request to follow...

from gff3toembl.

Could you reopen this issue as discussed on #68 please?

from gff3toembl.

It seems that break_long_words=False in Python's textwrap considers hyphenated terms as a single word, and thus will try to avoid breaking them.

That's generally fine, but we have a problem if the hyphenated term itself is about 60+ characters, meaning even when put on a line on its own once the 21 character FT indent is added it exceeds the 80 character limit.

The "about" is because things are much tighter if this is the first word as you also have the prefix /product=" to consider, while for the final word there is the extra " to include.

Sadly if we stick with the default of break_long_words=True, then Python does not seem to take advantage of hyphens when deciding where to line-break in this corner case - Python bug filed: http://bugs.python.org/issue28660

from gff3toembl.

Thanks for filing the python bug. This format is a royal pain in the modern age.

from gff3toembl.