[EMBOSS] Bug with file names containing a dash.

David Bauer david.bauer at bayer.com
Thu Jun 5 06:51:30 UTC 2014

Hi Charles,

this is a side effect of the automatic parsing of EMBOSS USAs.
If there is a ":" present it can be interpreted as file:entry or dbname:entry.
And if the part before the ":" stands for a database, then you can specify to search in indexed database fields by adding "-<index>".
In this case EMBOSS interprets the present directory "test" as a database (I'm not sure if this is a feature or a bug :-) ) and then assumes that "1.fa" is the name of an index where it should look for "A".


-----Ursprüngliche Nachricht-----
Von: emboss-bounces at lists.open-bio.org [mailto:emboss-bounces at lists.open-bio.org] Im Auftrag von Charles Plessy
Gesendet: 05 June 2014 07:36
An: emboss at lists.open-bio.org
Betreff: [EMBOSS] Bug with file names containing a dash.

Dear all,

I just encountered the following bug:

If a file name contains a dash, then EMBOSS will be disturbed when a directory
named as the string before that dash is present in the same directory.

echo -e '>A\nAAAAAA' > test-1.fa

cat test-1.fa 
# >A

seqret test-1.fa:A stdout
# Read and write (return) sequences
# >A

mkdir test

seqret test-1.fa:A stdout
# Read and write (return) sequences
# Error: Query 'test-1.fa:A' query field '1.fa' not defined for datatype 'sequence'
# Error: Unable to read sequence 'test-1.fa:A'
# Died: seqret terminated: Bad value for '-sequence' and no prompt


Charles Plessy, Tsurumi, Kanagawa, Japan.
EMBOSS mailing list
EMBOSS at lists.open-bio.org

More information about the EMBOSS mailing list