Quick hack for a Biopython conversion question http://mailman.open-bio.org/pipermail/biopython/2015-June/015648.html

genbank_to_table.py

from Bio import SeqIO
with open("CP008802.txt", "w") as output:
    output.write("Seqname\tSource\tfeature\tStart\tEnd\tScore\tStrand\tFrame\tAttributes\n")
    for record in SeqIO.parse("CP008802.gbk", "genbank"):
        print("Converting %s" % record.name)
        for f in record.features:
            if f.type != "gene":
                continue
            locus_tag = f.qualifiers["locus_tag"][0]
            if len(f.location.parts) > 1:
                print("What should we do for %s (compound location)? %s" % (locus_tag, f.location))
                continue
            output.write('%s\tGenBank\t%s\t%i\t%i\t0,000000\t%s\t.\tlocus_tag\t"%s"; transcript_id "%s"\n'
                         % (record.name, f.type,
                            f.location.start + 1, f.location.end, f.location.strand,
                            locus_tag, locus_tag))
print("Done")

Assuming you've saved http://www.ncbi.nlm.nih.gov/nuccore/CP008802 as a plain text GenBank format file in the current directory as CP008802.gbk, this will write CP008802.txt and print the following on screen:

$ python genbank_to_table.py
Converting CP008802
What should we do for FB03_00005 (compound location)? join{[0:6](-), [2158322:2159306](-)}
Done