#!/usr/bin/env python

import subprocess
import sys

# 14 Dec. 1998


# given a set of names or accessions, create a file
# containing the corresponding genbank entries
#
#  $1 file containing LOCUS names or ACCESSION numbers
#  $2 file to contain all of the GenBank entries

# From P. Rioux, Dec. 14, 1998:
# NCBI moved all Entrez access function to their WEB server,
# and the old TCP server is no longer maintained (I think it just reroutes
# the queries to the web server). Since the web server is stateless, this
# forces a tool such as nclever to reconnect all the time. At least, that's
# what I think is happening, I'm not sure. In the meantime, what you can
# do is NOT use trunctation mode for your queries. The server side is
# supposed (according to them) do the truncation for you, if necessary.
# This means that at the beginning of your nclever scripts, insert the
# command
#
#    option notruncation

p = subprocess.Popen(['nclever', '-b'], stdin=subprocess.PIPE)
p.stdin.write("option notruncation\n")

# This is allowed by the documentation, but
# doesn't work.
#p.stdin.write("option noTruncationMode\n")

p.stdin.write("database nucleotide\n")
p.stdin.write("report GenBank\n")

file_h = open(sys.argv[1], 'r')
for line in file_h:
    for name in line.split():
        p.stdin.write("accession " + name + "\n")
file_h.close()

p.stdin.write("union all\n")
p.stdin.write("save $2  all\n")
p.stdin.write("quit\n")
p.stdin.close()
p.wait()