[Bio] / FigKernelScripts / embl2gff_addmd5.py Repository:
ViewVC logotype

View of /FigKernelScripts/embl2gff_addmd5.py

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1.2 - (download) (as text) (annotate)
Mon Dec 5 18:56:37 2005 UTC (13 years, 11 months ago) by olson
Branch: MAIN
CVS Tags: mgrast_dev_08112011, rast_rel_2009_05_18, mgrast_dev_08022011, rast_rel_2014_0912, rast_rel_2008_06_18, myrast_rel40, rast_rel_2008_06_16, mgrast_dev_05262011, rast_rel_2008_12_18, mgrast_dev_04082011, rast_rel_2008_07_21, rast_rel_2010_0928, rast_2008_0924, mgrast_version_3_2, mgrast_dev_12152011, rast_rel_2008_04_23, mgrast_dev_06072011, rast_rel_2008_09_30, rast_rel_2009_0925, rast_rel_2010_0526, rast_rel_2014_0729, mgrast_dev_02212011, rast_rel_2010_1206, caBIG-05Apr06-00, mgrast_release_3_0, mgrast_dev_03252011, rast_rel_2010_0118, mgrast_rel_2008_0924, mgrast_rel_2008_1110_v2, rast_rel_2009_02_05, rast_rel_2011_0119, mgrast_rel_2008_0625, mgrast_release_3_0_4, mgrast_release_3_0_2, mgrast_release_3_0_3, mgrast_release_3_0_1, mgrast_dev_03312011, mgrast_release_3_1_2, mgrast_release_3_1_1, mgrast_release_3_1_0, mgrast_dev_04132011, rast_rel_2008_10_09, mgrast_dev_04012011, rast_release_2008_09_29, mgrast_rel_2008_0806, mgrast_rel_2008_0923, mgrast_rel_2008_0919, rast_rel_2009_07_09, rast_rel_2010_0827, mgrast_rel_2008_1110, myrast_33, rast_rel_2011_0928, rast_rel_2008_09_29, mgrast_rel_2008_0917, rast_rel_2008_10_29, mgrast_dev_04052011, mgrast_dev_02222011, caBIG-13Feb06-00, rast_rel_2009_03_26, mgrast_dev_10262011, rast_rel_2008_11_24, rast_rel_2008_08_07, HEAD
Changes since 1.1: +17 -0 lines
Add license words.

#
# Copyright (c) 2003-2006 University of Chicago and Fellowship
# for Interpretations of Genomes. All Rights Reserved.
#
# This file is part of the SEED Toolkit.
# 
# The SEED Toolkit is free software. You can redistribute
# it and/or modify it under the terms of the SEED Toolkit
# Public License. 
#
# You should have received a copy of the SEED Toolkit Public License
# along with this program; if not write to the University of Chicago
# at info or the Fellowship for Interpretation of
# Genomes at veronika or download a copy from
# http://www.theseed.org/LICENSE.TXT.
#

#
# embl2gff writes out a gff file that does not have the md5 sum in it.
# I do not want to cut/paste bob's code for computing md5 to avoid drift
# but that code requires the full gff file to be ready to run.  By the
# time we have the full gff file, it's often a GB long.  Pulling up vi
# to change 10 bytes in the first 200 bytes of a GB files is a tedious
# waste of time; hence, this little helper utility.
#

import sys

usage = "Usage: embl2gff_addmd5 file.gff md5value"

checksumPlaceholderText = "checksum_placeholder_xxxxxxxxxxx"
md5len=32    #the length in bytes of an md5 sum

if (not len(checksumPlaceholderText) == md5len):
    print "Program error: bad checksumPlaceholderText"
    sys.exit(1)

##
# Get args
##

if not len(sys.argv)==3:
    print usage
    sys.exit(1)


fileName = sys.argv[1]
md5val   = sys.argv[2]

if (not len(md5val)==md5len):
    print "md5 checksums must be $d characters long. %s is %d characters." % (md5len, md5val, len(md5val))
    sys.exit(1)



f=open(fileName, "r+")

##
# find the little bugger
##

bytes=0
seekAddr=0
line=f.readline()
while( line ):
    n=line.find( checksumPlaceholderText )
    if (not -1 == n):
        seekAddr = bytes+n
        break
    bytes += len(line)
    line=f.readline()

if (seekAddr==0):
    print "Checksum placeholder '%s' not found in the file." % checksumPlaceholderText
    sys.exit(1)
else:
    print "Seek address for checksum is ", seekAddr


##
# Poke in the new value
##

f.seek( seekAddr )
f.write( md5val )
f.close()

print "New checksum value inserted into file succesfully."

MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3