Following is a simple sample code that will generate the signature from your amazon credentials and add Authorization header in request to S3 server. The request to S3 service is send using linux wget command utility.

fetch_from_s3.py

#!/usr/local/bin/python2.6
import hashlib,hmac,base64
from urlparse import urlparse
from email.utils import formatdate
import sys,subprocess
 
#Update credentials
AWSAccessKeyId = "<your access key>"
AWSSecretAccessKey = "<your secret key>"
 
url = urlparse(str(sys.argv[1]))
resource_url = '/'+url.netloc.replace(".s3.amazonaws.com", "")+url.path
 
amazons3time = formatdate(usegmt=True)
 
HTTPVerb = "GET"
ContentMD5 = ""
ContentType = ""
CanonicalizedAmzHeaders = ""
CanonicalizedResource = resource_url
 
string_to_sign = HTTPVerb + "\n" + ContentMD5 + "\n" + ContentType + \
                 "\n" + amazons3time + "\n" + CanonicalizedAmzHeaders + \
                 CanonicalizedResource
 
sig = base64.b64encode(hmac.new(AWSSecretAccessKey, \
      string_to_sign, hashlib.sha1).digest())
 
auth_header = "Authorization:AWS "+AWSAccessKeyId+":"+sig
date_header = "Date:"+ amazons3time
 
cmd = "wget -S "+url.geturl()+" --header=\""+auth_header+"\" \
       --header=\""+date_header+"\""
 
p = subprocess.Popen(cmd, shell=True, stderr=subprocess.PIPE)
while True:
    outbuf = p.stderr.read(1)
    if outbuf == '' and p.poll() != None:
        break
    if outbuf != '':
        sys.stdout.write(outbuf)
        sys.stdout.flush()