#! /usr/bin/env python2.2

# This is an ugly little script for automatically grabbing all the datasheets
# for avr devices from atmel's website. Summary doc files are skipped.

# I don't intend for this to be robust in any way.

import os, re

non_verbose = ''
#non_verbose = '-nv'

get_page = 'wget %s -O - http://www.atmel.com/atmel/products/prod200.htm' %(non_verbose)

pg = os.popen(get_page).read()

re_body = re.compile (r'<!--START OF BODY COPY-->(.*)<!--END OF BODY COPY-->',re.S)
body = re_body.search(pg).group(1)

# For debugging
#open('prod200.html','w').write(body)

re_file = re.compile (r'<a href=/atmel/acrobat/(doc[0-9]*\.pdf)>.*?<b>(.*?)</b>(.*?)<br>',re.S)
re_rm = re.compile (r'\s*\(.*?\)') # for removing parens

for mo in re_file.finditer(body):
	file = mo.group(1)
	device = mo.group(2)
	desc = mo.group(3)

	ofile = re_rm.sub('', device.strip())
	ofile = ofile.replace(' ', '_')
	ofile = ofile.replace('.', '')
	ofile = ofile.replace('/', '_')
	ofile = ofile + '-' + file
	url = 'http://www.atmel.com/atmel/acrobat/%s' % (file)

	cmd = 'wget -O %s %s' % (ofile, url)

	os.system (cmd)