#! /usr/bin/env python

import urllib, re, subprocess, zipfile, time
from itertools import izip

disclaimer = """
Project Euler is protected under
Attribution-Non-Commercial-Share Alike 2.0 UK: England & Wales
"""

prefix = "http://projecteuler.net/index.php?section=problems&id="
problems = range(1, 203)
suffix = ''

#extras = ['http://projecteuler.net/project/' + i for i in [
#            'names.txt','words.txt', 'poker.txt', 'cipher1.txt',
#            'triangle.txt', 'keylog.txt', 'matrix.txt', 'roman.txt',
#            'sudoku.txt', 'base_exp.txt', 'triangles.txt', 'sets.txt',
#            'network.txt']]

def source_html():
    "generates original sources"
    prob_names = ('%s%i%s' % (prefix, i, suffix) for i in problems)
    return (urllib.urlopen(name).read() for name in prob_names)

def tag_extract(text, start, end):
    return re.search('%s.*?%s' % (start, end), text).group()

def prob_html(body):
    "extracts relivant part of full source"
    open_tag  = '<div class="problem_content">'
    close_tag = '</div><br />'
    body = body.replace('\r', '')
    body = body.replace('\n', '')
    return tag_extract(body, open_tag, close_tag)

def html_to_text(html):
    "just to make this hard to port"
    html_render = 'elinks -dump -force-html'
    p1 = subprocess.Popen(['echo', '"%s"' % html], stdout=subprocess.PIPE)
    p2 = subprocess.Popen(html_render.split(), stdin=p1.stdout, stdout=subprocess.PIPE)
    return p2.communicate()[0].strip().strip('"')

def textify(s):
    "replaces commonly used images with strings"
    r = {"<img src='images/symbol_times.gif' width='9' height='9' alt='' border='0' style='vertical-align:middle;' />" : ' * ',
         "<img src='images/symbol_lt.gif' width='10' height='10' alt='' border='0' style='vertical-align:middle;' />"  : ' < ',
         "<img src='images/symbol_gt.gif' width='10' height='10' alt='' border='0' style='vertical-align:middle;' />"  : ' > '}
    for img in r:
        s = s.replace(img, r[img])
    return s

def pretty_heading(n):
    l1 = 'Problem %i' % n
    l2 = '=' * len(l1)
    return '\n\n\n' + l1 + '\n' + l2 + '\n'

def pretty_time():
    return '%i-%i-%i at %i:%i' % tuple(time.localtime()[:5])

#def zip_extras():
#    f = zipfile.ZipFile('project_euler.zip', 'w')
#    for name in extras:
#        info   = zipfile.ZipInfo(name.split('/')[-1])
#        source = urllib.urlopen(name).read()
#        f.writestr(info, source)
#    f.close()

def add_extra(body, zfile):
    "adds linked text to zipfile"
    name = re.search('/\w.*?\.txt', body).group()
    name = 'http://projecteuler.net/project' + name
    info = zipfile.ZipInfo(name.split('/')[-1])
    text = urllib.urlopen(name).read()
    zfile.writestr(info, text)
    return zfile

def save_euler(path):
    f = file(path, 'w')
    z = zipfile.ZipFile('project_euler.zip', 'w')
    f.write(disclaimer + '\n\n')
    f.write('generated on ' + pretty_time() + '\n')
    for source,i in izip(source_html(), problems):
        print 'problem', i
        if '.txt' in source:
            z = add_extra(source, z)
        f.write(pretty_heading(i))
        f.write(html_to_text(prob_html(textify(source))))
    f.close()
    z.close()

save_euler('project_euler.txt')
#zip_extras()