google

#!/usr/bin/python
import requests
import sys
import getopt


def main():
    term = "+".join(sys.argv[1:])
    results = google_search(term)
    for r in results:
        print 'Title: %s\nLink: %s\nDesc: %s\n' % r

def google_search(term):
    url = "http://www.google.com/search?hl=en&safe=off&q=%s" % term
    results_page = get_webpage(url)
    results_list = extract_results(results_page)
    return results_list

def get_webpage(url):
    user_agent = 'Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.7) Gecko/2009021910 Firefox/3.0.7'
    headers = {'User-Agent': user_agent,} 
    r = requests.get(url, headers=headers)
    return r.content

def extract_results(results_page):
    start_tag = '<li class="g">'
    end_tag = '</div></li>'
    indices = get_index_list(start_tag, results_page, [])
    results = []
    for i in indices:
        snippet_end = results_page[i:].find(end_tag)
        snippet = results_page[i:i + snippet_end]
        link = pull_from_snippet(snippet, '<a href="/url?q=', '&amp;')
        title = pull_from_snippet(pull_from_snippet(snippet, '<a href="/url?q='), '>', '</a>')
        formatted_title = title.replace('<b>', '\033[1m').replace('</b>', '\033[0m')
        desc = pull_from_snippet(snippet, '<span class="st">', '</span>')
        formatted_desc = desc.replace('<b>', '\033[1m').replace('</b>', '\033[0m')
        results.append((formatted_title, link, formatted_desc, ))

    return results

def pull_from_snippet(snippet, start_tag, end_tag=None):
    if not start_tag in snippet:
        return snippet
    start_index = snippet.find(start_tag) + len(start_tag)
    if end_tag and end_tag in snippet:
        end_index = snippet[start_index:].find(end_tag) + start_index
    else:
        end_index = len(snippet) - 1
    return snippet[start_index:end_index]

def get_index_list(tag, content, indices):
    if tag in content:
        index = content.find(tag)
        previous_index = indices[len(indices) - 1] + len(tag) if indices else 0
        indices.append(index + previous_index)
        return get_index_list(tag, content[index + len(tag):], indices)
    else:
        return indices


if __name__ == "__main__":
    main()