import urllib
import string
import htmllib
import formatter

query_string = 'Cottonwood+arizona+geology+kids+winter+snowshoe'
page_URL = 'http://google.yahoo.com/bin/query?p=' + query_string
# page_URL = 'http://google.yahoo.com/bin/query?p=cottonwood+az&hc=1&hs=9'
try:
    search_page = urllib.urlopen(page_URL)
except IOError:
    print 'Error: ' + page_URL
    end
page_list = search_page.readlines()
search_page.close()
flag = 0
count = 0
link_number_line = ''
link_number_string = ''
links = open('c:\\links.txt','r+')
links2 = open('c:\\links2.txt','r+')
total_links = 0
start_point = 0
page_text = ''
def write_links(page_list):    
    refference = ''
    ln = ''
    flag = 0
    for ln in page_list:
        if string.find(ln,'Web Page Matches') <> -1:
            flag =1
        if flag == 1:
            if string.find(ln,'a href="') <> -1:
                refference = ln[string.find(ln,'a href="') + 8: string.rfind(ln,'"')]
                if string.find(refference,'srd.yahoo.com') <> -1:
                    links.write(refference[string.find(refference,'*')+1:] + '\n')
class tomParser(htmllib.HTMLParser):
    def __init__(self, formatter):
        htmllib.HTMLParser.__init__(self, formatter)

        # flag to determine if we are in an anchor tag
        self.in_anchor = 0
        self.headflag = 0
        self.bodyflag = 0
        self.titleflag = 0
        self.link_list = []
    def start_head(self, attrs):
        self.headflag = 1
        
    def end_head(self):
        self.headflag = 0
        
    def start_body(self, attrs):
        self.bodyflag = 1
        self.link_list  attrs
        
    def end_body(self):
        self.bodyflag = 0
        
    def start_title(self, attrs):
        self.titleflag = 1 
        
    def end_title(self):
        self.titleflag = 0 
        
    #def start_(self, attrs):
    #def end_(self, attrs):
       
    
    def start_a(self, attrs):
    #    """Signal when we get to an <a> tag.
    #    """
        self.in_anchor = 1
        print 'Anchor: ',attrs

    def end_a(self):
    #    """Signal when we are out of the anchor -- a </a> tag"""
        self.in_anchor = 0

    def handle_data(self, text):
        #print self.headflag,self.bodyflag,self.titleflag
        """This is called everytime we get to text data (ie. not tags) """
        if self.in_anchor:
            print text
        #if self.headflag:
        #    print text
        if self.bodyflag:
            if string.strip(text) <> '/n' or '' or chr(160):
                print string.strip(text)
                if len(string.strip(text)) == 1:
                    print ord(string.strip(text)) 

        if self.titleflag:
            print 'Title: ' , text
page_format = formatter.NullFormatter()
page_parser =  tomParser(page_format)
for ln in page_list:
    if string.find(ln,'Web Page Matches') <> -1:
        flag =1
        count = 1
    if flag == 1 and count <> 1:
        link_number_line = ln
    if flag == 1 and count == 2:
        flag = 0
        count = 0
    count = count +1
print link_number_line

for ch in link_number_line:
    if ch in string.digits:
        link_number_string = link_number_string + ch
total_links = int(link_number_string)
for c in range(1,total_links,20):
    page_URL = 'http://google.yahoo.com/bin/query?p=' + query_string + '&b=' + str(c)
    try:
        search_page = urllib.urlopen(page_URL)
    except IOError:
        print 'Error: ' + page_URL
    page_list = search_page.readlines()
    search_page.close()
    write_links(page_list)
links.seek(0)
page_URL = links.readline()  
while page_URL <> '':
    start_point = 0
    print page_URL
    try:
        search_page = urllib.urlopen(page_URL)
    except IOError:
        print 'Error: ' + page_URL
        page_URL = links.readline()
    else:    
        page_text = search_page.read()
        page_parser.feed(page_text)
        number_query = string.count(page_text,query_string[:string.find(query_string,'+')]) 
        for i in range(number_query):
            start_point = start_point + string.find(page_text,query_string[:string.find(query_string,'+')],start_point)
            info_line = page_text[(start_point - string.rfind(page_text[:start_point],'>')):(start_point + string.find(page_text[:start_point],'<'))]
            links2.write(info_line)
        page_URL = links.readline()  
links.close()
links2.close()