#!/usr/bin/env python

import sys



class Mag_BSoup(object):

    urlsExtracted=[]

    allh3tags=[];

    

#    import re

#    url_re=re.compile(r'<a href="(.*)" onmousedown=')

    def __init__(self,mag_file):     # mag_file is string, not file obj

        from BeautifulSoup import BeautifulSoup

        soup = BeautifulSoup(mag_file)

        self.__class__.allh3tags=soup('h3')        # shd be 13, 1st 10 are bookmark tags, last 3 are spurious

        self.__class__.allh3tags=self.__class__.allh3tags[0:10]



    def selecth3tag(self):

        for thish3tag in self.__class__.allh3tags:

#            urlextracted=self.__class__.url_re.match(string(h3tag) )   #TypeErr: tag is not str, can't be fed to regex

             print thish3tag.aTag['href']

             self.__class__.urlsExtracted.append(thish3tag.aTag.href)

#######    Reddit  ###########

class Red_BSoup(object):

    allcolspan3=[]

    urlsExtracted=[]

    

    def __init__(self,red_file):

        from BeautifulSoup import BeautifulSoup

        soup = BeautifulSoup(red_file)

        self.__class__.allcolspan3=soup.findAll(name='td',attrs={colspan:"3"})  

        

def build_dict(filename):

    ''' it's just too annoying trying to get Python SYCK or other YAML parsers to work'''

    import re

    blank_line=r'--- '      ## shd only be 1st line

    blankline_re=re.compile(blank_line)

    dict2return={}

    for inp_line in open(filename, "r"):

        if blankline_re.match(inp_line): continue

        (url, numoccur)=inp_line.split(r': ')

        dict2return[url]=int(numoccur)

    return dict2return

                

if __name__ == "__main__":

    import os,fileinput, glob

    [pyurls,rburls,railsurls]=map(build_dict, ["pyurls.yml","rburls.yml","railsurls.yml"])

    sys.stdout.write("Len of py, rb URLs dicts: " +repr(len(pyurls.keys()))+", "+repr(len(rburls.keys()))+"\n")

    sys.stdout.write("Len of rails urls dict: " +repr(len(railsurls.keys()))+"\n")

#    pyurls=pyurls.keys();       rburls=rburls.keys();       railsurls=railsurls.keys();

    allurls=pyurls.keys() + rburls.keys() + railsurls.keys()

                                                        ## EDIT "cwd" as neces

#    os.chdir("C:\\kacc\\pyrb\\magnol_pages")                    

    os.chdir("C:\\kacc\\pyrb\\reddit_pages")

    for magfilename in glob.glob("rub*.htm*"):      ## EDIT filename glob as neces

        magfilehand=open(magfilename)

        ent_magfile=magfilehand.read()        

        magfileparsed=Mag_BSoup(ent_magfile)

        magfileparsed.selecth3tag()