#jofish  13 sept 2007
#
#takes as input a series of files named conferencename1998.htm, etc. in the same directory.
#these can be downloaded from the proceedings section of the ACM Digital Library
#case of the conferencename matters!
#
#generates a series of files called conferencename1998.txt with one name per line.
#doesn't change html characters (ie é) to any other representation.
#that could be fixed.  or not.

conferencename = "CSCW" #change this to your conference name

from os import listdir
filelist = listdir(".")
filelist.sort()

for filename in filelist: #looks for cscw2001.htm cscw2002.htm etc
    if filename.lower().split('.')[-1]!='htm' or conferencename not in filename: continue #ignore other files
    outfile = open(filename.split('.')[0]+'.txt','w')
    year = filename.split()[0].split(conferencename)[1].split('.')[0] #sloppy, i know, but it'll work
    print "\nyear :",year
    for line in open(filename,'r'): #Incantation to the Gods of the ACM DL who believeth not in XML begins.
        if "citation.cfm" not in line: continue                #I wish this was XML
        for phrase in line.lower().split('<tr'):                       #Remember, this is the ACM.
            phrase=phrase.lower()
            if "medium-text" not in phrase: continue           #Not, like, historians.  I feel so dirty.
            if """</strong></a><br>""" not in phrase.lower(): continue #I can't believe I'm doing this.
            #Ok, i think we've got one.  Incantation over.
            names= phrase.split("</strong></a><br>")[1].split("<br><small>")[0] #Uggg.  Still dirty.
            for fullname in names.split(","):
                name = fullname.strip().lower() #might want to take this out in the future. not for now.
                if len(name) > 50: continue; #we get some crud in CHI1983 for some reason.
                outfile.write(name + "\n"); #so we're not dealing with duplicates or sorting.
    outfile.close()