# This example need updating for Python 3 ... sample files not easily to hand to re-test

"""We're running Google Analytics on some areas of our website, and I'm delighted to read reports
of xxx visitors from Sweden, yyy from Romania and zzz from Argentina. But these are countries with
very different populations; I would be very interested to know how the figures stack up as a
proportion of the population. In other words, take an average city with a million people and ask
"if it's in xxxx country, how many of its people have we reached?"

By taking three files - table of top level domain names (may not be needed, but useful for tabulating
the results), a table of country name to population mappings, and our own data from Analystics, I
answered my question.

"""

# Demonstration of correlating multiple text files - web Site penetration by country.

import re

class country(object):
        def __init__(self,tlc,name = None):
                self.tlc = tlc
                self.name = name
                self.traffic = None
                self.population = None
# Setting up a static list of countries
# be careful of this technique as it's a singleton table
                country.ctable[name] = self

        def setPopulation(self,citizens):
                self.population = citizens

        def setTraffic(self,visitors):
                self.traffic = visitors

        def hitrate(self):
                # Returns "1 in xxxxx" of the poplution as a number
                # Sets up a string for no. in a city of 1,000,000 for later use
                try:
                        rate = self.population / self.traffic
                        country.millions = " {0:4.1f} ".format(1000000./rate)
                        return rate
                except:
                        country.millions = " ---- "
                        return 0

        def __str__(self):
                try:
                        # Provide some country size grading to make big countries stick out
                        stars = ". "
                        if self.population > 1000000: stars = "* "
                        if self.population > 7000000: stars = "** "
                        if self.population > 35000000: stars = "*** "

                        # Change mathematically incorrect zeros
                        hitsay = str(self.hitrate())
                        if hitsay == "0": hitsay = " n/a "

                        return stars + self.tlc + ": " + country.millions + hitsay + " [" + self.name + "]"
                except:
                        return self.tlc

        def __cmp__(this, that):
                # Redefine natural sort order to sort by penetration rate
                vr1 = this.hitrate()
                vr2 = that.hitrate()
                if vr1 == 0 and vr2 == 0: return 0
                if vr1 == 0: return 1
                if vr2 == 0: return -1
                return vr1 - vr2

        @staticmethod
        def get(name):
                # Find a country by name and return its instance variable
                return country.ctable.get(name)

country.ctable = {}

# -------------------------------------------------

# Sample record:
# BE = Belgium
# from http://www.thrall.org/domains.htm

tld_record = re.compile(r'(\w+)\s=\s(.+)')

countries = []
for record in open("tld"):
        have = tld_record.match(record)
        if have:
                code = have.group(1).lower()
                name = have.group(2)
                countries.append(country(code,name))

# -------------------------------------------------

# Sample record:
# 75 Belgium 10,839,905 January 1, 2010 0.15% Official estimate
# from http://en.wikipedia.org/wiki/List_of_countries_by_population

pop_tab = re.compile(r'\t')
bracket_trim = re.compile(r'\s*[\[\(].*')

for record in open("wid"):
        have = pop_tab.split(record)
        cname = have[1].strip()
        # Remove footnotes, etc
        cname = bracket_trim.sub("",cname)
        c = country.get(cname)
        if not c:
                # uncomment next 2 lines in debug
                # print "[wid] failed on", cname, "((", have[1], "))"
                # print have[2]
                continue
        # remove commas within population numbers
        c.setPopulation(int(have[2].replace(",","")))

# -------------------------------------------------

# Sample record:
# 22. Belgium 434 1.30 00:00:40 91.24% 86.64%
# From Google Analytics

ana_tab = re.compile(r'\t')

for record in open("gad"):
        have = ana_tab.split(record)
        cname = have[1]
        c = country.get(cname)
        if not c:
                # print "[gad] failed on",cname
                continue
        c.setTraffic(int(have[2].replace(",","")))

# ----------------------------------------------------

# Now let's use the data. Sorry it's been made so easy!

countries.sort()

for c in countries:
        print c

# ------------------- Sample Output
#
# wizard:anaproj graham$ python visitors | head -20
# . is: 81.4 12291 [Iceland]
# * fi: 75.8 13185 [Finland]
# ** hk: 75.0 13327 [Hong Kong]
# * si: 71.0 14090 [Slovenia]
# ** se: 68.9 14505 [Sweden]
# *** uk: 66.5 15037 [United Kingdom]
# ** ch: 56.8 17606 [Switzerland]
# . mc: 55.7 17940 [Monaco]
# . li: 55.3 18078 [Liechtenstein]
# ** il: 53.3 18747 [Israel]
# * ee: 50.9 19649 [Estonia]
# * dk: 49.8 20073 [Denmark]
# * no: 48.0 20836 [Norway]
# . lu: 46.9 21326 [Luxembourg]
# ** nl: 45.6 21952 [Netherlands]
# . mt: 43.1 23200 [Malta]
# * ie: 42.3 23614 [Ireland]
# ** be: 40.0 24976 [Belgium]
# * sg: 39.7 25163 [Singapore]
# * lt: 37.2 26854 [Lithuania]
# wizard:anaproj graham$