import pandas as pd
import re

# read data from data.csv saved by the parser
class Analyzer:

    def __init__(self):
        self.skills = pd.read_csv("skills.txt",header=None,delimiter="\t")
        self.data = pd.read_csv('data.csv', index_col=0)

    """
    skills sorted by its salary
    """
    def salary(self):
        dictionary = {}
        for idx,line in self.data.iterrows():
            if line['Average'] > 0:
                skill_set = self.findSkill(line['Summary'])
                for s in skill_set:
                    pair = dictionary.get(s,(0,0))
                    dictionary[s] = (pair[0]+line['Average'],pair[1]+1)
        for key, val in dictionary.items():
            dictionary[key] = (val[0]/val[1], val[1])
        sorted_dic = sorted(dictionary.items(), key = lambda pair:-pair[1][0])
        res = ''
        for skill, salary in sorted_dic:
            res += skill + ': $' + str(int(salary[0])) + ', sample size: '+str(salary[1]) + '\n'
        return res

    """
    skills sorted by its frequency
    """
    def frequency(self):
        dictionary = {}
        for idx,line in self.data.iterrows():
            skill_set = self.findSkill(line['Summary'])
            for s in skill_set:
                dictionary[s] = dictionary.get(s,0) + 1
        sorted_dic = sorted(dictionary.items(), key = lambda pair:-pair[1])
        res = ''
        for skill, freq in sorted_dic:
            res += skill + ': ' +str(freq) + '\n'
        return res

    def findSkill(self, summary):
        summary = re.split('\(|\)|\.|; |, |\/|!| ',summary)
        return set([x for x in summary if x in list(self.skills[0])])