get_overlapping_resume_histories.py

import csv
import json
import sqlite3
from sqlite3 import Error
import csv
import datetime

'''Assumes there is a file resumes.json in the path with the input data.

1. Load input data and flatten it into list
2. Load list into SQLITE table
3. Join table on itself to figure out any employees that have the same company 
'''

query = """SELECT 
		    a.person_id  AS a_person_id, 
		       b.person_id  AS b_person_id, 
		       a.company_id AS a_company_id, 
		       b.company_id AS b_company_id, 
		       a.start_date AS a_start_date, 
		       b.start_date AS b_start_date, 
		       a.end_date   AS a_end_date, 
		       b.end_date   AS b_end_date 
		FROM   resumes AS a 
		       LEFT JOIN resumes AS b 
		              ON a.company_id = b.company_id 
		                AND a.start_date <= b.end_date 
		                AND b.start_date <= a.end_date
		WHERE  a.person_id != b.person_id"""
def prep_flattened_data(input_file_name):
		#Flatten the data so we can load it into a sqlite DB
		with open(input_file_name, 'r') as f:
			data = json.load(f)
		flattened = []
		for x in data:
			person_id = x['person_id']
			for w in x['work_history']:
				flattened.append([w['company_id'], datetime.datetime.strptime(w['start_date'], "%Y-%m-%d"), datetime.datetime.strptime(w['end_date'], "%Y-%m-%d"), x['person_id']])


		return flattened

def load_resume_data(db_file):
    conn = sqlite3.connect(db_file)
    to_db = prep_flattened_data('resumes.json')
    conn.text_factory=str
    cur = conn.cursor()
    #cur.execute("DROP TABLE locations;")
    #veh_id,lat,lon,located_at
    cur.execute("DROP TABLE resumes;")
    cur.execute("CREATE TABLE IF NOT EXISTS resumes ('person_id' text, 'company_id' text, 'start_date' date, end_date date);")
    cur.executemany("INSERT INTO resumes (company_id, start_date, end_date, person_id) VALUES (?,?,?,?);",to_db)
    conn.commit()
    conn.close()

def select_overlapping_work_histories(db_file,query):
	conn = sqlite3.connect(db_file)
	conn.text_factory=str
	cur = conn.cursor()
	cur.execute(query)
	rows = cur.fetchall()
	print ("{} rows generated in query, first 10 rows: ".format(len(rows)))
	print([x[0] for x in cur.description])
	for row in rows[:10]:
		print(row)
	return [x[0] for x in cur.description] + rows
if __name__ == '__main__':
    load_resume_data("resumes.db")
    matching_employment_histories = select_overlapping_work_histories("resumes.db", query)