from mechanize import Browser import requests from bs4 import BeautifulSoup, Comment import time from datetime import datetime import csv import os from py_db import db import argparse import ast db = db('personal') base_path = os.getcwd() br = Browser() br.set_handle_robots(False) br.set_handle_referer(False) br.set_handle_refresh(False) br.addheaders = [("User-agent", "Mozilla/5.0 (iPad; CPU OS 6_0 like Mac OS X) AppleWebKit/536.26 (KHTML, like Gecko) Version/6.0 Mobile/10A5355d Safari/8536.25"), ("Accept","text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8"), ("Keep-Alive","115"), ("Accept-Charset","ISO-8859-1,utf-8;q=0.7,*;q=0.7") ] headers = {"User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_11_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/50.0.2661.102 Safari/537.36"} def initiate(): url = "https://www.8a.nu/scorecard/andy-the-eagle-god/boulders/?AscentClass=0&AscentListTimeInterval=0&AscentListViewType=1&ListByAscDate=1&GID=4e15446b352fd1a7852da8accd84d52c" table_name = "boulders_completed"
import requests import urllib import csv import os import sys from time import time, sleep from py_data_getter import data_getter from py_db import db db = db('nba_shots') start_time = time() getter = data_getter() url = 'http://stats.nba.com/stats/commonallplayers' print "\n\nscraper_players.py" print "-------------------------" parameters = { 'IsOnlyCurrentSeason': 0, 'LeagueID': '00', 'Season': '0000-00', } json = getter.get_url_data(url, "json", nba=True, params=parameters) if json is None: sys.exit('\n\n\nNo data acquired') else:
from py_db import db import requests from bs4 import BeautifulSoup, Comment import argparse from time import time,sleep db = db('CBBSN') def initiate(): start_time = time() table = 'draft_pick_analysis' # db.query("TRUNCATE TABLE %s;" % (table)) # db.conn.commit() for yr in range(2000, 2019): print '\n', yr for rnd in range (1,51): print '\t', yr, '- round', rnd url = "https://www.baseball-reference.com/draft/?year_ID=%s&draft_round=%s&draft_type=junreg&query_type=year_round&from_type_jc=0&from_type_hs=0&from_type_4y=0&from_type_unk=0" % (yr, rnd) entries = process_draftRound(yr, rnd, url) if entries not in ([], None): raw_input(entries) # db.insertRowDict(entries, table, replace=True, insertMany=True, rid=0) # db.conn.commit()
from py_db import db import pandas as pd import numpy as np import matplotlib.pyplot as plt import seaborn as sns from scipy.stats import linregress import argparse import csv import NSBL_helpers as helper import os # Investigating how well zips projects babip onto the sim db = db('NSBL') def initiate(): path = os.getcwd() + '/' for hb in ('hitters', 'pitchers'): zips_list = [] observed_list = [] process(zips_list, observed_list, hb) plot(zips_list, observed_list, path, 'zips', 'observed', 'babip', hb) def process(zips_list, observed_list, hb): if hb == 'hitters':
import argparse from time import time import csv import os from py_db import db db = db("nba_shots") def initiate(): start_time = time() print "\nexporting to .csv" for statType in ('Player', 'Team', 'PlayerCareer'): for rangeType in ('Reg', 'Pre', 'Post'): print '\t', statType, rangeType if statType == 'PlayerCareer': isCareer = True dataType = 'Player' else: isCareer = False dataType = statType export_table(dataType, rangeType, isCareer=isCareer) end_time = time() elapsed_time = float(end_time - start_time) print "\n\nNBA table_exporter.py" print "time elapsed (in seconds): " + str(elapsed_time) print "time elapsed (in minutes): " + str(elapsed_time / 60.0)
from py_db import db #Creates MySQL tables for the mlb_prospects database #This is an initialization script, and only needs to be run once, prior to running every other script db = db('mlb_prospects') q = """ """ db.query(q)
import sys import os import csv import argparse from datetime import timedelta, date, datetime import smtplib from email.MIMEMultipart import MIMEMultipart from email.MIMEText import MIMEText from py_db import db db = db('craigslist') key_file = os.getcwd() + "/un_pw.csv" key_list = {} with open(key_file, 'rU') as f: mycsv = csv.reader(f) for row in mycsv: un, pw = row key_list[un] = pw def generate_body(type): mesg = '' sub = 'New Apartment Listings [%s]' % (str(date.today())) i = 1 ids = [] for cat_filt in (1, 0): for room_type in ('apartment', 'room'): if room_type == 'room':
import argparse from time import time import csv from xlsxwriter.workbook import Workbook import os from py_db import db db = db("mlb_prospects") year = 2021 def initiate(): start_time = time() print "\n\ndeleting _master_prospects" del_qry = """DROP TABLE IF EXISTS _master_prospects;""" del_query = del_qry db.query(del_query) db.conn.commit() print "\ncreating _master_prospects" query = "CREATE TABLE _master_prospects" for yr in range(2013, year+1): query += process_prospects(yr) print "writing _master_prospects" query += ";" db.query(query) db.conn.commit() print "\nupdating tables" update_tables(year)
import argparse from time import time import csv import os from py_db import db db = db("NSBL") def initiate(year): start_time = time() print "\nexporting to .csv" export_current_bracket(year) export_current_probabilities(year) end_time = time() elapsed_time = float(end_time - start_time) print "\n\nexport_playoff_brackets.py" print "time elapsed (in seconds): " + str(elapsed_time) print "time elapsed (in minutes): " + str(elapsed_time / 60.0) def export_current_bracket(year): print "\t current bracket" qry = """SELECT year, series_id, team, opponent, series_wins, series_losses, CONCAT( ROUND(100*(team_seriesProb),1), "%%") AS win_series, CONCAT( ROUND(100*(team_in3),1), "%%") AS in3,