635 lines
28 KiB
Python
Executable File
635 lines
28 KiB
Python
Executable File
#!/usr/bin/python3
|
|
# -*- coding: utf-8 -*-
|
|
''' Utility to load WPA results sheets into a MySQL database.'''
|
|
__author__ = 'Timothy Allen'
|
|
__email__ = 'tim@allen.org.za'
|
|
__license__ = 'MIT'
|
|
|
|
import bs4
|
|
import urllib.request
|
|
import urllib.parse
|
|
import json
|
|
import mailbox
|
|
import email
|
|
import mimetypes
|
|
import csv
|
|
import xlrd
|
|
import MySQLdb
|
|
import MySQLdb.cursors
|
|
import argparse
|
|
import datetime as dt
|
|
import dateutil.parser as dtp
|
|
import logging
|
|
import os
|
|
import re
|
|
import sys
|
|
import tempfile
|
|
from collections import defaultdict
|
|
import pprint
|
|
|
|
# Set up MySQL database, if not done
|
|
# Read Excel/ODS/CSV database into MySQL
|
|
# Check MIME attachments in email for spreadsheet, and load that ***
|
|
# Then display the data in a (separate) web application
|
|
|
|
# The user is in /etc/dovecot/users
|
|
# Password is zi6ohYae0OeYie8eegei (not that you'll ever need it)
|
|
MAILDIR = '/var/mail/virtual/aac/Maildir'
|
|
|
|
log = logging.getLogger(__file__)
|
|
pp = pprint.PrettyPrinter(indent=4)
|
|
|
|
def main():
|
|
if sys.version_info < (3, 2):
|
|
raise Exception(
|
|
'Unsupported Python version, please use at least Python 3.2')
|
|
|
|
args = parse_arguments()
|
|
rows = []
|
|
|
|
if args.calculate:
|
|
#position_calculations('Winelands Marathon')
|
|
position_calculations(args.calculate)
|
|
return
|
|
elif args.scrapeurl:
|
|
requrls = args.scrapeurl
|
|
spreadsheets = []
|
|
uniqurl = []
|
|
wpa = 'http://www.wpa.org.za/Events/DynamicEvents.asmx/BuildEventDisplay'
|
|
for year in range(2016, dt.datetime.now().year + 1):
|
|
log.debug("Finding results for {}".format(year));
|
|
args = {"WPAExtra":"True","TimeColumn":"True","entityid":"674417","selectedyear":year,"selectedmonth":0,"commissionid":"0","selectedstate":"0","categoryid":0,"themeid":"46"}
|
|
data = bytes(json.dumps(args).encode('utf8'))
|
|
req = urllib.request.Request(wpa, data=data, headers={'content-type': 'application/json'})
|
|
with urllib.request.urlopen(req) as response:
|
|
data = json.loads(response.read().decode('utf8'))
|
|
page, *_ = data.values() # get the first value
|
|
soup = bs4.BeautifulSoup(page, 'html.parser')
|
|
for event in soup.find_all('tr'):
|
|
raceinfo = dict()
|
|
link = event.find('a', href=re.compile('.xlsx?$'))
|
|
name = event.find('td', class_=re.compile('EventHeadline'))
|
|
date = event.find('td', class_=re.compile('EventDate'), string=re.compile('^\s*\d+[/-]\d+[/-]\d+'))
|
|
dist = event.find('td', class_=re.compile('Events?Distance'), string=re.compile('^\s*[\d\.,]+\s*(KM)?\s*$', flags=re.IGNORECASE))
|
|
if link is not None and name is not None:
|
|
if not link['href'] in uniqurl:
|
|
uniqurl.append(link['href'])
|
|
raceinfo['url'] = link['href']
|
|
raceinfo['event'] = name.string
|
|
raceinfo['date'] = dtp.parse(date.string, dayfirst=True)
|
|
raceinfo['distance'] = None
|
|
if dist is not None:
|
|
raceinfo['distance'] = dist.string
|
|
spreadsheets.append(raceinfo)
|
|
for race in spreadsheets:
|
|
url = race['url']
|
|
event = race['event']
|
|
''' Only parse one spreadsheet from the WPA website, from the commandline '''
|
|
isthisevent = False
|
|
for checkurl in requrls:
|
|
if checkurl and re.search(checkurl, event, flags=re.IGNORECASE):
|
|
isthisevent = True
|
|
if requrls[0] and url not in requrls and not isthisevent:
|
|
continue
|
|
if file_in_db(url):
|
|
continue
|
|
with urllib.request.urlopen(url) as response, tempfile.TemporaryDirectory() as tmpdir:
|
|
data = response.read()
|
|
urlparts = urllib.parse.urlparse(url)
|
|
filename = os.path.basename(urlparts.path)
|
|
if re.search('WALKRESULTS', filename, flags=re.IGNORECASE):
|
|
continue
|
|
log.info("Loading data from URL {}".format(url))
|
|
filepath = os.path.join(tmpdir, filename)
|
|
with open(filepath, 'wb') as fp:
|
|
fp.write(data)
|
|
try:
|
|
rows = read_spreadsheet(filepath, src=url, eventname=race['event'], eventdate=race['date'], eventdistance=race['distance'])
|
|
except:
|
|
log.warning("ERROR: Unable to load data from URL {}".format(url))
|
|
raise
|
|
else:
|
|
load_into_db(rows, event)
|
|
log.debug("\n")
|
|
|
|
elif args.input_file:
|
|
rows = read_spreadsheet(args.input_file, src=args.input_file)
|
|
log.info("Loading data from file {}".format(args.input_file))
|
|
load_into_db(rows)
|
|
|
|
else:
|
|
for message in mailbox.Maildir(MAILDIR):
|
|
counter = 1
|
|
for part in message.walk():
|
|
if part.get_content_maintype() == 'multipart':
|
|
continue
|
|
filename = part.get_filename()
|
|
ext = mimetypes.guess_extension(part.get_content_type())
|
|
if not filename:
|
|
if not ext:
|
|
ext = '.xls' # attempt to decode as a spreadsheet
|
|
filename = 'part-{:03}{}'.format(counter, ext)
|
|
counter += 1
|
|
if re.search('.xl(b|s)x?$', filename, flags=re.IGNORECASE) is not None:
|
|
with tempfile.TemporaryDirectory() as tmpdir:
|
|
filepath = os.path.join(tmpdir, filename)
|
|
with open(filepath, 'wb') as fp:
|
|
fp.write(part.get_payload(decode=True))
|
|
log.info("Loading data from file {}".format(filename))
|
|
try:
|
|
rows = read_spreadsheet(filepath, src=message['from'])
|
|
except:
|
|
log.info("Unable to load data from file {}".format(filename))
|
|
pass
|
|
else:
|
|
load_into_db(rows)
|
|
|
|
return
|
|
|
|
def position_calculations(events=None):
|
|
db = MySQLdb.connect(user='aac', passwd='saOAcCWHg4LaoSSA', db='AAC',
|
|
use_unicode=True, charset="utf8", cursorclass=MySQLdb.cursors.DictCursor)
|
|
c = db.cursor()
|
|
|
|
where = ''
|
|
wheres = []
|
|
if isinstance(events, list):
|
|
for event in events:
|
|
if event:
|
|
wheres.append('event LIKE "%{}%"'.format(event))
|
|
elif isinstance(events, str):
|
|
wheres.append('event LIKE "%{}%"'.format(events))
|
|
if wheres:
|
|
where = 'WHERE ' + ' OR '.join(wheres)
|
|
sql = 'SELECT event, date, distance FROM `results` {} GROUP BY event, date, distance'.format(where)
|
|
c.execute(sql)
|
|
#log.debug(c._last_executed)
|
|
eventlist = [e for e in c.fetchall()]
|
|
|
|
for race in eventlist:
|
|
log.debug("Recalculating position information for {}".format(race['event']))
|
|
''' Calculate total finishers per race '''
|
|
sql = 'UPDATE `results` AS r, (SELECT event, date, distance, COUNT(distance) AS finishers FROM `results` WHERE event LIKE "{}" AND date = "{}" AND distance LIKE "{}" GROUP BY event, date, distance) AS f SET r.finishers = f.finishers WHERE r.event = f.event AND r.date = f.date AND r.distance = f.distance AND r.finishers IS NULL;'.format(race['event'], race['date'], race['distance'])
|
|
c.execute(sql)
|
|
result = c.fetchall()
|
|
|
|
''' Update total finishers per sex per race '''
|
|
sql = 'UPDATE `results` AS r, (SELECT event, date, distance, sex, COUNT(sex) as sexfinishers FROM `results` WHERE event LIKE "{}" AND date = "{}" AND distance LIKE "{}" GROUP BY sex) AS s SET r.sexfinishers = s.sexfinishers WHERE r.event = s.event AND r.date = s.date AND r.distance = s.distance AND r.sexfinishers IS NULL AND r.sex = s.sex;'.format(race['event'], race['date'], race['distance'])
|
|
#print(sql)
|
|
c.execute(sql)
|
|
result = c.fetchall()
|
|
|
|
# ''' Update individual positions per sex per race '''
|
|
# sql = 'SELECT DISTINCT sex from `results` WHERE event LIKE "{}" AND date = "{}" AND distance LIKE "{}" AND sex IS NOT NULL;'.format(race['event'], race['date'], race['distance'])
|
|
# c.execute(sql)
|
|
# result = c.fetchall()
|
|
# for row in result:
|
|
# sql = 'UPDATE `results` AS r, (SELECT result_key, position, @rank := @rank+1 AS rank FROM `results` WHERE event LIKE "{}" AND date = "{}" AND distance LIKE "{}" AND sex LIKE "{}" ORDER BY position) AS c, (SELECT @rank := 0) AS n SET r.sexposition = c.rank WHERE r.result_key = c.result_key;'.format(race['event'], race['date'], race['distance'], row['sex'])
|
|
# print(sql)
|
|
# c.execute(sql)
|
|
# result = c.fetchall()
|
|
# # ''' This seems to generate a universal ranking on 1 the first time the statement is run '''
|
|
# #c.execute('SET @rank = 0;')
|
|
# #sql = 'UPDATE `results` AS r, (SELECT result_key, position, @rank := IF(@sex = sex, @rank+1, 1) AS srank, @sex := sex FROM `results` WHERE event LIKE "{}" AND date = "{}" AND distance LIKE "{}" AND sex IS NOT NULL ORDER BY sex, position) AS s SET r.sexposition = s.srank WHERE r.result_key = s.result_key;'.format(race['event'], race['date'], race['distance'])
|
|
# #print(sql)
|
|
# #c.execute(sql)
|
|
# #result = c.fetchall()
|
|
|
|
''' Update total finishers per category per race '''
|
|
sql = 'UPDATE `results` AS r, (SELECT event, date, distance, category, COUNT(category) as catfinishers FROM `results` WHERE event LIKE "{}" AND date = "{}" AND distance LIKE "{}" GROUP BY category) AS c SET r.catfinishers = c.catfinishers WHERE r.event = c.event AND r.date = c.date AND r.distance = c.distance AND r.catfinishers IS NULL AND r.category = c.category;'.format(race['event'], race['date'], race['distance'])
|
|
#print(sql)
|
|
c.execute(sql)
|
|
result = c.fetchall()
|
|
|
|
# ''' Update individual positions per category per race '''
|
|
# sql = 'SELECT DISTINCT category from `results` WHERE event LIKE "{}" AND date = "{}" AND distance LIKE "{}" AND category IS NOT NULL;'.format(race['event'], race['date'], race['distance'])
|
|
# c.execute(sql)
|
|
# result = c.fetchall()
|
|
# for row in result:
|
|
# sql = 'UPDATE `results` AS r, (SELECT result_key, position, @rank := @rank+1 AS rank FROM `results` WHERE event LIKE "{}" AND date = "{}" AND distance LIKE "{}" AND category LIKE "{}" ORDER BY position) AS c, (SELECT @rank := 0) AS n SET r.catposition = c.rank WHERE r.result_key = c.result_key;'.format(race['event'], race['date'], race['distance'], row['category'])
|
|
# print(sql)
|
|
# c.execute(sql)
|
|
# result = c.fetchall()
|
|
# # ''' This seems to generate a universal ranking on 1 the first time the statement is run '''
|
|
# #c.execute('SET @rank = 0;')
|
|
# #sql = 'UPDATE `results` AS r, (SELECT result_key, position, @rank := IF(@cat = category, @rank+1, 1) AS crank, @cat := category FROM `results` WHERE event LIKE "{}" AND date = "{}" AND distance LIKE "{}" AND category IS NOT NULL ORDER BY category, position) AS c SET r.catposition = c.crank WHERE r.result_key = c.result_key;'.format(race['event'], race['date'], race['distance'])
|
|
# #print(sql)
|
|
# #c.execute(sql)
|
|
# #result = c.fetchall()
|
|
|
|
db.commit()
|
|
return
|
|
|
|
def file_in_db(url):
|
|
db = MySQLdb.connect(user='aac', passwd='saOAcCWHg4LaoSSA', db='AAC',
|
|
use_unicode=True, charset="utf8")
|
|
c = db.cursor()
|
|
|
|
''' Check for duplicate values by DATE and POSITION and RACE and EVENT '''
|
|
sql = 'SELECT COUNT(*) FROM `results` WHERE source LIKE %s'
|
|
c.execute(sql, (url,))
|
|
#log.debug(c._last_executed)
|
|
if (c.fetchone()[0] > 0):
|
|
return True
|
|
return False
|
|
|
|
def load_into_db(rows, event=None):
|
|
'''
|
|
CREATE TABLE `results` (
|
|
`result_key` int(11) NOT NULL AUTO_INCREMENT,
|
|
`date` datetime DEFAULT NULL,
|
|
`distance` decimal(4,2) DEFAULT NULL,
|
|
`event` varchar(100) COLLATE utf8_unicode_ci NOT NULL,
|
|
`position` int(5) NOT NULL,
|
|
`finishers` int(5) DEFAULT NULL,
|
|
`time` time NOT NULL,
|
|
`name` varchar(75) COLLATE utf8_unicode_ci DEFAULT NULL,
|
|
`surname` varchar(75) COLLATE utf8_unicode_ci DEFAULT NULL,
|
|
`licence` varchar(20) COLLATE utf8_unicode_ci DEFAULT NULL,
|
|
`club` varchar(80) COLLATE utf8_unicode_ci DEFAULT NULL,
|
|
`age` int(3) DEFAULT NULL,
|
|
`sex` varchar(10) COLLATE utf8_unicode_ci DEFAULT NULL,
|
|
`sexposition` int(5) NOT NULL,
|
|
`sexfinishers` int(5) DEFAULT NULL,
|
|
`category` varchar(15) COLLATE utf8_unicode_ci DEFAULT NULL,
|
|
`catposition` int(5) NOT NULL,
|
|
`catfinishers` int(5) DEFAULT NULL,
|
|
`source` varchar(200) COLLATE utf8_unicode_ci DEFAULT NULL,
|
|
PRIMARY KEY (`result_key`)
|
|
) ENGINE=InnoDB CHARSET=utf8 COLLATE=utf8_unicode_ci;
|
|
'''
|
|
if rows is None or len(rows) < 1:
|
|
log.warning("**** No data found in spreadsheet ****")
|
|
else:
|
|
db = MySQLdb.connect(user='aac', passwd='saOAcCWHg4LaoSSA', db='AAC',
|
|
use_unicode=True, charset="utf8")
|
|
c = db.cursor()
|
|
|
|
''' Check for duplicate values by DATE and POSITION and RACE and EVENT '''
|
|
sql = 'SELECT COUNT(*) FROM `results` WHERE source LIKE %s'
|
|
c.execute(sql, (rows[0].get('source'),))
|
|
#log.debug(c._last_executed)
|
|
if (c.fetchone()[0] > 0):
|
|
log.info("Spreadsheet data already loaded")
|
|
return
|
|
|
|
sql = 'SELECT COUNT(*) FROM `results` WHERE date=%s AND position=%s AND distance LIKE %s AND event LIKE %s'
|
|
c.execute(sql, (rows[0].get('date'), rows[0].get('position'), rows[0].get('distance'), rows[0].get('event'),))
|
|
#log.debug(c._last_executed)
|
|
if (c.fetchone()[0] > 0):
|
|
log.info("Spreadsheet data already loaded")
|
|
return
|
|
|
|
for r in rows:
|
|
fields = ', '.join(r.keys())
|
|
values = ', '.join(['%s'] * len(r)) # placeholder values
|
|
sql = 'INSERT into `results` ( {} ) VALUES ( {} )'.format(fields, values)
|
|
try:
|
|
c.execute(sql, r.values())
|
|
except :
|
|
e = sys.exc_info()[0]
|
|
log.debug("ERROR: {}".format(e))
|
|
log.debug("Last query was: {}".format(c._last_executed))
|
|
raise
|
|
#pass
|
|
|
|
db.commit()
|
|
#position_calculations(event)
|
|
|
|
return
|
|
|
|
def read_spreadsheet(spreadsheet, src=None, eventname=None, eventdate=None, eventdistance=None):
|
|
rows = []
|
|
filename = os.path.basename(spreadsheet)
|
|
if re.search('.xlsx?$', spreadsheet, flags=re.IGNORECASE) is not None:
|
|
book = xlrd.open_workbook(spreadsheet)
|
|
for sheetname in book.sheet_names():
|
|
if re.search('walk', sheetname, flags=re.IGNORECASE) is not None:
|
|
continue
|
|
sheet = book.sheet_by_name(sheetname)
|
|
log.debug("Processing sheet {}".format(sheetname))
|
|
|
|
''' Look for the header in the first 15 rows, searching from the top '''
|
|
fields = []
|
|
for row in range(0, 15):
|
|
try:
|
|
if re.search('((pos\w*|no\w*|num\w*|(last\s*|sur)name|(first\s*)?name|time|club)\s*){2,}', ' '.join(str(x) for x in (sheet.row_values(row))), flags=re.IGNORECASE) is not None:
|
|
fields = sheet.row_values(row)
|
|
log.debug("Spreadsheet fields: {}".format(', '.join(str(x) for x in fields)))
|
|
break
|
|
except:
|
|
''' Probably a blank sheet, let's skip '''
|
|
continue
|
|
''' Translate field names, and delete unwanted fields '''
|
|
position_idx = None
|
|
time_idx = {}
|
|
for i in range(len(fields)):
|
|
if 'position' not in fields and re.search('^\s*(overall)?\s*(pos|place|index)', str(fields[i]), flags=re.IGNORECASE) is not None:
|
|
fields[i] = 'position'
|
|
''' Store the index of this field for later processing '''
|
|
position_idx = i
|
|
elif 'time' not in fields and re.search('^\s*(race\s*)?(finish|elapsed_?|f\S?|net|chip)?\s*(time|h:?m:?s?)', str(fields[i]), flags=re.IGNORECASE) is not None:
|
|
fields[i] = 'time'
|
|
''' Store the index of this field for later processing '''
|
|
time_idx[fields[i]] = i
|
|
elif re.search('^\s*start\s*time', str(fields[i]), flags=re.IGNORECASE) is not None:
|
|
fields[i] = 'starttime'
|
|
time_idx[fields[i]] = i
|
|
elif re.search('^\s*(age\s*)?cat\S*\s*pos(\.|\w+)?\s*$', str(fields[i]), flags=re.IGNORECASE) is not None:
|
|
fields[i] = 'catposition'
|
|
elif re.search('^\s*(sex|gender)\s*pos(\.|\w*)\s*$', str(fields[i]), flags=re.IGNORECASE) is not None:
|
|
fields[i] = 'sexposition'
|
|
elif re.search('^\s*pos(\.|\w+)\s*(sex|gender)\s*$', str(fields[i]), flags=re.IGNORECASE) is not None:
|
|
fields[i] = 'sexposition'
|
|
elif re.search('^\s*(sur|last)\s*name', str(fields[i]), flags=re.IGNORECASE) is not None:
|
|
fields[i] = 'surname'
|
|
elif re.search('^\s*((first|nick)?\s*name|participant)', str(fields[i]), flags=re.IGNORECASE) is not None:
|
|
fields[i] = 'name'
|
|
elif re.search('^\s*(club(\.|\w*)|team)\s*(name)?\s*$', str(fields[i]), flags=re.IGNORECASE) is not None:
|
|
fields[i] = 'club'
|
|
elif re.search('^\s*age(\.|\w*)\s*$', str(fields[i]), flags=re.IGNORECASE) is not None:
|
|
fields[i] = 'age'
|
|
elif re.search('^\s*(sex|gender|m.?f|male|female)(\.|\w*)\s*$', str(fields[i]), flags=re.IGNORECASE) is not None:
|
|
fields[i] = 'sex'
|
|
elif re.search('^\s*((age\s*)?cat|extra group)(\.|\w*)\s*$', str(fields[i]), flags=re.IGNORECASE) is not None:
|
|
fields[i] = 'category'
|
|
elif re.search('^\s*(race)?\s*(lic|no|num)(\.|\S*)\s*\S*\s*$', str(fields[i]), flags=re.IGNORECASE) is not None:
|
|
fields[i] = 'licence'
|
|
elif re.search('^\s*(race)?date', str(fields[i]), flags=re.IGNORECASE) is not None:
|
|
fields[i] = 'date'
|
|
elif re.search('^\s*(race)?dist(ance)?\s*$', str(fields[i]), flags=re.IGNORECASE) is not None:
|
|
fields[i] = 'distance'
|
|
elif re.search('^\s*(race)?(event|name)\s*$', str(fields[i]), flags=re.IGNORECASE) is not None:
|
|
fields[i] = 'event'
|
|
|
|
''' If there isn't a position field or a time field, we don't want this sheet '''
|
|
if position_idx is None or time_idx is None:
|
|
continue
|
|
|
|
''' Look for the date in the file name, and then look the first 15 rows and override it '''
|
|
if eventdate is None:
|
|
eventdate = dt.datetime.min
|
|
filedate = re.search('(20\d{2})', str(filename), flags=re.IGNORECASE)
|
|
if filedate is not None:
|
|
eventdate = filedate.group(1)
|
|
for row in range(0, 15):
|
|
sheetdate = re.search('(\d+\s*(Jan|Feb|Mar|Apr|May|Jun|Jul|Aug|Sep|Oct|Nov|Dec)\w*\s+\d{4})', str(sheet.cell(row, 0).value), flags=re.IGNORECASE)
|
|
if sheetdate is not None:
|
|
eventdate = sheetdate.group(1)
|
|
break
|
|
log.info("Race date: {}".format(eventdate))
|
|
#log.info("Race date: {}".format(eventdate.strftime('%Y-%m-%d')))
|
|
|
|
if eventname is None:
|
|
''' Use the filename as the event name :-( '''
|
|
eventname, *_ = os.path.splitext(filename)
|
|
eventname = str(eventname)
|
|
''' Clean up common patterns '''
|
|
eventname = re.sub('[\-_]', ' ', eventname, flags=re.IGNORECASE)
|
|
eventname = re.sub('results?(\s*book)?', ' ', eventname, flags=re.IGNORECASE)
|
|
eventname = re.sub('export', ' ', eventname, flags=re.IGNORECASE)
|
|
eventname = re.sub('excel', ' ', eventname, flags=re.IGNORECASE)
|
|
eventname = re.sub('\(\d\)', ' ', eventname, flags=re.IGNORECASE)
|
|
eventname = re.sub('\d{0,4}20\d{2}\d{0,4}', ' ', eventname, flags=re.IGNORECASE)
|
|
eventname = re.sub('\s\s+', ' ', eventname, flags=re.IGNORECASE)
|
|
eventname = re.sub('(^\s*|\s*$)', '', eventname, flags=re.IGNORECASE)
|
|
log.info("Event name: {}".format(eventname))
|
|
|
|
''' Look for the race distance in the sheet name, or in the filename '''
|
|
distance = eventdistance
|
|
log.info("Race distance: {}".format(distance))
|
|
eventnamedistance = re.search('(Half)?[\s-]*(Marathon)', eventname, flags=re.IGNORECASE)
|
|
if eventnamedistance is not None:
|
|
if eventnamedistance.group(1) is not None:
|
|
distance = 21.1
|
|
else:
|
|
distance = 42.2
|
|
filedistance = re.search('(\d{1,2}([\.,]\d)?)\s*KM', filename, flags=re.IGNORECASE)
|
|
if not distance and filedistance is not None:
|
|
distance = filedistance.group(1)
|
|
eventnamedistance = re.search('([\d\.,]{2,3})\s*KM', eventname, flags=re.IGNORECASE)
|
|
if eventnamedistance is not None:
|
|
distance = eventnamedistance.group(1)
|
|
sheetdistance = re.search('([\d\.,]+\s*KM)', sheetname, flags=re.IGNORECASE)
|
|
if sheetdistance is not None:
|
|
distance = sheetdistance.group(1)
|
|
sheetdistance = re.search('(helper|marshal)', sheetname, flags=re.IGNORECASE)
|
|
if sheetdistance is not None:
|
|
distance = sheetname
|
|
log.info("Race distance: {}".format(distance))
|
|
|
|
for row in range(sheet.nrows):
|
|
''' TODO: don't assume that the position is the first cell '''
|
|
if re.search('(^\s*$|[A-Za-z])', str(sheet.cell(row, position_idx).value), flags=re.IGNORECASE) is None:
|
|
item = dict()
|
|
data = []
|
|
for col in range(sheet.ncols):
|
|
data.append(sheet.cell(row, col).value)
|
|
item = dict(zip(fields, data))
|
|
''' If the time has been modified by Excel, unmodify it '''
|
|
timecols = ( 'time', 'starttime' )
|
|
for timecol in timecols:
|
|
if timecol in item and isinstance(item[timecol], float):
|
|
try:
|
|
item[timecol] = xlrd.xldate_as_tuple(sheet.cell(row, time_idx[timecol]).value, book.datemode)
|
|
except:
|
|
try:
|
|
if book.datemode == 1:
|
|
flipdatemode = 0
|
|
else:
|
|
flipdatemode = 1
|
|
item[timecol] = xlrd.xldate_as_tuple(sheet.cell(row, time_idx[timecol]).value, flipdatemode)
|
|
except:
|
|
continue
|
|
''' Skip this row if the date can't be parsed, as it's probably wrong anyway (41 hours or something silly) '''
|
|
continue
|
|
item['source'] = src
|
|
item['date'] = eventdate
|
|
item['event'] = eventname
|
|
if not 'distance' in item:
|
|
item['distance'] = distance
|
|
rows.append(item)
|
|
|
|
rows = clean_data(rows)
|
|
if len(rows) > 0:
|
|
log.debug("Sample output: {}".format(pp.pformat(rows[0])))
|
|
return rows
|
|
|
|
|
|
def clean_data(input_rows):
|
|
rows = []
|
|
for ir in input_rows:
|
|
r = dict()
|
|
|
|
''' Fix date '''
|
|
date = ir.get('date')
|
|
if isinstance(date, str):
|
|
today = dt.datetime.now()
|
|
year = dt.datetime.combine(dt.date(year=today.year, month=1, day=1), dt.time(hour=0, minute=0, second=0))
|
|
date = dtp.parse(date, default=year)
|
|
r['date'] = date
|
|
|
|
''' Check time '''
|
|
time = ir['time']
|
|
#print("1: {} {}".format(time, type(time)))
|
|
''' Deal with various formats that xlrd might give us. Note that floats should already be converted to tuples '''
|
|
if isinstance(time, tuple):
|
|
time = dt.datetime.combine(dt.date(year=1900, month=1, day=1), dt.time(hour=time[3], minute=time[4], second=time[5]))
|
|
#print("2: {} {}".format(time, type(time)))
|
|
elif isinstance(time, float):
|
|
for fmt in ( '%H.%M.%S', '%M.%S', ):
|
|
try:
|
|
time = dt.datetime.strptime(time, fmt)
|
|
#print("3: {} {} {}".format(time, type(time), fmt))
|
|
except:
|
|
pass
|
|
elif isinstance(time, str):
|
|
for fmt in ( '%H:%M:%S', '%M:%S', '%H.%M.%S', '%M.%S', ):
|
|
try:
|
|
time = dt.datetime.strptime(time, fmt)
|
|
#print("4: {} {} {}".format(time, type(time), fmt))
|
|
except:
|
|
pass
|
|
''' If time is still a string, ignore it, as it's probably blank/DNF '''
|
|
if isinstance(time, str):
|
|
continue
|
|
if 'starttime' in ir:
|
|
starttime = ir['starttime']
|
|
#print("Start 1: {} {}".format(starttime, type(starttime)))
|
|
if isinstance(starttime, tuple):
|
|
starttime = dt.datetime.combine(dt.date(year=1900, month=1, day=1), dt.time(hour=starttime[3], minute=starttime[4], second=starttime[5]))
|
|
#print("Start 2: {} {}".format(starttime, type(starttime)))
|
|
if isinstance(starttime, float):
|
|
for fmt in ( '%H.%M.%S', '%M.%S', ):
|
|
try:
|
|
starttime = dt.datetime.strptime(starttime, fmt)
|
|
#print("Start 3: {} {} {}".format(starttime, type(starttime), fmt))
|
|
except:
|
|
pass
|
|
elif isinstance(starttime, str):
|
|
for fmt in ( '%H:%M:%S', '%M:%S', '%H.%M.%S', '%M.%S', ):
|
|
try:
|
|
starttime = dt.datetime.strptime(starttime, fmt)
|
|
#print("Start 4: {} {} {}".format(starttime, type(starttime), fmt))
|
|
except:
|
|
pass
|
|
''' If starttime is still a string, ignore it, as it's probably blank '''
|
|
if not isinstance(time, str) and not isinstance(starttime, str):
|
|
timedelta = time - starttime
|
|
time = dt.datetime.min + timedelta
|
|
#print("5: {} {}".format(time, type(time)))
|
|
r['time'] = time.time()
|
|
|
|
''' Fix distance '''
|
|
length = re.search('([\d\.,]+)(?:\s*km)?', str(ir.get('distance')), flags=re.IGNORECASE)
|
|
if length is not None:
|
|
distance = re.sub(",", ".", length.group(1))
|
|
r['distance'] = float(distance)
|
|
else:
|
|
r['distance'] = 0
|
|
|
|
''' Should be an int '''
|
|
for key in ( 'position', 'sexposition', 'catposition', 'age', ):
|
|
val = ir.get(key)
|
|
if val is not None:
|
|
try:
|
|
r[key] = int(val)
|
|
except:
|
|
pass
|
|
|
|
''' Should be a string '''
|
|
for key in ( 'event', 'name', 'surname', 'licence', 'club', 'category', ):
|
|
val = ir.get(key)
|
|
if isinstance(val, float):
|
|
val = int(val)
|
|
if val is not None:
|
|
try:
|
|
r[key] = re.sub('(^\s*|\s*$)', '', str(val))
|
|
except:
|
|
pass
|
|
|
|
''' Leave alone '''
|
|
for key in ( 'event', 'source', ):
|
|
r[key] = ir.get(key)
|
|
|
|
''' Fix sex '''
|
|
if 'sex' in r:
|
|
sex = 'sex' in r and re.search('^\s*(F|M)', str(r.get('sex')), flags=re.IGNORECASE)
|
|
if sex is not None:
|
|
if sex.group(1) == 'F':
|
|
r['sex'] = 'F'
|
|
else:
|
|
r['sex'] = 'M'
|
|
elif 'category' in r:
|
|
sex = 'category' in r and re.search('^\s*(F|M)', str(r.get('category')), flags=re.IGNORECASE)
|
|
if sex is not None:
|
|
if sex.group(1) == 'F':
|
|
r['sex'] = 'F'
|
|
else:
|
|
r['sex'] = 'M'
|
|
|
|
''' Fix club '''
|
|
if re.search('^\s*(AAC$|Atlantic\s*Athletic)', str(r.get('club')), flags=re.IGNORECASE) is not None:
|
|
r['club'] = 'AAC'
|
|
|
|
rows.append(r)
|
|
|
|
''' sort rows by position, then populate sexposition and catposition for each sex and category '''
|
|
sorted(rows, key=lambda r: r['position'])
|
|
totals = defaultdict(int)
|
|
for r in rows:
|
|
totals['positions'] += 1
|
|
if 'sex' in r:
|
|
totals[r['sex']] += 1
|
|
#log.debug("{} {}".format(totals[r['sex']], r['sex']))
|
|
if 'sexposition' not in r:
|
|
r['sexposition'] = totals[r['sex']]
|
|
if 'category' in r:
|
|
totals[r['category']] += 1
|
|
#log.debug("{} {}".format(totals[r['category']], r['category']))
|
|
if 'catposition' not in r:
|
|
r['catposition'] = totals[r['category']]
|
|
for i in rows:
|
|
r['finishers'] = totals['positions']
|
|
if 'sex' in r:
|
|
r['sexfinishers'] = totals[r['sex']]
|
|
if 'category' in r:
|
|
r['catfinishers'] = totals[r['category']]
|
|
return rows
|
|
|
|
|
|
def parse_arguments():
|
|
parser = argparse.ArgumentParser(description='Load a spreadsheet containing WPA results into a database')
|
|
parser.add_argument(
|
|
'--url', '-u', action='append', required=False, dest='scrapeurl', nargs="?",
|
|
help='Scrape WPA website, or, if a link is specified, the spreadsheet at that link')
|
|
parser.add_argument(
|
|
'--calc', '-c', action='append', required=False, dest='calculate', nargs="?",
|
|
help='Calculate unset positions in the database, optionally just for specified races')
|
|
parser.add_argument(
|
|
'--input', '-i', action='store', required=False, type=str, dest='input_file',
|
|
help='Manually select the spreadsheet to be imported')
|
|
parser.add_argument(
|
|
'--verbose', '-v', action='count', required=False, dest='verbose',
|
|
help='Print more information')
|
|
args = parser.parse_args()
|
|
|
|
if args.input_file:
|
|
if not os.path.exists(args.input_file) or not os.access(args.input_file, os.R_OK):
|
|
raise
|
|
|
|
logging.basicConfig(format='%(message)s')
|
|
if args.verbose is not None and args.verbose == 1:
|
|
log.setLevel(logging.INFO)
|
|
elif args.verbose is not None and args.verbose >= 2:
|
|
log.setLevel(logging.DEBUG)
|
|
else:
|
|
log.setLevel(logging.WARNING)
|
|
|
|
return args
|
|
|
|
if __name__ == "__main__":
|
|
main()
|
|
|
|
# vim: set expandtab shiftwidth=2 softtabstop=2 tw=0 :
|