logger = logging.getLogger("geolife")
#stdout = logging.StreamHandler()
#stdout.setLevel(logging.INFO)
#logger.addHandler(stdout)
from utils import ETACalculator
from schema import get_users
from schema import RecordsOnOneDay
import argparse
from sqlalchemy.sql import func
from sqlalchemy import Index
import csv
from raw import record
import config
engine = config.getEngine()
from sqlalchemy.orm import sessionmaker
Session = sessionmaker()
Session.configure(bind=engine)
def get_arguments():
parser = argparse.ArgumentParser(
description='Extract centroids of user movements.')
parser.add_argument(
'-t',
'--dry-run',
dest="dry_run",
action="store_true",
help=('Boolean indicator controlling if data should be added to the'
if __name__ == "__main__":
args = get_arguments()
directory = args.input_directory
logger.info("Loading database with raw Geolife records")
if args.weekday is None:
logger.info("All weekdays")
weekday = None
else:
weekday = int(args.weekday)
logger.info("Weekday: {0}".format(record.WEEKDAY_STRINGS[weekday]))
logger.info("Source: {0}".format(directory))
engine = config.getEngine()
Session = sessionmaker()
Session.configure(bind=engine)
session = Session()
logger.info("-"*50)
logger.info("Database will be created and populated from files"
" in {0}".format(directory))
record.initialize_table(engine)
logger.info("Table initialized")
timer = ETACalculator(iterations=geolife.get_num_files(directory))
user_weekday_counts = defaultdict(int)
for u in user.from_directory(directory):
logger.info("Beginning yielding of records from user {0.id}".format(u))
for f in u.files:
from sqlalchemy.ext.declarative import declarative_base
from sqlalchemy import Column
from sqlalchemy import Integer
from sqlalchemy import BigInteger
from sqlalchemy import Float
from sqlalchemy import Time
from sqlalchemy import Table
from sqlalchemy import MetaData
metadata = MetaData()
import config
RecordsOnOneDay = Table('day_records_view',
metadata,
autoload=True,
autoload_with=config.getEngine())
Base = declarative_base()
class HomogenizedRecord(Base):
__tablename__ = "time_homogenized"
id = Column(Integer, primary_key=True)
user = Column(BigInteger) #, index=True)
latitude = Column(Float)
longitude = Column(Float)
time = Column(Time)
def __repr__(self):
return "<HomogenizedTimeRecord(name={0}, (x,y)=({1}, {2}), time={3})>".format(
self.user, self.latitude, self.longitude, self.time)
from raw.record import GeoLifeUser
from raw.record import RawRecord
from schema import HomogenizedRecord
from schema import HomogenizedGeoLifeUser
from sqlalchemy import create_engine
from config import getEngine
engine = getEngine()
GeoLifeUser.__table__.drop(engine, checkfirst=True)
engine.execute("DROP VIEW day_records_view;")
RawRecord.__table__.drop(engine, checkfirst=True)
HomogenizedGeoLifeUser.__table__.drop(engine, checkfirst=True)
HomogenizedRecord.__table__.drop(engine, checkfirst=True)
import argparse
from utils import timerange
from utils import ETACalculator
from utils import num_elements_in_time_range
from datetime import time
from datetime import timedelta
from sqlalchemy.orm import sessionmaker
import os
from one import ExternalMovementReaderConverter
from utils import timeDifferenceSeconds
import messages
import pystache
import csv
Session = sessionmaker()
engine = getEngine()
Session.configure(bind=engine)
# Parse the command-line arguments.
def get_arguments():
parser = argparse.ArgumentParser(
description='Write out files for simulation.'
)
parser.add_argument(
'-d', '--time-delta',
dest='time_delta',
help="Number of seconds that should be between any two consecutive records",
type=lambda x: timedelta(seconds=int(x)),
default=timedelta(seconds=5),
)