def get_all_mails_for_all_days(mails):
mails = sort_mails(mails)
for i in range(len(mails)):
if utils.dateParser(mails[i].headers['Date']).year < 1998:
pass
else:
mails = mails[i:]
break
first_date = utils.dateParser(mails[0].headers['Date'])
last_date = utils.dateParser(mails[-1].headers['Date'])
ret_list = []
index = 0
for d in range((last_date - first_date).days):
current_date = first_date + datetime.timedelta(d)
new_index, mm = get_mail_by_day(current_date, mails[index:])
index += new_index
#print index
ret_list.append((current_date, mm))
return ret_list
def group_mails_by_week(mails):
mails = sorted(mails, key = lambda x: utils.dateParser(x.headers['Date']).toordinal())
start_date = utils.dateParser(mails[0].headers['Date']).isocalendar()
ret_list = []
current_year = start_date[0]
current_week = start_date[1]
new_mails = []
for i in range(len(mails)):
iso_date = utils.dateParser(mails[i].headers['Date']).isocalendar()
if iso_date[0] == current_year and iso_date[1] == current_week:
new_mails.append(mails[i])
else:
ret_list.append((current_year, current_week, new_mails))
new_mails = []
if current_week == 53:
current_week = 1
current_year += 1
else:
current_week += 1
i -= 1
return ret_list
def get_mail_by_day(date, mails):
ret_list = []
found_one = False
year, month, day = date.year, date.month, date.day
for i in range(len(mails)):
mail_date = utils.dateParser(mails[i].headers['Date'])
if mail_date.year == year and mail_date.month == month and mail_date.day == day:
ret_list.append(mails[i])
found_one = True
else:
if found_one == True:
return (i ,ret_list)
if found_one:
return (i, ret_list)
return (0, ret_list)
def getDataframe(ticker,
dateRange=False,
dataFields=False,
sorting=0,
printError=False):
'''
Returns the dataframe corresponding to the passed ticker.
Returns False if dataframe is not present or values requested are not valid.
Daterange must be a 2 element list, in the following format: [[<start date>], [<end date>]], date format = string "YYYY-MM-DD".
dateFields must be a list of strings corresponding to the columns of the dataframe you want returned.
By default the dataframe is descending i.e. df.iloc[0] = 2018-01-01 df.iloc[1] = 2018-01-02
'''
global DatabaseDictionary
try:
DatabaseLock.acquire()
dataframe = DatabaseDictionary[ticker]
DatabaseLock.release()
if (dateRange):
if type(dateRange[0]) is str:
dateRange = [
dateParser(dateRange[0]),
dateParser(dateRange[1])
]
if (dataFields):
mask = (dataframe.index >= dateRange[0]) & (dataframe.index <=
dateRange[1])
dataframeReturn = dataframe.loc[mask, dataFields]
else:
mask = (dataframe.index >= dateRange[0]) & (dataframe.index <=
dateRange[1])
dataframeReturn = dataframe.loc[mask]
elif (dataFields):
dataframeReturn = dataframe[dataFields]
else:
dataframeReturn = dataframe
if (len(dataframeReturn) == 0):
#Data is not available
return False
else:
return dataframeReturn
except:
#Dataframe has not yet been loaded into memory. Get from file
DatabaseLock.release()
try:
if (dateRange):
if type(dateRange[0]) is str:
dateRange = [
dateParser(dateRange[0]),
dateParser(dateRange[1])
]
if (dataFields):
dataframe = loadDataframeFromFile(ticker + ".csv")[1]
mask = (dataframe.index >=
dateRange[0]) & (dataframe.index <= dateRange[1])
dataframe = dataframe.loc[mask, dataFields]
else:
dataframe = loadDataframeFromFile(ticker + ".csv")[1]
mask = (dataframe.index >=
dateRange[0]) & (dataframe.index <= dateRange[1])
dataframe = dataframe.loc[mask]
elif (dataFields):
dataframe = loadDataframeFromFile(ticker +
".csv")[1][dataFields]
else:
dataframe = loadDataframeFromFile(ticker + ".csv")[1]
if (len(dataframe) == 0):
#Data is not available
return False
else:
#dataframe.sort_values('date',ascending=sorting)
return dataframe
except Exception as e:
if printError:
print(e)
print(str(e))
return False
def sort_mails(mails):
return sorted(mails, key = lambda x: utils.dateParser(x.headers['Date']).toordinal())
def get_mails_by_day(mails, day, WEEK_DAY_dict = WEEK_DAYS):
ret_mails = []
for mail in mails:
if WEEK_DAY_dict[utils.dateParser(mail.headers['Date']).isoweekday()] == day:
ret_mails.append(mail)
return ret_mails
def filter_by_hour(mails, hour): # day is a value of WEEK_DAYS
ret_mail = []
for mail in mails:
if utils.dateParser(mail.headers['Date']).hour == hour:
ret_mail.append(mail)
return ret_mail
def count_mails_in_day(mails, day, WEEK_DAY_dict = WEEK_DAYS): # day is a value of WEEK_DAYS
count = 0
for mail in mails:
if WEEK_DAY_dict[utils.dateParser(mail.headers['Date']).isoweekday()] == day:
count += 1
return count
