решено: Это была магическая функция %%timeit, которую я поместил в начале ячейки. Я даже забыл упомянуть об этом в вопросе ниже. Я удалил его, и все стало работать нормально.
У меня есть список файлов CSV, которые я собираюсь зациклить и отправить в функцию. Список файлов называется 1404.csv, 1405.csv, ....., 1905.csv, 1906.csv
Когда я использую цикл без функции, все работает нормально.
for i in os.listdir("dispensing_csv"):
db_file = "05.db"
csv_file = "dispensing_csv/" + i
month = "01/" + i[2:4] + "/20" + i[0:2]
# dispensing_csv_to_sql(db_file, csv_file, month)
print("Period " + month + " written.")
Выводы становятся:
Period 01/04/2014 written.
Period 01/05/2014 written.
....
....
Period 01/05/2019 written.
Period 01/06/2019 written.
Но когда я пытаюсь использовать функцию внутри цикла, цикл никогда не заканчивается.
for i in os.listdir("dispensing_csv"):
db_file = "05.db"
csv_file = "dispensing_csv/" + i
month = "01/" + i[2:4] + "/20" + i[0:2]
dispensing_csv_to_sql(db_file, csv_file, month)
print("Period " + month + " written.")
Выходыстать:
Period 01/04/2014 written.
Period 01/05/2014 written.
....
....
Period 01/05/2019 written.
Period 01/06/2019 written.
Period 01/04/2014 written.
Period 01/05/2014 written.
....
....
....
Я использую Python 3.7.3 в Jupyter Lab. Как следует из названия, функция пытается записать CSV-файл в файл sqlite.
Заранее благодарен за помощь.
ДОБАВЛЕНО dispensing_csv_to_sql ()
def dispensing_csv_to_sql(db_file, csv_file, month):
# Same as in ContractorTypes in SQL db
ContractorTypes = {"Abated": 1, "Appliance": 2, "Pharmacy": 3}
# Connecting to the db
conn = sqlite3.connect(db_file)
c = conn.cursor()
# Reading the CSV file in dataframe
df = pd.read_csv(csv_file)
df.columns = df.columns.str.strip() # to remove possible whitespaces
# Checking if period exists in the db, update if not. Return period id
query_period = c.execute("SELECT 1 FROM Periods WHERE Period = ?", (month, ))
query_period = list(query_period)
try:
query_period = query_period[0][0]
except:
c.execute("INSERT INTO Periods (Period) VALUES (?)", (month, ))
# Getting the period_id to use in DispensingNumbers table
period_id = c.execute("SELECT PeriodID FROM Periods WHERE Period = ?", (month, ))
period_id = list(period_id)
period_id = period_id[0][0] # period_id
# Now iterating rows in df
for index, rows in df.iterrows():
# Checking if contractor exists in the db, update if not. Return contractor id
query_contractor = c.execute("SELECT 1 FROM Contractors WHERE ContractorCode = ?", (rows["ContractorCode"], ))
query_contractor = list(query_contractor)
try:
query_contractor = query_contractor[0][0]
except:
# Checking if area exists in the db, update if not. Return area id
query_area = c.execute("SELECT 1 FROM Areas WHERE AreaCode = ?", (rows["AreaCode"], ))
query_area = list(query_area)
try:
query_area = query_area[0][0]
except:
# Inserting non-existing area data
insert_area = "INSERT INTO Areas (AreaCode, AreaName) VALUES (?, ?)"
c.execute(insert_area, (rows["AreaCode"], rows["Area"]))
# Getting the area_id to use in Contractors table
area_id = c.execute("SELECT AreaID FROM Areas WHERE AreaCode = ?", (rows["AreaCode"], ))
area_id = list(area_id)
area_id = area_id[0][0] # area_id
# Inserting non-existing Contractor data
ContractorCode = rows["ContractorCode"]
ContractorName = rows["ContractorName"]
ContractorAddress = " ".join(rows[5:9].dropna().tolist())
ContractorPostcode = rows["Postcode"]
AreaID = area_id
ContractorTypeID = ContractorTypes[rows["Contractor Type"].strip()]
insert_contractor = "INSERT INTO Contractors (ContractorCode, ContractorName, ContractorAddress, \
ContractorPostcode, AreaID, ContractorTypeID) VALUES (?, ?, ?, ?, ?, ?)"
contractor_data_to_be_inserted = (ContractorCode, ContractorName, ContractorAddress, ContractorPostcode, AreaID, ContractorTypeID)
c.execute(insert_contractor, contractor_data_to_be_inserted)
# Getting the contractor_id to use in DispensingNumbers table
contractor_id = c.execute("SELECT ContractorID FROM Contractors WHERE ContractorCode = ?", (rows["ContractorCode"], ))
contractor_id = list(contractor_id)
contractor_id = contractor_id[0][0] # contractor_id
# Inserting dispensing numbers data
ContractorID = contractor_id
PeriodID = period_id
NumberofForms = rows["NumberofForms"] or 0
NumberofItems = rows["NumberofItems"] or 0
TotalNofPresPF = rows["TotalnumberofPrescriptions(ProfessionalFees)"] or 0
NofPresPFStdDis = rows["NumberofPrescriptions(ProfessionalFees)(Standarddiscountrate)"] or 0
NofPresPFZeroDis = rows["NumberofPrescriptions(ProfessionalFees)(Zerodiscountrate)"] or 0
NofFormEPS = rows["NumberofformsforElectronicPrescriptionService(EPS)"] or 0
NofItemEPS = rows["NumberofItemsprocessedviaElectronicPrescriptionService(EPS)"] or 0
MUR = rows["NumberofMedicineUseReviews(MURs)declared"] or 0
NMS = rows["NumberofNewMedicineService(NMS)interventionsdeclared"] or 0
AURHome = rows["NumberofApplianceUseReviews(AURs)conductedinusershomes"] or 0
AURPremise = rows["NumberofApplianceUseReviews(AURs)conductedatpremises"] or 0
Stoma = rows["NumberofStomaCustomisationFees"] or 0
insert_dispensingno = "INSERT INTO DispensingNumbers (ContractorID, PeriodID, NumberofForms, \
NumberofItems, TotalNofPresPF, NofPresPFStdDis, NofPresPFZeroDis, NofFormEPS, NofItemEPS, MUR, \
NMS, AURHome, AURPremise, Stoma) VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)"
dispensingno_data_to_be_inserted = (ContractorID, PeriodID, NumberofForms, NumberofItems, TotalNofPresPF, NofPresPFStdDis,
NofPresPFZeroDis, NofFormEPS, NofItemEPS, MUR, NMS, AURHome, AURPremise, Stoma)
c.execute(insert_dispensingno, dispensingno_data_to_be_inserted)
conn.commit()
conn.close()