def MarriageBeforeDivorce(row):
msg = []
if isinstance(row['Married'], str) and isinstance(row['Divorced'], str):
married_datetime = convertStringToDatetime(row['Married'])
divorce_datetime = convertStringToDatetime(row['Divorced'])
if married_datetime > divorce_datetime:
msg.append('ERROR: FAMILY: US04: ' + str(row['ID']) +
': Divorced: ' + str(divorce_datetime.date()) +
' occurs before Marriage: ' +
str(married_datetime.date()))
return msg
def BirthBeforeDeath(individualsDF):
msg = [] #empty list, this is where error messages will go
for index, row in individualsDF.iterrows(): #checking each individual in the family tree
if row['Alive'] == False: #only if the individual is dead
birth_datetime = convertStringToDatetime(row['Birthday'])
death_datetime = convertStringToDatetime(row['Death'])
#The rows above isolate the birth and death
if birth_datetime > death_datetime: #if the birth is after the death
msg.append('ERROR: INDIVIDUAL: US03: ' + str(row['ID']) + ' Died: ' + str(death_datetime.date()) +
' occurs before Birth: ' + str(birth_datetime.date()))
#append the error message
return msg
def siblings_spacing(individualsDF, familiesDF):
errors = []
for index, famRow in familiesDF.iterrows():
if isinstance(famRow["Children"], str) and len(literal_eval(famRow["Children"])) > 1: # No need to check if there is only one child in this family
childList = literal_eval(famRow['Children'])
for childId in childList:
child1Bday_dt = convertStringToDatetime(individualsDF.iloc[int(childId[1:])-1]["Birthday"])
for childId2 in childList[(childList.index(childId)+1):]:
child2Bday_dt = convertStringToDatetime(individualsDF.iloc[int(childId2[1:])-1]["Birthday"])
if (child1Bday_dt-child2Bday_dt).total_seconds() >= 2*24*60*60 and (child1Bday_dt-child2Bday_dt).total_seconds() <= 8*31*24*60*60:
errors.append("ERROR: FAMILY: US13: {}: Birthday of siblings {} & {} are between 2 days and 8 months".format(famRow["ID"], childId, childId2))
return errors
def birth_before_parent_marriage(fam_row, indi_df):
errors = []
if isinstance(fam_row["Married"], str):
marriage_dt = convertStringToDatetime(fam_row['Married'])
children_list = fam_row['Children']
for index, row in indi_df.iterrows():
#try:
if type(children_list) != float and row['ID'] in children_list:
birth_dt = convertStringToDatetime(row['Birthday'])
if birth_dt < marriage_dt:
errors.append("ANOMOLY: FAMILY: US08: {}: Child {} born {} before marriage on {}".format(fam_row['ID'], row['ID'], birth_dt.date(), marriage_dt.date()))
#print(errors)
return errors
def birth_before_marriage_check(individuals2, families2):
errors = []
for index, row_indi in individuals2.iterrows():
for index, row_fam in families2.iterrows():
if row_indi["ID"] == row_fam["Husband ID"] or row_indi[
"ID"] == row_fam["Wife ID"]:
if isinstance(row_fam["Married"], str):
birth_dt = convertStringToDatetime(row_indi['Birthday'])
marriage_dt = convertStringToDatetime(row_fam['Married'])
if birth_dt > marriage_dt:
errors.append(
"ERROR: INDIVIDUAL: US02: {}: Married {} before born {}"
.format(row_indi["ID"], marriage_dt.date(),
birth_dt.date()))
return errors
def marriage_after_fourteen(individuals2, families2):
errors = []
for index, row_indi in individuals2.iterrows():
for index, row_fam in families2.iterrows():
if row_indi["ID"] == row_fam["Husband ID"] or row_indi[
"ID"] == row_fam["Wife ID"]:
if isinstance(row_fam["Married"], str):
# compare marriage and birthdate and get the difference
# if difference is less than 14 then print an error
birth_dt = convertStringToDatetime(row_indi['Birthday'])
marriage_dt = convertStringToDatetime(row_fam['Married'])
duration_in_s = (marriage_dt - birth_dt).total_seconds()
years = divmod(duration_in_s, 31536000)[0]
if years < 14:
errors.append(
"ERROR: INDIVIDUAL: US10: {}: Married {} before age 14"
.format(row_indi["ID"], marriage_dt.date()))
return errors
def birth_before_current_check(row):
errors = []
currDate = datetime.datetime.now()
# row['Birthday'] comes in as a string from the csv, we convert it and store it as a list 'birth' here
birth_datetime = convertStringToDatetime(row['Birthday'])
if birth_datetime > currDate:
errors.append("ERROR: INDIVIDUAL: US01: {}: Birthday {} occurs in the future".format(row['ID'], birth_datetime.date()))
#print(errors)
return errors
def unique_child_name_and_birth(individuals, fam_row):
errors = []
seen = {}
children = fam_row['Children']
if not isinstance(children ,float):
fam_id = fam_row["ID"]
for index, row in individuals.iterrows():
if row["ID"] in children:
if not isinstance(row['Birthday'] ,float):
birth_dt = convertStringToDatetime(row['Birthday'])
if row["Name"] in seen and seen.get(row["Name"]) == birth_dt:
errors.append("ANOMOLY: FAMILY: US25: {}: More than one child is not unique".format(fam_id))
else:
seen[row["Name"]] = birth_dt
return errors
def divorce_before_death_check(famRow, individualsDF):
errors = []
# Only runs if the family was actually married and does not have a value of "NaN"
if isinstance(famRow["Divorced"], str):
# Retrieves the married date and converts it to a number using "datetime.datetime"
divorced_dt = convertStringToDatetime(famRow['Divorced'])
# Checks if the husband in the family is dead, if so, proceed
if individualsDF.iloc[int(famRow["Husband ID"][1:]) -
1]["Alive"] == False:
# Retrieves the husband's death date and converts it to a number using "datetime.datetime"
husbandDeath_dt = convertStringToDatetime(
individualsDF.iloc[int(famRow["Husband ID"][1:]) - 1]["Death"])
# Finally checks if divorce occured after the husband died, if so, append an error
if divorced_dt > husbandDeath_dt:
errors.append(
"ERROR: INDIVIDUAL: US06: {}: Death {} before Divorced {}".
format(famRow["Husband ID"], husbandDeath_dt.date(),
divorced_dt.date()))
# Checks if the wife in the family is dead, if so, proceed
if individualsDF.iloc[int(famRow["Wife ID"][1:]) -
1]["Alive"] == False:
# Retrieves the wife's death date and converts it to a number using "datetime.datetime"
wifeDeath_dt = convertStringToDatetime(
individualsDF.iloc[int(famRow["Wife ID"][1:]) - 1]["Death"])
# Finally checks if divorce occured after the wife died, if so, append an error
if divorced_dt > wifeDeath_dt:
errors.append(
"ERROR: INDIVIDUAL: US06: {}: Death {} before Divorced {}".
format(famRow["Wife ID"], wifeDeath_dt.date(),
divorced_dt.date()))
return errors
def no_more_than_5_kids_check(individualsDF, familiesDF):
errors = []
for index, famRow in familiesDF.iterrows():
numTwins = 0
if isinstance(famRow["Children"], str) and len(literal_eval(famRow["Children"])) > 5: # Needs to be greater than 5
birthList=[]
for childId in literal_eval(famRow['Children']):
childBday_dt = convertStringToDatetime(individualsDF.iloc[int(childId[1:])-1]["Birthday"])
birthList.append(childBday_dt)
for birth in birthList:
for birth2 in birthList[(birthList.index(birth)+1):]:
if abs((birth-birth2).days) < 2:
numTwins += 1
if(numTwins > 5):
errors.append("ERROR: FAMILY: US14: {}: More than 5 children born at the same time (Within 2 days of eachother)".format(famRow["ID"]))
return errors
def get_parents_death(individuals2, row_fam):
children = row_fam["Children"]
mother_death = [False, None]
father_death = [False, None]
for index, row_indi in individuals2.iterrows():
if isinstance(row_indi["Death"], str):
if row_indi["ID"] == row_fam["Wife ID"] or row_indi[
"ID"] == row_fam["Husband ID"]:
death_dt = convertStringToDatetime(row_indi['Death'])
if row_indi["ID"] == row_fam["Wife ID"]:
mother_death = [True, death_dt]
else:
father_death = [True, death_dt]
if not isinstance(children, float):
return birth_before_parents_death(individuals2, row_fam["ID"],
mother_death, father_death, children)
else:
return []
def birth_before_parents_death(individuals2, fam, mother_death, father_death,
children):
errors = []
for index, row_indi in individuals2.iterrows():
if row_indi["ID"] in children:
birth_dt = convertStringToDatetime(row_indi['Birthday'])
if mother_death[0] and birth_dt > mother_death[1]:
errors.append(
"ERROR: FAMILY: US09: {}: Child {} born {} after mother's death on {}"
.format(fam, row_indi["ID"], birth_dt.date,
mother_death[1].date))
nine_month = relativedelta(months=9)
if father_death[0] and (birth_dt - nine_month) > father_death[1]:
errors.append(
"ERROR: FAMILY: US09: {}: Child {} born {} after more than nine months after father's death on {}"
.format(fam, row_indi["ID"], birth_dt.date(),
father_death[1].date()))
return errors
def NoBigamy(dataframe):
msg = []
for index, row in dataframe.iterrows():
if isinstance(row['Husband ID'], str):
hus_id = row['Husband ID'] #save husband id
if isinstance(row['Wife ID'], str):
wife_id = row['Wife ID'] #save wife id
if isinstance(row['Married'], str):
married_datetime1 = convertStringToDatetime(row['Married'])
for index2, row2 in dataframe.iterrows():
if row2.all() == row.all(): #dont compare row to itself
continue
if wife_id == row2['Wife ID']:
if isinstance(row2['Married'],
str): #get second marriage date
married_datetime2 = convertStringToDatetime(
row2['Married'])
if isinstance(
row['Divorced'], str
): #different case if family is divorced or not
divorce_datetime1 = convertStringToDatetime(
row['Divorced']) #save divorce date
if married_datetime2 > married_datetime1 and married_datetime2 < divorce_datetime1:
msg.append('ERROR: FAMILY: US11: ' +
str(row['ID']) + ': ' + wife_id +
' married to both ' + hus_id +
' and ' + str(row2['Husband ID']) +
' at the same time')
elif isinstance(row2['Divorced'], str):
divorce_datetime2 = convertStringToDatetime(
row2['Divorced']) #save divorce date
if married_datetime1 > married_datetime2 and married_datetime1 < divorce_datetime2:
msg.append('ERROR: FAMILY: US11: ' +
str(row['ID']) + ': ' + wife_id +
' married to both ' + hus_id +
' and ' + str(row2['Husband ID']) +
' at the same time')
else:
msg.append('ERROR: FAMILY: US11: ' +
str(row['ID']) + ': ' + wife_id +
' married to both ' + hus_id + ' and ' +
str(row2['Husband ID']) +
' at the same time')
if hus_id == row2['Husband ID']:
if isinstance(row2['Married'],
str): #get second marriage date
married_datetime2 = convertStringToDatetime(
row2['Married'])
if isinstance(
row['Divorced'], str
): #different case if family is divorced or not
divorce_datetime1 = convertStringToDatetime(
row['Divorced']) #save divorce date
if married_datetime2 > married_datetime1 and married_datetime2 < divorce_datetime1:
msg.append('ERROR: FAMILY: US11: ' +
str(row['ID']) + ': ' + hus_id +
' married to both ' + wife_id +
' and ' + str(row2['Wife ID']) +
' at the same time')
elif isinstance(row2['Divorced'], str):
divorce_datetime2 = convertStringToDatetime(
row2['Divorced']) #save divorce date
if married_datetime1 > married_datetime2 and married_datetime1 < divorce_datetime2:
msg.append('ERROR: FAMILY: US11: ' +
str(row['ID']) + ': ' + hus_id +
' married to both ' + wife_id +
' and ' + str(row2['Wife ID']) +
' at the same time')
else:
msg.append('ERROR: FAMILY: US11: ' +
str(row['ID']) + ': ' + hus_id +
' married to both ' + wife_id +
' and ' + str(row2['Wife ID']) +
' at the same time')
return msg