I am trying to iterate over all nodes & child nodes in a tree using ElementTree. I would like to get the all the parent & its child XML tags as columns and values which could append the child nodes to parent in CSV format. I am using python 2.7. The header should be printed only once & below should be respective values
XML File :
<Customers>
<Customer CustomerID="GREAL">
<CompanyName>Great Lakes Food Market</CompanyName>
<ContactName>Howard Snyder</ContactName>
<ContactTitle>Marketing Manager</ContactTitle>
<Phone>(503) 555-7555</Phone>
<FullAddress>
<Address>2732 Baker Blvd.</Address>
<City>Eugene</City>
<Region>OR</Region>
<PostalCode>97403</PostalCode>
<Country>USA</Country>
</FullAddress>
</Customer>
<Customer CustomerID="HUNGC">
<CompanyName>Hungry Coyote Import Store</CompanyName>
<ContactName>Yoshi Latimer</ContactName>
<ContactTitle>Sales Representative</ContactTitle>
<Phone>(503) 555-6874</Phone>
<Fax>(503) 555-2376</Fax>
<FullAddress>
<Address>City Center Plaza 516 Main St.</Address>
<City>Elgin</City>
<Region>OR</Region>
<PostalCode>97827</PostalCode>
<Country>USA</Country>
</FullAddress>
</Customer>
<Customer CustomerID="LAZYK">
<CompanyName>Lazy K Kountry Store</CompanyName>
<ContactName>John Steel</ContactName>
<ContactTitle>Marketing Manager</ContactTitle>
<Phone>(509) 555-7969</Phone>
<Fax>(509) 555-6221</Fax>
<FullAddress>
<Address>12 Orchestra Terrace</Address>
<City>Walla Walla</City>
<Region>WA</Region>
<PostalCode>99362</PostalCode>
<Country>USA</Country>
</FullAddress>
</Customer>
<Customer CustomerID="LETSS">
<CompanyName>Let's Stop N Shop</CompanyName>
<ContactName>Jaime Yorres</ContactName>
<ContactTitle>Owner</ContactTitle>
<Phone>(415) 555-5938</Phone>
<FullAddress>
<Address>87 Polk St. Suite 5</Address>
<City>San Francisco</City>
<Region>CA</Region>
<PostalCode>94117</PostalCode>
<Country>USA</Country>
</FullAddress>
</Customer>
</Customers>
My Code:
#Import Libraries
import csv
import xmlschema
import xml.etree.ElementTree as ET
#Define the variable to store the XML Document
xml_file = 'C:/Users/391648/Desktop/BOSS_20190618_20190516_18062019141928_CUMA/source_Files_XML/CustomersOrders.xml'
#using XML Schema Library validate the XML against XSD
my_schema = xmlschema.XMLSchema('C:/Users/391648/Desktop/BOSS_20190618_20190516_18062019141928_CUMA/source_Files_XML/CustomersOrders.xsd')
SchemaCheck = my_schema.is_valid(xml_file)
print(SchemaCheck) #Prints as True if the document is validated with XSD
#Parse XML & get root
tree = ET.parse(xml_file)
root = tree.getroot()
#Create & Open CSV file
xml_data_to_csv = open('C:/Users/391648/Desktop/BOSS_20190618_20190516_18062019141928_CUMA/source_Files_XML/PythonXMl.csv','w')
#create variable to write to csv
csvWriter = csv.writer(xml_data_to_csv)
#Create list contains header
count =0
#Loop for each node
for element in root.findall('Customers/Customer'):
List_nodes = []
#Get head by Tag
if count ==0:
list_header =[]
Full_Address = []
CompanyName = element.find('CompanyName').tag
list_header.append(CompanyName)
ContactName = element.find('ContactName').tag
list_header.append(ContactName)
ContactTitle = element.find('ContactTitle').tag
list_header.append(ContactTitle)
Phone = element.find('Phone').tag
list_header.append(Phone)
print(list_header)
csvWriter.writerow(list_header)
count = count + 1
#Get the data of the Node
CompanyName = element.find('CompanyName').text
List_nodes.append(CompanyName)
ContactName = element.find('ContactName').text
List_nodes.append(ContactName)
ContactTitle = element.find('ContactTitle').text
List_nodes.append(ContactTitle)
Phone = element.find('Phone').text
List_nodes.append(Phone)
print(List_nodes)
#Write List_Nodes to CSV
csvWriter.writerow(List_nodes)
xml_data_to_csv.close()
Expected CSV output:
CompanyName,ContactName,ContactTitle,Phone, Address, City, Region, PostalCode, Country
Great Lakes Food Market,Howard Snyder,Marketing Manager,(503) 555-7555, City Center Plaza 516 Main St., Elgin, OR, 97827, USA
Hungry Coyote Import Store,Yoshi Latimer,Sales Representative,(503) 555-6874, 12 Orchestra Terrace, Walla Walla, WA, 99362, USA
You might be better off using lxml. It has most of the desired functionality for finding elements built in.
from lxml import etree
import csv
with open('file.xml') as fp:
xml = etree.fromstring(fp.read())
field_dict = {
'CompanyName': 'CompanyName',
'ContactName': 'ContactName',
'ContactTitle': 'ContactTitle',
'Phone': 'Phone',
'Address': 'FullAddress/Address',
'City': 'FullAddress/City',
'Region': 'FullAddress/Region',
'PostalCode': 'FullAddress/PostalCode',
'Country': 'FullAddress/Country'
}
customers = []
for customer in xml:
line = {k: customer.find(v).text for k, v in field_dict.items()}
customers.append(line)
with open('customers.csv', 'w') as fp:
writer = csv.DictWriter(fp, field_dict)
writer.writerows(customers)
You can use xmltodict to convert data to JSON format instead of parsing XML:
import xmltodict
import pandas as pd
with open('data.xml', 'r') as f:
data = xmltodict.parse(f.read())['Customers']['Customer']
data_pd = {'CompanyName': [i['CompanyName'] for i in data],
'ContactName': [i['ContactName'] for i in data],
'ContactTitle': [i['ContactTitle'] for i in data],
'Phone': [i['Phone'] for i in data],
'Address': [i['FullAddress']['Address'] for i in data],
'City': [i['FullAddress']['City'] for i in data],
'Region': [i['FullAddress']['Region'] for i in data],
'PostalCode': [i['FullAddress']['PostalCode'] for i in data],
'Country': [i['FullAddress']['Country'] for i in data]}
df = pd.DataFrame(data_pd)
df.to_csv('result.csv', index=False)
Output CSV file:
CompanyName,ContactName,ContactTitle,Phone,Address,City,Region,PostalCode,Country
Great Lakes Food Market,Howard Snyder,Marketing Manager,(503) 555-7555,2732 Baker Blvd.,Eugene,OR,97403,USA
Hungry Coyote Import Store,Yoshi Latimer,Sales Representative,(503) 555-6874,City Center Plaza 516 Main St.,Elgin,OR,97827,USA
Lazy K Kountry Store,John Steel,Marketing Manager,(509) 555-7969,12 Orchestra Terrace,Walla Walla,WA,99362,USA
Let's Stop N Shop,Jaime Yorres,Owner,(415) 555-5938,87 Polk St. Suite 5,San Francisco,CA,94117,USA
A couple of things I have changed:
Removed schema validation since I do not have the XSD. You may include it
Made the child node traversal dynamic instead of statically referring each child node
The main for loop condition changed to for customer in root.findall('Customer') from for customer in root.findall('Customers/Customer')
However, I tried to keep your program structure, library usage intact. Here is the modified program:
import xml.etree.ElementTree as et
import csv
tree = et.parse("../data/customers.xml")
root = tree.getroot()
headers = []
count = 0
xml_data_to_csv = open('../data/customers.csv', 'w')
csvWriter = csv.writer(xml_data_to_csv)
for customer in root.findall('Customer'):
data = []
for detail in customer:
if(detail.tag == 'FullAddress'):
for addresspart in detail:
data.append(addresspart.text.rstrip('/n/r'))
if(count == 0):
headers.append(addresspart.tag)
else:
data.append(detail.text.rstrip('/n/r'))
if(count == 0):
headers.append(detail.tag)
if(count == 0):
csvWriter.writerow(headers)
csvWriter.writerow(data)
count = count + 1
With the given input XML content it produces:
CompanyName,ContactName,ContactTitle,Phone,Address,City,Region,PostalCode,Country
Great Lakes Food Market,Howard Snyde,Marketing Manage,(503) 555-7555,2732 Baker Blvd.,Eugene,OR,97403,USA
Hungry Coyote Import Store,Yoshi Latime,Sales Representative,(503) 555-6874,(503) 555-2376,City Center Plaza 516 Main St.,Elgi,OR,97827,USA
Lazy K Kountry Store,John Steel,Marketing Manage,(509) 555-7969,(509) 555-6221,12 Orchestra Terrace,Walla Walla,WA,99362,USA
Let's Stop N Shop,Jaime Yorres,Owne,(415) 555-5938,87 Polk St. Suite 5,San Francisco,CA,94117,USA
Note: Instead of writing to CSV in the loop you may append to an array and write it at one go. It depends on your content size and performance.
Update: When you have customers and their orders in the XML
The XML processing and CSV writing code structure remains the same. Additionally, process Orders element while processing customers. Now, under Orders Order elements can be processed exactly like Customer. As you mentioned each Order has ShipInfo as well.
The input XML is assumed to be (based on the comment below):
<Customers>
<Customer CustomerID="GREAL">
<CompanyName>Great Lakes Food Market</CompanyName>
<ContactName>Howard Snyder</ContactName>
<ContactTitle>Marketing Manager</ContactTitle>
<Phone>(503) 555-7555</Phone>
<FullAddress>
<Address>2732 Baker Blvd.</Address>
<City>Eugene</City>
<Region>OR</Region>
<PostalCode>97403</PostalCode>
<Country>USA</Country>
</FullAddress>
<Orders>
<Order>
<Param1>Value1</Param1>
<Param2>Value2</Param2>
<ShipInfo>
<ShipInfoParam1>Value3</ShipInfoParam1>
<ShipInfoParam2>Value4</ShipInfoParam2>
</ShipInfo>
</Order>
<Order>
<Param1>Value5</Param1>
<Param2>Value6</Param2>
<ShipInfo>
<ShipInfoParam1>Value7</ShipInfoParam1>
<ShipInfoParam2>Value8</ShipInfoParam2>
</ShipInfo>
</Order>
</Orders>
</Customer>
<Customer CustomerID="HUNGC">
<CompanyName>Hungry Coyote Import Store</CompanyName>
<ContactName>Yoshi Latimer</ContactName>
<ContactTitle>Sales Representative</ContactTitle>
<Phone>(503) 555-6874</Phone>
<Fax>(503) 555-2376</Fax>
<FullAddress>
<Address>City Center Plaza 516 Main St.</Address>
<City>Elgin</City>
<Region>OR</Region>
<PostalCode>97827</PostalCode>
<Country>USA</Country>
</FullAddress>
<Orders>
<Order>
<Param1>Value7</Param1>
<Param2>Value8</Param2>
<ShipInfo>
<ShipInfoParam1>Value9</ShipInfoParam1>
<ShipInfoParam2>Value10</ShipInfoParam2>
</ShipInfo>
</Order>
</Orders>
</Customer>
<Customer CustomerID="LAZYK">
<CompanyName>Lazy K Kountry Store</CompanyName>
<ContactName>John Steel</ContactName>
<ContactTitle>Marketing Manager</ContactTitle>
<Phone>(509) 555-7969</Phone>
<Fax>(509) 555-6221</Fax>
<FullAddress>
<Address>12 Orchestra Terrace</Address>
<City>Walla Walla</City>
<Region>WA</Region>
<PostalCode>99362</PostalCode>
<Country>USA</Country>
</FullAddress>
</Customer>
<Customer CustomerID="LETSS">
<CompanyName>Let's Stop N Shop</CompanyName>
<ContactName>Jaime Yorres</ContactName>
<ContactTitle>Owner</ContactTitle>
<Phone>(415) 555-5938</Phone>
<FullAddress>
<Address>87 Polk St. Suite 5</Address>
<City>San Francisco</City>
<Region>CA</Region>
<PostalCode>94117</PostalCode>
<Country>USA</Country>
</FullAddress>
</Customer>
</Customers>
Here is the modified code that process both customers and orders:
import xml.etree.ElementTree as et
import csv
tree = et.parse("../data/customers-with-orders.xml")
root = tree.getroot()
customer_csv = open('../data/customers-part.csv', 'w')
order_csv = open('../data/orders-part.csv', 'w')
customerCsvWriter = csv.writer(customer_csv)
orderCsvWriter = csv.writer(order_csv)
customerHeaders = []
orderHeaders = ['CustomerID']
isFirstCustomer = True
isFirstOrder = True
def processOrders(customerId):
global isFirstOrder
for order in detail.findall('Order'):
orderData = [customerId]
for orderdetail in order:
if(orderdetail.tag == 'ShipInfo'):
for shipinfopart in orderdetail:
orderData.append(shipinfopart.text.rstrip('/n/r'))
if(isFirstOrder):
orderHeaders.append(shipinfopart.tag)
else:
orderData.append(orderdetail.text.rstrip('/n/r'))
if(isFirstOrder):
orderHeaders.append(orderdetail.tag)
if(isFirstOrder):
orderCsvWriter.writerow(orderHeaders)
orderCsvWriter.writerow(orderData)
isFirstOrder = False
for customer in root.findall('Customer'):
customerData = []
customerId = customer.get('CustomerID')
for detail in customer:
if(detail.tag == 'FullAddress'):
for addresspart in detail:
customerData.append(addresspart.text.rstrip('/n/r'))
if(isFirstCustomer):
customerHeaders.append(addresspart.tag)
elif(detail.tag == 'Orders'):
processOrders(customerId)
else:
customerData.append(detail.text.rstrip('/n/r'))
if(isFirstCustomer):
customerHeaders.append(detail.tag)
if(isFirstCustomer):
customerCsvWriter.writerow(customerHeaders)
customerCsvWriter.writerow(customerData)
isFirstCustomer = False
Output produced in customers-part.csv:
CompanyName,ContactName,ContactTitle,Phone,Address,City,Region,PostalCode,Country
Great Lakes Food Market,Howard Snyde,Marketing Manage,(503) 555-7555,2732 Baker Blvd.,Eugene,OR,97403,USA
Hungry Coyote Import Store,Yoshi Latime,Sales Representative,(503) 555-6874,(503) 555-2376,City Center Plaza 516 Main St.,Elgi,OR,97827,USA
Lazy K Kountry Store,John Steel,Marketing Manage,(509) 555-7969,(509) 555-6221,12 Orchestra Terrace,Walla Walla,WA,99362,USA
Let's Stop N Shop,Jaime Yorres,Owne,(415) 555-5938,87 Polk St. Suite 5,San Francisco,CA,94117,USA
Output produced in orders-part.csv:
CustomerID,Param1,Param2,ShipInfoParam1,ShipInfoParam2
GREAL,Value1,Value2,Value3,Value4
GREAL,Value5,Value6,Value7,Value8
HUNGC,Value7,Value8,Value9,Value10
Note: the code can be optimized further by reusing. I am leaving that part to you. Secondly, notice that in each order customer Id is added in order to distinguish.
Related
I am trying to create a dataframe out the XML code as shown below
<Structure>
<Field>
<Field_Name>GAMEREF</Field_Name>
<Field_Type>Numeric</Field_Type>
<Field_Len>4</Field_Len>
<Field_Dec>0</Field_Dec>
</Field>
...
<Field>
<Field_Name>WINLOSS</Field_Name>
<Field_Type>Character</Field_Type>
<Field_Len>1</Field_Len>
<Field_Dec>0</Field_Dec>
</Field>
</Structure>
<Records>
<Record>
<GAMEREF>1217</GAMEREF>
<YEAR>2021</YEAR>
(MORE ELEMENTS I DO NOT CARE ABOUT)
<GAMENO>1</GAMENO>
<WINLOSS>W</WINLOSS>
</Record>
...
<Record>
<GAMEREF>1220</GAMEREF>
<YEAR>2021</YEAR>
(MORE ELEMENTS I DO NOT CARE ABOUT)
<GAMENO>4</GAMENO>
<WINLOSS>L</WINLOSS>
</Record>
</Records>
The structure section of the XML code that is irrelevant to the dataframe I am trying to create.
I am trying to only use the XML elements of GAMEREF, YEAR, GAMENO, and WINLOSS as there are more in the XML for the Record elements.
I have tried using code as shown below to get this to work, but when I run the code I get the error of "AttributeError: 'NoneType' object has no attribute 'text'"
Code is below.
import pandas as pd
import xml.etree.ElementTree as et
xtree = et.parse("gamedata.xml")
xroot = xtree.getroot()
df_cols = ["GAME REF","YEAR", "GAME NO", "WIN LOSS"]
rows = []
for child in xroot.iter():
s_gameref = child.find('GAMEREF').text,
s_year = child.find('YEAR').text,
s_game_no = child.find('GAMENO').text,
s_winloss = child.find('WINLOSS').text
rows.append({"GAME REF": s_gameref,"YEAR": s_year,
"GAME NO": s_game_no, "WIN LOSS": s_winloss})
df = pd.DataFrame(rows, columns = df_cols)
The code is based off other stuff I have seen on the Stack and other sites, but nothing is working yet.
Ideal dataframe output is below
GAME REF
YEAR
GAME NO
WIN LOSS
1217
2021
1
W
1218
2021
2
W
1219
2021
3
L
1220
2021
4
L
Thanks
EDIT - NOT SURE WHAT IS GOING ON WITH MY TABLE, BUT IT SHOULD LOOK LIKE THIS
I think the below is what you are looking for. (Just loop over the "interesting" sub elements of Record). The logic of the code is in the line that starts with data = [.... The 2 loops can be found there.
import pandas as pd
import xml.etree.ElementTree as ET
xml = '''<r><Structure>
<Field>
<Field_Name>GAMEREF</Field_Name>
<Field_Type>Numeric</Field_Type>
<Field_Len>4</Field_Len>
<Field_Dec>0</Field_Dec>
</Field>
...
<Field>
<Field_Name>WINLOSS</Field_Name>
<Field_Type>Character</Field_Type>
<Field_Len>1</Field_Len>
<Field_Dec>0</Field_Dec>
</Field>
</Structure>
<Records>
<Record>
<GAMEREF>1217</GAMEREF>
<YEAR>2021</YEAR>
<GAMENO>1</GAMENO>
<WINLOSS>W</WINLOSS>
</Record>
<Record>
<GAMEREF>1220</GAMEREF>
<YEAR>2021</YEAR>
<GAMENO>4</GAMENO>
<WINLOSS>L</WINLOSS>
</Record>
</Records></r>'''
fields = {'GAMEREF':'GAME REF', 'YEAR':'YEAR', 'GAMENO':'GAME NO','WINLOSS':'WIN LOSS'}
root = ET.fromstring(xml)
data = [{display_name: rec.find(element_name).text for element_name,display_name in fields.items()} for rec in root.findall('.//Record')]
df = pd.DataFrame(data)
print(df)
output
GAME REF YEAR GAME NO WIN LOSS
0 1217 2021 1 W
1 1220 2021 4 L
import pandas as pd
import xml.etree.ElementTree as et
xtree = et.parse("gamedata.xml")
xroot = xtree.getroot()
df_cols = ["GAME REF","YEAR", "GAME NO", "WIN LOSS"]
rows = []
for record in xroot:
s_gameref = record.find('GAMEREF').text
s_year = record.find('YEAR').text
s_game_no = record.find('GAMENO').text
s_winloss = record.find('WINLOSS').text
rows.append({"GAME REF": s_gameref,"YEAR": s_year,
"GAME NO": s_game_no, "WIN LOSS": s_winloss})
df = pd.DataFrame(rows, columns = df_cols)
Remove .iter()
I'm using Jupyter Notebook and ElementTree (Python 3) to create a dataframe and save as csv from an XML file. Here is the XML format (in Estonian):
<asutused hetk="2020-04-14T03:53:33" ver="2">
<asutus>
<registrikood>10000515</registrikood>
<nimi>Osaühing B.Braun Medical</nimi>
<aadress />
<tegevusload>
<tegevusluba>
<tegevusloa_number>L04647</tegevusloa_number>
<alates>2019-12-10</alates>
<kuni />
<loaliik_kood>1</loaliik_kood>
<loaliik_nimi>Eriarstiabi</loaliik_nimi>
<haiglaliik_kood />
<haiglaliik_nimi />
<tegevuskohad>
<tegevuskoht>
<aadress>Harju maakond, Tallinn, Mustamäe linnaosa, J. Sütiste tee 17/1</aadress>
<teenused>
<teenus>
<kood>T0038</kood>
<nimi>ambulatoorsed üldkirurgiateenused</nimi>
</teenus>
<teenus>
<kood>T0236</kood>
<nimi>õe vastuvõtuteenus</nimi>
</teenus>
</teenused>
</tegevuskoht>
<tegevuskoht>
<aadress>Harju maakond, Tallinn, Mustamäe linnaosa, J. Sütiste tee 17/1</aadress>
<teenused>
<teenus>
<kood>T0038</kood>
<nimi>ambulatoorsed üldkirurgiateenused</nimi>
</teenus>
<teenus>
<kood>T0236</kood>
<nimi>õe vastuvõtuteenus</nimi>
</teenus>
</teenused>
</tegevuskoht>
</tegevuskohad>
</tegevusluba>
<tegevusluba>
<tegevusloa_number>L04651</tegevusloa_number>
<alates>2019-12-11</alates>
<kuni />
<loaliik_kood>2</loaliik_kood>
<loaliik_nimi>Õendusabi</loaliik_nimi>
<haiglaliik_kood />
<haiglaliik_nimi />
<tegevuskohad>
<tegevuskoht>
<aadress>Harju maakond, Tallinn, Mustamäe linnaosa, J. Sütiste tee 17/1</aadress>
<teenused>
<teenus>
<kood>T0038</kood>
<nimi>ambulatoorsed üldkirurgiateenused</nimi>
</teenus>
<teenus>
<kood>T0236</kood>
<nimi>õe vastuvõtuteenus</nimi>
</teenus>
</teenused>
</tegevuskoht>
<tegevuskoht>
<aadress>Harju maakond, Tallinn, Mustamäe linnaosa, J. Sütiste tee 17/1</aadress>
<teenused>
<teenus>
<kood>T0038</kood>
<nimi>ambulatoorsed üldkirurgiateenused</nimi>
</teenus>
<teenus>
<kood>T0236</kood>
<nimi>õe vastuvõtuteenus</nimi>
</teenus>
</teenused>
</tegevuskoht>
</tegevuskohad>
</tegevusluba>
</tegevusload>
<tootajad>
<tootaja>
<kood>D03091</kood>
<eesnimi>Evo</eesnimi>
<perenimi>Kaha</perenimi>
<kutse_kood>11</kutse_kood>
<kutse_nimi>Arst</kutse_nimi>
<erialad>
<eriala>
<kood>E420</kood>
<nimi>üldkirurgia</nimi>
</eriala>
</erialad>
</tootaja>
<tootaja>
<kood>N01146</kood>
<eesnimi>Karmen</eesnimi>
<perenimi>Mežulis</perenimi>
<kutse_kood>15</kutse_kood>
<kutse_nimi>Õde</kutse_nimi>
</tootaja>
<tootaja>
<kood>N01153</kood>
<eesnimi>Nele</eesnimi>
<perenimi>Terras</perenimi>
<kutse_kood>15</kutse_kood>
<kutse_nimi>Õde</kutse_nimi>
</tootaja>
<tootaja>
<kood>N02767</kood>
<eesnimi>Helena</eesnimi>
<perenimi>Tern</perenimi>
<kutse_kood>15</kutse_kood>
<kutse_nimi>Õde</kutse_nimi>
</tootaja>
<tootaja>
<kood>N12882</kood>
<eesnimi>Hanna</eesnimi>
<perenimi>Leemet</perenimi>
<kutse_kood>15</kutse_kood>
<kutse_nimi>Õde</kutse_nimi>
</tootaja>
</tootajad>
</asutus>
</asutused>
Each "asutus" is a hospital and I need some of the information inside. Here is my code:
tree = ET.parse("od_asutused.xml")
root = tree.getroot()
# open a file for writing
data = open('EE.csv', 'w')
# create the csv writer object
csvwriter = csv.writer(data, delimiter=';')
head = []
count = 0
for member in root.findall('asutus'):
hospital = []
if count == 0:
ident = member.find('registrikood').tag
head.append(id)
name = member.find('nimi').tag
head.append(name)
address = member.find('aadress').tag
head.append(address)
facility_type = member.find('./tegevusload/tegevusluba/haiglaliik_nimi').tag
head.append(facility_type)
site_address = member.find('./tegevusload/tegevusluba/tegevuskohad/tegevuskoht/aadress').tag
head.append(site_address)
for elem in member.findall('tegevusload'):
list_specs = elem.find('./tegevusluba/tegevuskohad/tegevuskoht/teenused/teenus/nimi').tag
head.append(list_specs)
csvwriter.writerow(head)
count = count + 1
ident = member.find('registrikood').text
hospital.append(ident)
name = member.find('nimi').text
hospital.append(name)
address = member.find('aadress').text
hospital.append(address)
facility_type = member.find('./tegevusload/tegevusluba/haiglaliik_nimi').text
hospital.append(facility_type)
site_address = member.find('./tegevusload/tegevusluba/tegevuskohad/tegevuskoht/aadress').text
hospital.append(site_address)
for spec in elem.findall('tegevusload'):
list_specs = spec.find('./tegevusluba/tegevuskohad/tegevuskoht/teenused/teenus/nimi').text
hospital.append(list_specs)
csvwriter.writerow(hospital)
data.close()
#Upload csv for geocoding
df = pd.read_csv(r'EE.csv', na_filter= False, delimiter=';')
#Rename columns
df.rename(columns = {'<built-in function id>':'id',
'nimi':'name',
'aadress':'address',
'haiglaliik_nimi':'facility_type',
'haiglaliik_kood':'facility_type_c',
'aadress.1':'site_address',
'nimi.1':'list_specs'},
inplace = True)
#Add columns
df['country'] = 'Estonia'
df['cc'] = 'EE'
df.head(10)
And the result of the df.head(10):
Result of dataframe
The "list_specs" is blank no matter what I do. How can I populate this field with a list of each 'nimi' for each site address? Thank you.
I found in your code the following points to change:
At least on my computer, calling csv.writer causes that newline chars
are doubled. The remedy I found is to open the output file with
additional parameters:
data = open('EE.csv', 'w', newline='\n', encoding='utf-8')
There is no sense to write head with Estonian column names and then
rename the columns. Note also that in head.append(id) you use an undeclared
variable (id).
But this is not so important, as I changed this whole section with writing
target column names (see below).
As you write the CSV file to be read by read_csv, it should contain a
fixed number of columns. So it is a bad practice to use a loop to write
one element.
Your instruction list_specs = elem.findall(...) was wrong, because
elem is not set in the current loop. Instead you should use member (but
I solved this detail other way).
There is no sense to create a variable only in order to use it once.
More concise and readable code is e.g. hospital.append(member.findtext('nimi')).
To avoid long XPath expressions, with repeated initial part, I decided
to set a temporary variable "in the middle" of this path, e.g.
tgvLb = member.find('tegevusload/tegevusluba') and then use a relative
XPath starting from this node.
Your rename instruction contains one not needed column, namely facility_type_c. You read only 6 columns, not 7.
So change the middle part of your code to:
data = open('EE.csv', 'w', newline='\n', encoding='utf-8')
csvwriter = csv.writer(data, delimiter=';')
head = ['id', 'name', 'address', 'facility_type', 'site_address', 'list_specs']
csvwriter.writerow(head)
for member in root.findall('asutus'):
hospital = []
hospital.append(member.findtext('registrikood'))
hospital.append(member.findtext('nimi'))
hospital.append(member.findtext('aadress'))
tgvLb = member.find('tegevusload/tegevusluba')
hospital.append(tgvLb.findtext('haiglaliik_nimi'))
tgvKoht = tgvLb.find('tegevuskohad/tegevuskoht')
hospital.append(tgvKoht.findtext('aadress'))
hospital.append(tgvKoht.findtext('teenused/teenus/nimi'))
csvwriter.writerow(hospital)
data.close()
df = pd.read_csv(r'EE.csv', na_filter= False, delimiter=';')
and drop df.rename from your code.
My goal is to convert an .XML file into a .CSV file.
This part of the code is already functional.
However, I also want to extract the sub-sub-nodes of one of the "father" nodes.
Maybe an example would be more self explanatory;
Here is the structure of my XML:
<nedisCatalogue>
<headerInfo>
<feedVersion>1-0</feedVersion>
<dateCreated>2018-01-22T23:37:01+0100</dateCreated>
<supplier>Nedis_BENED</supplier>
<locale>nl_BE</locale>
</headerInfo>
<productList>
<product>
<nedisPartnr><![CDATA[VS-150/63BA]]></nedisPartnr>
<nedisArtlid>17005</nedisArtlid>
<vendorPartnr><![CDATA[TONFREQ-ELKOS / BIPOL 150, 5390]]></vendorPartnr>
<brand><![CDATA[Visaton]]></brand>
<EAN>4007540053905</EAN>
<intrastatCode>8532220000</intrastatCode>
<UNSPSC>52161514</UNSPSC>
<headerText><![CDATA[Crossover Foil capacitor]]></headerText>
<internetText><![CDATA[Bipolaire elco met een ruwe folie en een zeer goede prijs/kwaliteits-verhouding voor de bouw van cross-overs. 63 Vdc, 10% tolerantie.]]></internetText>
<generalText><![CDATA[Dimensions 16 x 35 mm
]]></generalText>
<images>
<image type="2" order="15">767736.JPG</image>
</images>
<attachments>
</attachments>
<categories>
<tree name="Internet_Tree_ISHP">
<entry depth="001" id="1067858"><![CDATA[Audio]]></entry>
<entry depth="002" id="1067945"><![CDATA[Speakers]]></entry>
<entry depth="003" id="1068470"><![CDATA[Accessoires]]></entry>
</tree>
</categories>
<properties>
<property id="360" multiplierID="" unitID="" valueID=""><![CDATA[...]]></property>
</properties>
<status>
<code status="NORMAL"></code>
</status>
<packaging quantity="1" weight="8"></packaging>
<introductionDate>2015-10-26</introductionDate>
<serialnumberKeeping>N</serialnumberKeeping>
<priceLevels>
<normalPricing from="2017-02-13" to="2018-01-23">
<price level="1" moq="1" currency="EUR">2.48</price>
</normalPricing>
<specialOfferPricing></specialOfferPricing>
<goingPriceInclVAT currency="EUR" quantity="1">3.99</goingPriceInclVAT>
</priceLevels>
<tax>
</tax>
<stock>
<inStockLocal>25</inStockLocal>
<inStockCentral>25</inStockCentral>
<ATP>
<nextExpectedStockDateLocal></nextExpectedStockDateLocal>
<nextExpectedStockDateCentral></nextExpectedStockDateCentral>
</ATP>
</stock>
</product>
....
</nedisCatalogue>
And here is the code that I have now:
import xml.etree.ElementTree as ET
import csv
tree = ET.parse("/Users/BE07861/Documents/nedis_catalog_2018-01-23_nl_BE_32191_v1-0_xml")
root = tree.getroot()
f = open('/Users/BE07861/Documents/test2.csv', 'w')
csvwriter = csv.writer(f, delimiter='ç')
count = 0
head = ['Nedis Part Number', 'Nedis Article ID', 'Vendor Part Number', 'Brand', 'EAN', 'Header text', 'Internet Text', 'General Text', 'categories']
prdlist = root[1]
prdct = prdlist[5]
cat = prdct[12]
tree1=cat[0]
csvwriter.writerow(head)
for time in prdlist.findall('product'):
row = []
nedis_number = time.find('nedisPartnr').text
row.append(nedis_number)
nedis_art_id = time.find('nedisArtlid').text
row.append(nedis_art_id)
vendor_part_nbr = time.find('vendorPartnr').text
row.append(vendor_part_nbr)
Brand = time.find('brand').text
row.append(Brand)
ean = time.find('EAN').text
row.append(ean)
header_text = time.find('headerText').text
row.append(header_text)
internet_text = time.find('internetText').text
row.append(internet_text)
general_text = time.find('generalText').text
row.append(general_text)
categ = time.find('categories').find('tree').find('entry').text
row.append(categ)
csvwriter.writerow(row)
f.close()
If you run the code, you'll see that I only retrieve the first "entry" of the categories/tree; which is normal. However, I don't know how to create a loop that, for every node "categories", creates new columns such as categories1, categories2 & categories3 with the values: "entry".
My result should look like this
Nedis Part Number Nedis Article ID Vendor Part Number
VS-150/63BA 17005 TONFREQ-ELKOS / BIPOL 150, 5390
Brand EAN Header text Internet Text
Visaton 4,00754E+12 Crossover Foil capacitor Bipolaire elco …
General Text Category1 Categroy2 Category3
Dimensions 16 x 35 mm Audio Speakers Accessoires
I've really tried my best but didn't manage to find the solution.
Any help would be very much appreciated!!! :)
Thanks a lot,
Allan
I think this is what you're looking for:
for child in time.find('categories').find('tree'):
categ = child.text
row.append(categ)
Here's a solution that loops through the xml once to figure out how many headers to add, adds the headers, and then loops through each product's category list:
**Updated to iterate through images in addition to categories. This is the biggest difference:
for child in time.find('categories').find('tree'):
categ = child.text
row.append(categ)
curcat += 1
while curcat < maxcat:
row.append('')
curcat += 1
It's going to figure out the maximum number of categories on a single record and then and that many columns. If a particular record has less categories, this code will stick blank values in as placeholders so the column headers always line up with the data.
For instance:
Cat1 Cat2 Cat3 Img1 Img2 Img3
A B C 1 2 3
D E <blank> 4 <blank> <blank>
Here's the full solution:
import xml.etree.ElementTree as ET
import csv
tree = ET.parse("c:\\python\\xml.xml")
root = tree.getroot()
f = open('c:\\python\\xml.csv', 'w')
csvwriter = csv.writer(f, delimiter=',')
count = 0
head = ['Nedis Part Number', 'Nedis Article ID', 'Vendor Part Number', 'Brand', 'EAN', 'Header text', 'Internet Text', 'General Text']
prdlist = root[1]
maxcat = 0
for time in prdlist.findall('product'):
cur = 0
for child in time.find('categories').find('tree'):
cur += 1
if cur > maxcat:
maxcat = cur
for cnt in range (0, maxcat):
head.append('Category ' + str(cnt + 1))
maximg = 0
for time in prdlist.findall('product'):
cur = 0
for child in time.find('images'):
cur += 1
if cur > maximg:
maximg = cur
for cnt in range(0, maximg):
head.append('Image ' + str(cnt + 1))
csvwriter.writerow(head)
for time in prdlist.findall('product'):
row = []
nedis_number = time.find('nedisPartnr').text
row.append(nedis_number)
nedis_art_id = time.find('nedisArtlid').text
row.append(nedis_art_id)
vendor_part_nbr = time.find('vendorPartnr').text
row.append(vendor_part_nbr)
Brand = time.find('brand').text
row.append(Brand)
ean = time.find('EAN').text
row.append(ean)
header_text = time.find('headerText').text
row.append(header_text)
internet_text = time.find('internetText').text
row.append(internet_text)
general_text = time.find('generalText').text
row.append(general_text)
curcat = 0
for child in time.find('categories').find('tree'):
categ = child.text
row.append(categ)
curcat += 1
while curcat < maxcat:
row.append('')
curcat += 1
curimg = 0
for img in time.find('images'):
image = img.text
row.append(image)
curimg += 1
while curimg < maximg:
row.append('')
curimg += 1
csvwriter.writerow(row)
f.close()
My python 3 script takes an xml file and creates a csv file.
Small excerpt of xml file:
<?xml version="1.0" encoding="UTF-8" ?>
<metadata>
<dc>
<title>Golden days for boys and girls, 1895-03-16, v. XVI #17</title>
<subject>Children's literature--Children's periodicals</subject>
<description>Archives & Special Collections at the Thomas J. Dodd Research Center, University of Connecticut Libraries</description>
<publisher>James Elverson, 1880-</publisher>
<date>1895-06-15</date>
<type>Text | periodicals</type>
<format>image/jp2</format>
<handle>http://hdl.handle.net/11134/20002:860074494</handle>
<accessionNumber/>
<barcode/>
<identifier>20002:860074494 | local: 868010272 | local: 997186613502432 | local: 39153019382870 | hdl: | http://hdl.handle.net/11134/20002:860074494</identifier>
<rights>These Materials are provided for educational and research purposes only. The University of Connecticut Libraries hold the copyright except where noted. Permission must be obtained in writing from the University of Connecticut Libraries and/or theowner(s) of the copyright to publish reproductions or quotations beyond "fair use." | The collection is open and available for research.</rights>
<creator/>
<relation/>
<coverage/>
<language/>
</dc>
</metadata>
Python3 code:
import csv
import xml.etree.ElementTree as ET
tree = ET.ElementTree(file='ctda_set1_uniqueTags.xml')
doc = ET.parse("ctda_set1_uniqueTags.xml")
root = tree.getroot()
oaidc_data = open('ctda_set1_uniqueTags.csv', 'w', encoding='utf-8')
titles = 'dc/title'
subjects = 'dc/subject'
csvwriter = csv.writer(oaidc_data)
oaidc_head = ['Title', 'Subject', 'Description', 'Publisher', 'Date', 'Type', 'Format', 'Handle', 'Accession Number', 'Barcode', 'Identifiers', 'Rights', 'Creator', 'Relation', 'Coverage', 'Language']
count = 0
for member in root.findall('dc'):
if count == 0:
csvwriter.writerow(oaidc_head)
count = count + 1
dcdata = []
titles = member.find('title').text
dcdata.append(titles)
subjects = member.find('subject').text
dcdata.append(subjects)
descriptions = member.find('description').text
dcdata.append(descriptions)
publishers = member.find('publisher').text
dcdata.append(publishers)
dates = member.find('date').text
dcdata.append(dates)
types = member.find('type').text
dcdata.append(types)
formats = member.find('format').text
dcdata.append(formats)
handle = member.find('handle').text
dcdata.append(handle)
accessionNo = member.find('accessionNumber').text
dcdata.append(accessionNo)
barcodes = member.find('barcode').text
dcdata.append(barcodes)
identifiers = member.find('identifier').text
dcdata.append(identifiers)
rt = member.find('rights').text
print(member.find('rights').text)
dcdata.append('rt')
ct = member.find('creator').text
dcdata.append('ct')
rt = member.find('relation').text
dcdata.append('rt')
ce = member.find('coverage').text
dcdata.append('ce')
lang = member.find('language').text
dcdata.append('lang')
csvwriter.writerow(dcdata)
oaidc_data.close()
Everything works as expected except for rt, ce, and lang. What happens is that in the csv, all the data is written with the comma delimiter. For rt, the value is always rt, for ce, ce, lang, lang, etc.
Here's a snippet of the output:
Title,Subject,Description,Publisher,Date,Type,Format,Handle,Accession Number,Barcode,Identifiers,Rights,Creator,Relation,Coverage,Language
"Golden days for boys and girls, 1895-03-16, v. XVI #17",Children's literature--Children's periodicals,"Archives & Special Collections at the Thomas J. Dodd Research Center, University of Connecticut Libraries","James Elverson, 1880-",1895-06-15,Text | periodicals,image/jp2,hdl.handle.net/11134/20002:860074494,,,20002:860074494 | local: 868010272 | local: 997186613502432 | local: 39153019382870,**rt,ct,rt,ce,lang**
Some of the rights statements get very long - perhaps that's the issue. That's why I added the print(member.find('rights')) to see the output. The text is printed just fine. The text just isn't written to the csv. What I'd like is to have the value or text written for these xml tags. Any help would be appreciated.
Thanks.
Jennifer
In the line dcdata.append('rt') there is no need for the quotes. Try dcdata.append(rt). Similarly, there are unnecessary quotes in the ce and lang lines.
I'm trying to add edges between nodes.
I have a text file which I have put into a list.
The first list contains this:
Title , Rating
[('"$weepstake$" (1979) {(#1.2)}', '10.0'),
('"\'Til Death Do Us Part" (2006) {Pilot(#1.0)}', '3.7'),
('"\'Conversations with My Wife\'" (2010)', '4.2'),
('"\'Da Kink in My Hair" (2007)', '4.2').....much more here ]
I want to create nodes labeled with all the titles and when two titles have the same rating, then I want to create an edge between them, so I - in the end - get all titles with rating 10.0 together in one network and so on.
My code so far:
import networkx as nx
import string
from sys import maxint
import csv
import pprint
import re
def printStuff(labels,dG):
for index, node in enumerate(dG.nodes()):
print '%s:%d\n' % (labels[index],dG.node[node]['count'])
str1 = titleList
#print str1
get_user_info = titleList1
dG = nx.DiGraph()
for i, word in enumerate(str1):
try:
next_word = str1[i]
if not dG.has_node(word):
dG.add_node(word)
dG.node[word]['count'] = 1
else:
dG.node[word]['count'] += 1
if not dG.has_node(next_word):
dG.add_node(next_word)
dG.node[next_word]['count'] = 0
if not dG.has_edge(word, next_word):
dG.add_edge(word, next_word, weight=0)
else:
dG.edge[word][next_word]['weight'] += 1
except IndexError:
if not dG.has_node(word):
dG.add_node(word)
dG.node[word]['count'] = 1
else:
dG.node[word]['count'] += 1
except:
raise
printStuff(titleList, dG)
Output:
10.0:1
10.0:1
3.7:1
10.0:1
3.7:1
4.2:1
10.0:1
3.7:1
4.2:1
4.2:1
And for edges:
for edge in dG.edges():
print '%s:%d\n' % (edge,dG.edge[edge[0]][edge[1]]['weight'])
Output:
(('"\'Conversations with My Wife\'" (2010)', '4.2'), ('"\'Conversations with My Wife\'" (2010)', '4.2')):0
(('"\'Da Kink in My Hair" (2007)', '4.2'), ('"\'Da Kink in My Hair" (2007)', '4.2')):0
(('"$weepstake$" (1979) {(#1.2)}', '10.0'), ('"$weepstake$" (1979) {(#1.2)}', '10.0')):0
(('"\'Til Death Do Us Part" (2006) {Pilot (#1.0)}', '3.7'), ('"\'Til Death Do Us Part" (2006) {Pilot (#1.0)}', '3.7')):0
How about this:
data = [('"$weepstake$" (1979) {(#1.2)}', '10.0'),
('"\'Til Death Do Us Part" (2006) {Pilot(#1.0)}', '3.7'),
('"\'Conversations with My Wife\'" (2010)', '4.2'),
('"\'Da Kink in My Hair" (2007)', '4.2')]
import networkx as nx
G = nx.Graph()
G.add_edges_from(data)
nx.draw(G)
if you want a count of edges from a score.
len(G.edges('4.2'))
2