python lxml - 循环/遍历excel行并将每行保存为一个xml

Question

the problem is that the 2nd xml file contains also the data from the first iteration of the excel row and the third xml file every data from the first and 2nd rows

Working since hours on that and cant figure it out

from lxml import etree
import openpyxl


# Create root element with namespace information
xmlns = "http://xml.datev.de/bedi/tps/ledger/v040"
xsi = "http://www.w3.org/2001/XMLSchema-instance"
schemaLocation = "http://xml.datev.de/bedi/tps/ledger/v040 Belegverwaltung_online_ledger_import_v040.xsd"
version = "4.0"
generator_info = "DATEV Musterdaten"
generating_system = "DATEV manuell"

xmlRoot = etree.Element(
    "{" + xmlns + "}LedgerImport",
    version=version,
    attrib={"{" + xsi + "}schemaLocation": schemaLocation},
    generator_info=generator_info,
    generating_system=generating_system,
    nsmap={'xsi': xsi, None: xmlns}
)

####open excel file speadsheet
wb = openpyxl.load_workbook('import_spendesk_datev.xlsx')
sheet = wb['Import']

# build the xml tree
for i in range(2,6):
        consolidate = etree.SubElement(xmlRoot, 'consolidate', attrib={'consolidatedAmount': str(sheet.cell(row=i,column=16).value),'consolidatedDate': str(sheet.cell(row=i,column=2).value), 'consolidatedInvoiceId': str(sheet.cell(row=i,column=13).value), 'consolidatedCurrencyCode': str(sheet.cell(row=i,column=12).value) })
        accountsPayableLedger = etree.SubElement(consolidate, 'accountsPayableLedger')
        account = etree.SubElement(accountsPayableLedger, 'bookingText')
        account.text = sheet.cell(row=i,column=21).value
        invoice = etree.SubElement(accountsPayableLedger, 'invoiceId')
        invoice.text = sheet.cell(row=i,column=13).value
        date = etree.SubElement(accountsPayableLedger, 'date')
        date.text = sheet.cell(row=i,column=2).value
        amount = etree.SubElement(accountsPayableLedger, 'amount')
        amount.text = sheet.cell(row=i,column=16).value
        account_no = etree.SubElement(accountsPayableLedger, 'accountNo')
        account_no.text = sheet.cell(row=i,column=19).value
        cost1 = etree.SubElement(accountsPayableLedger, 'costCategoryId')
        cost1.text = sheet.cell(row=i,column=15).value
        currency_code = etree.SubElement(accountsPayableLedger, 'currencyCode')
        currency_code.text = sheet.cell(row=i,column=12).value
        party_id = etree.SubElement(accountsPayableLedger, 'partyId')
        party_id.text = sheet.cell(row=i,column=20).value
        bpaccount = etree.SubElement(accountsPayableLedger, 'bpAccountNo')
        bpaccount.text = sheet.cell(row=i,column=20).value
        doc = etree.ElementTree(xmlRoot)
        doc.write( str(sheet.cell(row=i,column=13).value)+".xml", xml_declaration=True, encoding='utf-8', pretty_print=True)

as described this for every single excel row and for each row one .xml file

<?xml version='1.0' encoding='UTF-8'?>
<LedgerImport xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns="http://xml.datev.de/bedi/tps/ledger/v040" generating_system="DATEV manuell" generator_info="DATEV Musterdaten" version="4.0" xsi:schemaLocation="http://xml.datev.de/bedi/tps/ledger/v040 Belegverwaltung_online_ledger_import_v040.xsd">
  <consolidate consolidatedAmount="1337.01">
    <accountsPayableLedger>
      <bookingText>amazon</bookingText>
      <invoiceId>1</invoiceId>
    </accountsPayableLedger>
  </consolidate>
</LedgerImport>

Answer 1

The same xmlRoot object is reused several times. You need to create a new root element for each iteration in the for loop.

The code that creates the root element can be put in a function. Here is a simplified example:

from lxml import etree

def makeroot():
    return etree.Element("LedgerImport")

for i in range(2, 6):
    xmlRoot = makeroot()
    consolidate = etree.SubElement(xmlRoot, 'consolidate',
                                   attrib={'consolidatedAmount': str(i)})
    doc = etree.ElementTree(xmlRoot)
    doc.write(str(i) + ".xml", xml_declaration=True, encoding='utf-8', pretty_print=True)

Answer 2

After @mzjn pointed out your basic mistake, here is a thing I made for fun - you can create nested XML with a declarative mapping, instead of laboriously calling etree.SubElement yourself.

Here is how. Assume this as the basic situation:

from lxml import etree
import openpyxl

ns = {
    None:  'http://xml.datev.de/bedi/tps/ledger/v040',
    'xsi': 'http://www.w3.org/2001/XMLSchema-instance',
}

mapping = {
    '_tag': '{' + ns[None] + '}LedgerImport',
    'attrib': {
        'version': '4.0',
        '{' + ns['xsi'] + '}schemaLocation': 'http://xml.datev.de/bedi/tps/ledger/v040 Belegverwaltung_online_ledger_import_v040.xsd',
        'generator_info': 'DATEV Musterdaten',
        'generating_system': 'DATEV manuell',
    },
    'nsmap': ns,
    '_children': [{
        '_tag': 'consolidate',
        'attrib': {
            'consolidatedAmount': lambda: sheet.cell(i, 16).value,
            'consolidatedDate': lambda: sheet.cell(i, 2).value,
            'consolidatedInvoiceId': lambda: sheet.cell(i, 13).value,
            'consolidatedCurrencyCode': lambda: sheet.cell(i, 12).value,
        },
        '_children': [{
            '_tag': 'accountsPayableLedger',
            '_children': [
                {'_tag': 'bookingText', '_text': lambda: sheet.cell(i, 21).value},
                {'_tag': 'invoiceId', '_text': lambda: sheet.cell(i, 13).value},
                {'_tag': 'date', '_text': lambda: sheet.cell(i, 2).value},
                {'_tag': 'amount', '_text': lambda: sheet.cell(i, 16).value},
                {'_tag': 'accountNo', '_text': lambda: sheet.cell(i, 19).value},
                {'_tag': 'costCategoryId', '_text': lambda: sheet.cell(i, 15).value},
                {'_tag': 'currencyCode', '_text': lambda: sheet.cell(i, 12).value},
                {'_tag': 'partyId', '_text': lambda: sheet.cell(i, 20).value},
                {'_tag': 'bpAccountNo', '_text': lambda: sheet.cell(i, 20).value},
            ]
        }]
    }],
}

The nested dict resembles your final XML document. Its keys also resemble the parameters that etree.Element() and etree.SubElement() take, with the addition of _text and _children .

Now we can define a single recursive helper function that takes this input tree and transforms it into a nested XML tree of the same configuration. As a bonus we can execute the lambda functions, which allows us to dynamically calculate attribute values and text:

def build_tree(template, parent=None):
    # prepare a dict for calling etree.Element()/etree.SubElement()
    params = {k: v for k, v in template.items() if k not in ['_children', '_text']}

    # calculate any dynamic attribute values
    for name in params.get('attrib', {}):
        value = params['attrib'][name]
        params['attrib'][name] = str(value() if callable(value) else value)

    if parent is None:
        node = etree.Element(**params)    
    else:
        params['_parent'] = parent
        node = etree.SubElement(**params)

    # calculate (if necessary) and set the node text
    if '_text' in template:
        if callable(template['_text']):
            node.text = str(template['_text']())
        else:
            node.text = str(template['_text']) if template['_text'] else template['_text']

    # recurse into children, if any    
    for child in template.get('_children', []):
        build_tree(child, node)

    return node

We can call this in a loop:

wb = openpyxl.load_workbook('import_spendesk_datev.xlsx')
sheet = wb['Import']

for i in range(2,6):
    root = build_tree(mapping)
    doc = etree.ElementTree(root)
    name = "%s.xml" % sheet.cell(i, 13).value
    doc.write(name, xml_declaration=True, encoding='utf-8', pretty_print=True)

This should generate a couple of nicely nested XML documents, and it should be a lot easier to manage if your XML structure changes or gets more complicated.

Answer 3

Alternatively, consider XSLT , the special-purpose declarative langauge designed to transform XML files, which lxml does support. Specifically, pass parameters from Python to the stylesheet to transform a template XML (not unlike passing parameters to a prepared SQL statement):

XML template (includes all top-level namespaces)

<?xml version='1.0' encoding='UTF-8'?>
<LedgerImport xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" 
              xmlns="http://xml.datev.de/bedi/tps/ledger/v040" 
              generating_system="DATEV manuell" 
              generator_info="DATEV Musterdaten" version="4.0" 
              xsi:schemaLocation="http://xml.datev.de/bedi/tps/ledger/v040 Belegverwaltung_online_ledger_import_v040.xsd">
  <consolidate consolidatedAmount="???">
    <accountsPayableLedger>
      <bookingText>???</bookingText>
      <invoiceId>???</invoiceId>
      <date>???</date>
      <amount>???</amount>
      <accountNo>???</accountNo>
      <costCategoryId>???</costCategoryId>
      <currencyCode>???</currencyCode>
      <partyId>???</partyId>
      <bpAccountNo>???</bpAccountNo>
    </accountsPayableLedger>
  </consolidate>
</LedgerImport>

XSLT (save as .xsl file, a little longer due to default namespace in XML)

<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
                              xmlns:doc="http://xml.datev.de/bedi/tps/ledger/v040">
  <xsl:output indent="yes"/>
  <xsl:strip-space elements="*"/>

  <!-- INITIALIZE PARAMETERS -->
  <xsl:param name="prm_consolidate" />
  <xsl:param name="prm_bookingText" /> 
  <xsl:param name="prm_invoiceId" /> 
  <xsl:param name="prm_date" /> 
  <xsl:param name="prm_amount" /> 
  <xsl:param name="prm_accountNo" /> 
  <xsl:param name="prm_costCategoryId" /> 
  <xsl:param name="prm_currencyCode" /> 
  <xsl:param name="prm_partyId" /> 
  <xsl:param name="prm_bpAccountNo" /> 

  <!-- IDENTITY TRANSFORM -->
  <xsl:template match="@*|node()">
    <xsl:copy>
      <xsl:apply-templates select="@*|node()"/>
    </xsl:copy>
  </xsl:template>

  <!-- REWRITE TITLE TEXT -->
  <xsl:template match="doc:accountsPayableLedger">
    <xsl:copy>
      <xsl:element name="consolidate" namespace="http://xml.datev.de/bedi/tps/ledger/v040">
           <xsl:attribute name="consolidatedAmount"><xsl:value-of select="$prm_consolidate"/></xsl:attribute>
      </xsl:element>
      <xsl:element name="bookingText" namespace="http://xml.datev.de/bedi/tps/ledger/v040"><xsl:value-of select="$prm_bookingText"/></xsl:element>
      <xsl:element name="invoiceId" namespace="http://xml.datev.de/bedi/tps/ledger/v040"><xsl:value-of select="$prm_invoiceId"/></xsl:element>
      <xsl:element name="date" namespace="http://xml.datev.de/bedi/tps/ledger/v040"><xsl:value-of select="$prm_date"/></xsl:element>
      <xsl:element name="amount" namespace="http://xml.datev.de/bedi/tps/ledger/v040"><xsl:value-of select="$prm_amount"/></xsl:element>
      <xsl:element name="accountNo" namespace="http://xml.datev.de/bedi/tps/ledger/v040"><xsl:value-of select="$prm_accountNo"/></xsl:element>
      <xsl:element name="costCategoryId" namespace="http://xml.datev.de/bedi/tps/ledger/v040"><xsl:value-of select="$prm_costCategoryId"/></xsl:element>
      <xsl:element name="currencyCode" namespace="http://xml.datev.de/bedi/tps/ledger/v040"><xsl:value-of select="$prm_currencyCode"/></xsl:element>
      <xsl:element name="partyId" namespace="http://xml.datev.de/bedi/tps/ledger/v040"><xsl:value-of select="$prm_partyId"/></xsl:element>
      <xsl:element name="bpAccountNo" namespace="http://xml.datev.de/bedi/tps/ledger/v040"><xsl:value-of select="$prm_bpAccountNo"/></xsl:element>
    </xsl:copy>
  </xsl:template>

</xsl:stylesheet>

Python (no DOM element building)

import lxml.etree as et

# LOAD XML AND XSL
xml = et.parse('/path/to/Template.xml')
xsl = et.parse('/path/to/XSLTScript.xsl')

### OPEN EXCEL SPREADSHEET
wb = openpyxl.load_workbook('import_spendesk_datev.xlsx')
sheet = wb['Import']

# LOOP THROUGH ROWS
for i in range(2, 6):
   consolidate = et.XSLT.strparam(sheet.cell(row=i,column=16).value)
   account = et.XSLT.strparam(sheet.cell(row=i,column=21).value)
   invoice = et.XSLT.strparam(sheet.cell(row=i,column=13).value)
   date = et.XSLT.strparam(sheet.cell(row=i,column=2).value)
   amount = et.XSLT.strparam(sheet.cell(row=i,column=16).value)
   account_no = et.XSLT.strparam(sheet.cell(row=i,column=19).value)
   cost1 = et.XSLT.strparam(sheet.cell(row=i,column=15).value)
   currency_code = et.XSLT.strparam(sheet.cell(row=i,column=12).value)
   party_id = et.XSLT.strparam(sheet.cell(row=i,column=20).value)
   bpaccount = et.XSLT.strparam(sheet.cell(row=i,column=20).value)

   # PASS PARAMETER TO XSLT
   transform = et.XSLT(xsl)
   result = transform(xml, prm_consolidate = consolidate,
                           prm_bookingText=account,       
                           prm_invoiceId = invoice,
                           prm_date = date,
                           prm_amount = amount,
                           prm_account_no = account_no,
                           prm_costCategoryId = cost1,
                           prm_currencyCode = currency_code,
                           prm_partyId = party_id,
                           prm_bpAccountNo = bpaccount)

   # SAVE XML TO FILE
   with open('/path/to/Output_Row{}.xml'.format(i), 'wb') as f:
       f.write(result)