C# Calculate SHA256 value for SEPA (XML) paymentfile

Question

in the specification of the "DFÜ agreement" is described to calculate the hashvalue for a SEPA container payment file by SHA256.

The hash value is created using the entire contained document, including the opening and closing "document" tag.

The document is canonised according to Canonical XML, version 1.0. (http://www.w3.org/TR/2001/REC-xml-c14n-20010315).

In the case of included documents, the canonisation has also to be executed according to the main document.

SHA-256 is used as hash algorithm.

The hash value is entered in hexadecimal form in the tag, capital characters are used for the hexadecimal digits A to F. When using an XML container within the SRZ procedure it is mandatory to specify the hash value (the abbreviation SRZ stands for the German term „Servicerechenzentrum“ meaning “data processing service centre”).

I tryed different ways to do that but i am not able to get the correct value.

    [Test]
    public void GetHashTest()
    {
        // load document
        XmlDocument sepaContainer = new XmlDocument();
        sepaContainer.PreserveWhitespace = true;
        sepaContainer.Load("PathToFile");

        // do canonical 
        XmlDsigC14NTransform transformer = new XmlDsigC14NTransform();
        transformer.LoadInput(document);

        string compare = string.Empty;
        foreach (byte b in transformer.GetDigestedOutput(new SHA256Managed()))
        {
            compare += b.ToString("X2");
        }

        Assert.That(compare, Is.EqualTo("1FA8EE4F1E9551C82E1C7A82A88140325453A52BD08FF9FA5D13CA40F04AB305"));
    }

Also not working to read the file into a string and calculate:

    private string GetSHA256(string text)
    {
        UTF8Encoding UE = new UTF8Encoding();
        byte[] message = UE.GetBytes(text);

        SHA256Managed hashString = new SHA256Managed();
        string hex = string.Empty;

        byte[] hashValue = hashString.ComputeHash(message);
        foreach (byte x in hashValue)
        {
            hex += string.Format("{0:X2}", x);
        }

        return hex;
    }

Currently i work with the SEPA XML container from here: http://entwickler-forum.de/showthread.php/64499-Auslesen-einer-XML-Datei

<?xml version="1.0" encoding="UTF-8"?>
<conxml xmlns="urn:conxml:xsd:container.nnn.002" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="urn:conxml:xsd:container.nnn.002 container.nnn.002.xsd">
  <CreDtTm>2010-10-26T14:48:17Z</CreDtTm>
  <MsgPain008>
    <HashValue>38B862BD35B95D8CB20448153B5F6F73B3657EB1BC81CF11D3491D16EEA94009</HashValue>
    <HashAlgorithm>SHA256</HashAlgorithm>
    <Document xmlns="urn:swift:xsd:$pain.008.002.01">
      <pain.008.001.01>
        <GrpHdr>
          <MsgId>D004201010261648081</MsgId>
          <CreDtTm>2010-10-26T14:48:08Z</CreDtTm>
          <NbOfTxs>1</NbOfTxs>
          <CtrlSum>0.20</CtrlSum>
          <Grpg>MIXD</Grpg>
          <InitgPty>
            <Nm>Testauftraggeber SEPA-LS</Nm>
          </InitgPty>
        </GrpHdr>
        <PmtInf>
          <PmtInfId>D00420101024444444081</PmtInfId>
          <PmtMtd>DD</PmtMtd>
          <PmtTpInf>
            <SvcLvl>
              <Cd>SEPA</Cd>
            </SvcLvl>
            <LclInstrm>
              <Cd>CORE</Cd>
            </LclInstrm>
            <SeqTp>OOFF</SeqTp>
          </PmtTpInf>
          <ReqdColltnDt>2010-11-02</ReqdColltnDt>
          <Cdtr>
            <Nm>Testauftraggeber SEPA-LS</Nm>
          </Cdtr>
          <CdtrAcct>
            <Id>
              <IBAN>DE352505018444402014</IBAN>
            </Id>
          </CdtrAcct>
          <CdtrAgt>
            <FinInstnId>
              <BIC>SPKHDE2HXXX</BIC>
            </FinInstnId>
          </CdtrAgt>
          <ChrgBr>SLEV</ChrgBr>
          <DrctDbtTxInf>
            <PmtId>
              <EndToEndId>Lastschriftreferenz 002</EndToEndId>
            </PmtId>
            <InstdAmt Ccy="EUR">0.20</InstdAmt>
            <DrctDbtTx>
              <MndtRltdInf>
                <MndtId>Mandat002</MndtId>
                <DtOfSgntr>2010-10-26</DtOfSgntr>
                <AmdmntInd>false</AmdmntInd>
              </MndtRltdInf>
              <CdtrSchmeId>
                <Id>
                  <PrvtId>
                    <OthrId>
                      <Id>DE98ZZZ09999999999</Id>
                      <IdTp>SEPA</IdTp>
                    </OthrId>
                  </PrvtId>
                </Id>
              </CdtrSchmeId>
            </DrctDbtTx>
            <DbtrAgt>
              <FinInstnId>
                <BIC>SPKHDE2HXXX</BIC>
              </FinInstnId>
            </DbtrAgt>
            <Dbtr>
              <Nm>Test Zahlungspflichtiger</Nm>
            </Dbtr>
            <DbtrAcct>
              <Id>
                <IBAN>DE132505018005555552</IBAN>
              </Id>
            </DbtrAcct>
            <UltmtDbtr>
              <Nm>Test Miete</Nm>
            </UltmtDbtr>
            <RmtInf>
              <Ustrd>Test SEPA-LS Einzug durch 10002014 Zahlungspflichtiger 10002022</Ustrd>
            </RmtInf>
          </DrctDbtTxInf>
        </PmtInf>
      </pain.008.001.01>
    </Document>
  </MsgPain008>
</conxml>

I extracted the "Document"-node by hand and by xpath but still got a different value then there is calculated for the check hash.

Hopefully any of you has an idea?

EDIT: 26.08.2013 - changed (corrected) the given hashvalue of the xml file

Answer 1

Ok i want to inform about the solution, maybe any one will find these topic:
The document node has to selected directly in that way from the container: "XmlNodeList documentNodeList = sepaContainer.GetElementsByTagName("Document");"

then the outer xml of the first entry have to be canonized and convert to hash.

also the first posted value: 7DDB2138E8C91037DA7A6E9552478C59A2AACE26F8112EBC8012A8DE11592025 is wrong the correct value will be 38B862BD35B95D8CB20448153B5F6F73B3657EB1BC81CF11D3491D16EEA94009