Parse file that does not have closing tags

[deepak-agendaboa]

My purpose is to read the .OFX file which is also a file with attributes.
Some OFX files don't have the closing tags. So they cannot be parsed.
is there a way to parse the file that is missing closing tags?

This is the content that I'm trying to parse

<OFX>
  <SIGNONMSGSRSV1>
    <SONRS>
      <STATUS>
        <CODE>0
        <SEVERITY>INFO
      </STATUS>
      <DTSERVER>20071015021529.000[-8:PST]
      <LANGUAGE>ENG
      <DTACCTUP>19900101000000
      <FI>
        <ORG>MYBANK
        <FID>01234
      </FI>
    </SONRS>
  </SIGNONMSGSRSV1>
  <BANKMSGSRSV1>
      <STMTTRNRS>
        <TRNUID>23382938
        <STATUS>
          <CODE>0
          <SEVERITY>INFO
        </STATUS>
        <STMTRS>
          <CURDEF>USD
          <BANKACCTFROM>
            <BANKID>987654321
            <ACCTID>098-121
            <ACCTTYPE>SAVINGS
          </BANKACCTFROM>
          <BANKTRANLIST>
            <DTSTART>20070101
            <DTEND>20071015
            <STMTTRN>
              <TRNTYPE>CREDIT
              <DTPOSTED>20070315
              <DTUSER>20070315
              <TRNAMT>200.00
              <FITID>980315001
              <NAME>DEPOSIT
              <MEMO>automatic deposit
            </STMTTRN>
            <STMTTRN>
              <TRNTYPE>CREDIT
              <DTPOSTED>20070329
              <DTUSER>20070329
              <TRNAMT>150.00
              <FITID>980310001
              <NAME>TRANSFER
              <MEMO>Transfer from checking
            </STMTTRN>
            <STMTTRN>
              <TRNTYPE>PAYMENT
              <DTPOSTED>20070709
              <DTUSER>20070709
              <TRNAMT>-100.00
              <FITID>980309001
                <CHECKNUM>1025
              <NAME>John Hancock
            </STMTTRN>
          </BANKTRANLIST>
          <LEDGERBAL>
            <BALAMT>5250.00
            <DTASOF>20071015021529.000[-8:PST]
          </LEDGERBAL>
          <AVAILBAL>
            <BALAMT>5250.00
            <DTASOF>20071015021529.000[-8:PST]
          </AVAILBAL>
        </STMTRS>
      </STMTTRNRS>
  </BANKMSGSRSV1>
</OFX>

Attaching the OFX files as zip folder.
ofx-files.zip

[renggli]

The input you parse is not well-formed XML, which is not supported by this library. Still, you could possibly use the event-driven parser and insert close-events as needed before converting to a DOM tree?

[deepak-agendaboa]

I have tried the following to add the missing closing tags

                      ofxString = ofxString
                      // remove whitespace in between tag close/open
                      .replaceAll(RegExp(r">\s+<"), '><')
                      // remove whitespace before a close tag
                      .replaceAll(RegExp(r"\s+<"), '<')
                      // remove whitespace after a close tag
                      .replaceAll(RegExp(r">\s+"), '>');

                      // parse the string as xml events to add the missing closing nodes
                      // TODO: this can be improved
                      bool isLastTextNode = false;
                      StringBuffer ofxBuffer = StringBuffer();
                      parseEvents(ofxString, withParent: true).map((event) {
                        return {
                          "event": event.toString(),
                          "type": event.nodeType,
                          "parent": event.parent?.name
                        };
                      }).forEach(
                        (element) {
                          final event = element['event'];
                          final type = element['type'];
                          final parent = element['parent'];

                          if (type == XmlNodeType.ELEMENT) {
                            // check if the last node was text or not
                            if (isLastTextNode) {
                              final parentElement = "</$parent>";
                              if (!event.toString().startsWith(parentElement)) {
                                // the last node was text but current is not the closing
                                // tag of the previous node
                                ofxBuffer.write(parentElement);
                              }
                            }
                          }
                          ofxBuffer.write(event);
                          isLastTextNode = type == XmlNodeType.TEXT;
                        },
                      );

                      debugPrint(ofxBuffer.toString());