Reading Data from a multi sequenced File

[MumbleB]

Hi Guys. I have a file with two sequences, Sequence A and a Sequence B. Sequence A only appears once in the file and Sequence B can appear once or multiple times in the file. See Sample data below: :21:REFERENCE1 :23E:OTHR/DMST :32B:ZAR5005,00 :57A:CODEXXXX :59:/12345600002345678 TEST NAME2 ADDRESS1 ADDRESS2 ADDRESS3 :21:REFERENCE2 :23E:OTHR/DMST :32B:ZAR7005,00 :57A:CODEAAAA :59:/12345657002345678 TEST NAME3 ADDRESS1 ADDRESS2 ADDRESS3 :71A:OUR -} The portion in Bold is Sequence A and the remainder is sequence B data. As you can see by the "TAGS" the fields starting with ':', that there are multiple Sequence B data. I need a way to pick up the data for the multiple Sequence B records. Sequence B start with :21: and ends with :59: followed by the address details. I tried a straight forward StreamReader and then populating the data but my Steamreader always goes from top to bottom in my while loop. Sample below.

        string inFile;
        inFile = textBox1.Text;
        StreamReader sr = new StreamReader(inFile);
        string holdline;
        int lineCount = 0;
        int clientname = 0;
        int creditname = 0;
        while (!sr.EndOfStream)
        {
            holdline = sr.ReadLine();

            if (holdline.Length > 4)
            {
                string tag = holdline.Substring(0, 4);
                string tag2 = holdline.Substring(0, 5);
                if (tag == "{1:F")
                {
                    lblBIC1.Text = holdline.Substring(6, 8).Trim();
                    lblClientBIC.Text = holdline.Substring(46, 8).Trim();
                    lblMsgType.Text = "MT" + holdline.Substring(33, 3).Trim();
                    lblDate.Text = holdline.Substring(36, 10).Trim();
                    lblSWIFTRef.Text = holdline.Substring(58, 10).Trim();
                }

                if (tag == ":20:")
                {
                    int results = holdline.Length - 4;
                    lblA20.Text = holdline.Substring(4, results).Trim();
                }

                else if (tag2 == ":21R:")
                {
                    int r1 = holdline.Length - 5;
                    lblA20.Tex

[OriginalGriff]

Try adding "+" characters:

                    lblA20.Text += holdline.Substring(4, results).Trim();
                    ...
                    lblA20.Text += holdline.Substring(4, r1).Trim();
                    ...
                    lblA50H.Text += holdline.Substring(6, r3).Trim();

Would be better if you used StringBuilder, but hey ho!

Real men don't use instructions. They are only the manufacturers opinion on how to put the thing together.

[MumbleB]

Hi Griff. Not sure if that will work for what I have planned. What I am currently doing is writing the data to labels. Sequence B can be made up of multiple records/lines, each containing a set of TAGS. Tags are anything that start and end with ":". From the sample I supplied, one sequence B records starts with TAG :21: and ends with TAG :70:

Excellence is doing ordinary things extraordinarily well.

[PIEBALDconsult]

If it's not too big, you could read the whole thing in and use a RegularExpression to parse it. I would likely read it one character at a time rather than as lines. You can then write a little Finite State Machine. You are also trying to do two things in one method here; I recommend splitting the method in two -- one to read/parse, one to display. The reader/parse could return a List of Tuples, each Tuple wouyld contain the type/number of the field and the data.

[jschell]

Kwagga wrote:

Any advice would be greatly apreciated.

Write a parser which reads the file and creates a data structure (presuming that the file is reasonably sized.) Your data struct is represented by the following pseudo code

  class RecordA
  {
    // Fields from record
    ...
  }
  class RecordB
  {
    // Fields from record
    ...
  }

  class Data
  {
    RecordA RecordA
    List RecordBList
  }

After you have populated Data above then you pass it to something else.