MF Cobol File Parsing - Not reading all of the records when switching from Trial to Licensed

Posted January 20th, 2009 by hhichoyer

We have evaluated the MF Cobol Parser software and it worked well for our evaluation purposes and we are now trying to use the same parser to process our production files...

With the trial verision we got a seemingly random number of records and when we purchased the real license, the same behavior is present - specifically, I have a 80K+ record file that I get 3500 records from in either trial or licensed mode!!!

Is there specific swtiches or some procedure that we need to do to configure the software to process everything in the file?

Thanks

Carl Hoyer

‹ cannot save map Microfocus ISAM files ›

RE: MF Cobol File Parsing...

On January 20th, 2009 jterrell said:

Hi Carl,

The trial license limits the output from the Data Parsers to 100 records, while a purchased license should export all of the records in your data file. There are no procedures required to switch from the trial license to the purchased license, except to be sure that the product you are launching is the purchased license in the DataTools Launcher.

A possible cause of the problem you described is that the Micro Focus COBOL data file and/or index file is "corrupt". And the best way to know this is to verify that the application from which the files originated is operating correctly with the data and index files. If the application is no longer operable and/or does not operate properly with the files, then there is a possibility that one or both of the files have been corrupted. If this is the case, the exported data file may not contain all of the records.

- Jo

Okay...using the binary parser works except....

On January 21st, 2009 hhichoyer said:

When I am using the binary parser to process our MF data files for the previous issue it worked and gave me all of the records...we have 36 files to convert here....so when I try another, it gives me the following:

Possible loss of precision from data type mapping in RunButtonClicked(dparser.bas)(25020)

Checked the source data browser and all of the data is available with the combination of the data file and cpy file...so when I try to export, the above error message occurs immediately and generates no other log messages - also no empty target file either!!!

This seems to be an issue with trying to export to an ASCII file?

I have no other options here...if the source data browser is happy, then what is the deal?

Thanks

Carl

RE: Okay...using the binary parser works except....

On January 21st, 2009 jterrell said:

Hi Carl,

The error seems to indicate that one or more of the numeric fields in the source file may be slightly different; i.e., the number of decimals, or something similar. Make sure all the data looks perfect in the source data browser. If it doesn't, you may need to make a change to one or more of the data fields' properties before exporting the data.

- Jo

Yes all of the data looks just fine...So now what?

On January 22nd, 2009 hhichoyer said:

So when we set the data browser to display all the records - in this case around 24K - they look just fine...

So why can I not get an export and why is there nothing in the log file telling me some form of error?

Also, I am using the parser product here....is there something else that will allow me to transfer sequential flat file data that has copy book descriptors (cpy file) to an intermediate format - csv or something else - that will be loaded into a sql database?

Thanks

Carl

RE: Yes all of the data looks just fine...So now what?

On January 23rd, 2009 jterrell said:

Hi hhichoyer,

Couple of questions:

1. When you open the source data browser and view the unpacked data records, each "page" of the browser displays only 500 records. But you can scroll through the pages by clicking on the left and right arrows in the lower right portion of the browser window. If you scroll "right", how many total data records can you view?

2. How many data records are exporting to the CSV data file?

- Jo

Data exporting problems continued...

On January 26th, 2009 hhichoyer said:

Actually if you ajuast the range from 500 to 50000, we get only one "page" of data that contains 29,470 records that are what we expect...

We do not get a CSV file at all...there is no file nor any records exported...

Thanks

Carl Hoyer

RE: Data exporting problems continued...

On January 26th, 2009 jterrell said:

Hi hhichoyer,

After verifying that the data looks correct in the data browser, are you clicking on the "Run Map" button (looks like the Run button on a CD player), entering a target file path and filename, and actually exporting the data to an output file?

- Jo

Still having data export problems...

On January 27th, 2009 hhichoyer said:

Yes it is that step that generates the original error message that we have asked about...

Carl

RE: Still having data export problems...

On January 27th, 2009 jterrell said:

Since the trial license allows the export of only the first 100 records, and if those 100 records export correctly -- without the error, then it seems probable that somewhere after the 100th record there is an unexpected character that is preventing the Data Parser from handling the remainder of the file.

Although it is a tedious task, you may want to examine the data in the source browser very closely and try to find the record that may be corrupted. If you can find it, you can then use the record filtering feature in the export dialog box to export records 1 through "n" -- where "n" is the record just BEFORE the corrupted record. Then perform a second export using the record filtering feature to export records "n+2" through the last record -- where "n+2" is the record just AFTER the corrupted record.

Please let us know if this helps resolve the problem for you, or not.

- Jo

Yes we try to export but get the error without any output....

On January 27th, 2009 hhichoyer (not verified) said:

and for other files this seems to be working okay...