Consultor Eletrônico

Status: Verified

GOAL:

What to do when the records were loaded into a Database having chosen the wrong code page during the import?

GOAL:

What are the corrective steps to take when the wrong code page have been chose during the load of the records?

CAUSE:

The import of the records was done without specifying the correct code page and outside of the Data Dictionary.
The Database has been loaded with records that were dumped with a different code page outside of the Data Administration tool and the wrong codepage or no code page was specified with the -cpstream during the import of the records outside of Data Administration.
In example the records were IBM850 and the Data Base is ISO8859-1.
This will create a logical corruption of the records because some characters are not visible or not the correct ones for the specified code page.

FIX:

There are two possible situations for this problem and it might take a different approach depending on the case.
1. No new records have been created after the import.
In this case it is best to redo the import of the data into a new Data Base specifying the correct -cpstream codepage in way to correct conversion occurs during the load or import of records outside of the Data Administration.
If the data which is wanted to be loaded was IBM850, then specify for the load -cpstream IBM850 this will ensure the conversion of the characters to your actual code page.
2. Already created new records after the load in the Database new code page.
In this case part of the records showing the correct characters and part of the records showing incorrect characters.
The only corrective measure that can be taken is to dump the database with the Data Administration tool and using an editor do a search and replace of the incorrect characters in the .d files. This could be long and tedious depending on the amount of data to correct.
Once the files have been corrected reload them using once again the Data Administration tool.