Explain Processing A Data File?

Mainly data file applications needs that a data file be altered as it is being processed. For instance in an application involving the processing of customer records it perhaps desirable to add new records to the file either at the end of the file or interspersed among the existing records to delete existing records that to modify the contents of existing records or to rearrange the records and these requirements in turn suggest numerous different computational strategies.

Consider for instance the problem of updating the records within a data file. There are numerous approaches to this problem. Maybe the most obvious approach is to read each record from a data file update the record as necessary and then write the updated record to the same data file. But there are some troubles with this strategy. In particular it is hard to write and read formatted data to the same data file with no disrupting the arrangement of the data items within the file. Furthermore the original set of records may become unapproachable if something goes wrong during the program execution. Another approach is to work with two diverse data files - an old file (a source) and a new file. Every record is read from the old file the updated as necessary and then written to the new file. When each and every one of the records have been updated the old file is deleted or placed into archival storage and the new file renamed. Therefore the new file will become the source for the next round of updates.