Information for the product and data column tables will be extracted automatically from the PDS label without modification wherever possible. Ingest routines will scan the label to pull out the desired information and format it into tables and command files which can be loaded directly into the database using the SQL*Loader utility (from Oracle).
The Object references will be extracted by using the information in the label to locate the identification field(s) and then reading the data file to find the corresponding record numbers. As above, output will be in the form of a command file and data list to be loaded into the database.
When updates are needed (following reviews) existing data will be deleted and the new data inserted using these automatically generated command files.
Note
The processing routines will need to be able to make some assumptions about formats and field definition, especially with respect to object identifications. ID fields will have to use standard local conventions for NAME and content. This will require cosmetic editing of some of the current archive files.
We will also need to ensure that all our archive labels contain the information needed to fill these tables, even when not required by PDS. We should bear this in mind during reviews, especially in looking at DESCRIPTION fields. DESCRIPTIONs should always be: a) present; b) succinct; and c) a fairly thorough description of the field contents.