Method for transporting relational data

摘要

A method for transporting relational data from one relational database to another such database. The data transported may be a subset of data in the entire database and the relations between the root of this data and other data entities are preserved in a way allowing their reconstruction in the target database. This preservation, itself further directed through configuration, is able to recreate new data identifiers without collision, yet also without requiring maintenance of the original identifiers.

主权项

1. A computer-implemented method comprising the steps of:
accessing a data table in a source relational database management system (RDBMS); selecting a row of the table, the row including columnar data; and encoding the row for export; wherein encoding the row includes—
determining dependent data referenced in the row;generating portable row identifier substitutes; andcollecting the row data and the dependent data referenced in the row to form export data, except for dependent data stored in a table designated for exclusion; arranging the export data, utilizing the portable row identifier substitutes in lieu of the original identifiers, into a reconstructable sequence; and then importing the export data into a target database according to the reconstructable sequence; wherein encoding the row for export includes:
generating a unique secure listing of the portable row identifier substitutes corresponding to the original identifiers in the data for export;receiving parameters from a source client including an initial row identifier and table to export;encoding the initial row for export; repeating exporting additional rows until no more rows require exporting—bottomed out what is referenced and referencing; arranging the exported rows into a list in the reconstructable sequence for progressive reconstruction; and converting the list of rows into an external storage format; wherein each row encoding step includes: skipping export of the row in the case that export parameters indicate ignoring the corresponding table; creating an encoded row identifier which is unique and traceable to the table and the original row identifier; in the case that the encoded row identifier is used, skipping export of the row; fetching the columnar data of the exporting row from the source database; processing the columnar data into an internal representation; determining relations to and from the exporting row; examining each row referenced by the exporting row, and for each referenced row— IF—a referenced row is from an ignored table, storing a dummy value in the internal representation for later detection during import; ELSE—exporting the row and storing the corresponding encoded row identifier into the internal representation; examining each row that references the exporting row, and exporting each such referencing row.