Data Conversion - Information Systems Reengineering, Integration and Normalization - page 149

from the bottom up and map each record type into a sequential file. The Unload
algorithm is as follows:

Program Unload network database to sequentialfiles
begin
/* n=number of record types

m=number of levels in each path expression

*/
Getall record type N 1 , N 2 ….N n within inputnetwork schema;
Fori=1tondo/*for each target record type N i */
whileN i record occurrencefounddo
begin
If it is firstoccurrence
then obtain firstrecord N i within area
else obtain next record N i within area;
Forj=m-1 to 1do
/* read target record ownerrecords by databasenavigation
from levelm-1 to level1,asystem-owned records*/
Obtain ownerrecords keys K i (1), K i (2),...K i (j)
/* obtainthe record keys of allownersofrecordN i along
databaseaccess pathfrombottomuptothe system owned
record*/
end-for;
Case record identifier_typeof
'F': begin
Ifm=1
then output N i record with K i (m)as recordidentifierto
sequential file i
else output N i record with K i (1), K i (2)….K i (m-1), K i (m)
as foreign keytosequentialfilei/* K i (m)=keyof

ownerrecordkey in levelm*/

end;
'P': output N i record with K i (1), K i (2),..,K i (m-1), K i (m)as
record identifier to sequential file i;
'I': output N i record with K i (1), K i (2),..,K i (m-1),
Sequence# as record identifier to sequentialfilei;
end-case;
end-while;
end-for;
end;