TurboIMAGE/XL Database Management System Reference Manual MPE/iX V6.5 (30391-90011)

Chapter 2 43
Database Structure and Protection
Data Set Types and Relationships
follows all existing entries in the chain. If the path has a sort item, the entry is inserted in
the chain according to the value of that item.
If the entry's sort item value matches the sort item values of other entries in the chain, the
position of the entry is determined by an extended sort field consisting of the sort item
value and the values of all items following the sort item in the entry. If the extended sort
field matches another extended sort field, the entry is inserted chronologically following
the other entries with the same extended sort field value. This also occurs if the sort item
is the last item in the entry and its value matches another entry's sort item value. Note
that Native Language Support does not support extended sort items. The only database
language that supports extended sort fields is Native-3000 which uses US ASCII. If an
extended sort field is used, the sort is done in ASCII collating sequence (negative integers
sort higher than positive).
If you depend on extended sort fields to sort a chain, do not call DBUPDATE to modify any of
the values in the extended sort fields because the chain will not be automatically resorted
according to the new extended sort data values. Instead, call DBDELETE and DBPUT to
re-enter the records with modified values. DBUPDATE only recognizes extended sort items
when the actual search or sort item is changed. Chapter 5 describes DBUPDATE, DBDELETE,
and DBPUT in detail.
If you do not want TurboIMAGE/XL to sort chains by extended sort fields, structure the
data record so that the sort item is in the last field of the record.
When the database content is copied to magnetic tape using the TurboIMAGE/XL utility
program DBUNLOAD, the pointers that define an entry's position in a chain are not copied to
the tape. When the data is loaded back into the database, the chains are re-created.
Therefore, entries that were previously ordered chronologically will not necessarily be in
that same order. The new chronological ordering is based on the order the entries are read
from the tape. The chains of a primary path are an exception; the order of these chains is
preserved if the tape was created with DBUNLOAD in the chained mode. See chapter 8 for
more information about DBUNLOAD.
NOTE It is important to limit the use of sorted chains to paths consisting of
relatively short chains or chronological sort items that are usually added to
the end of chain (for example, date). Sorted paths should not be used for
multiple key sorts or for sorting entire data sets. These functions are handled
more efficiently by user-written routines or the MPE/iX HP Sort subsystem.
Jumbo Data Sets
You can create data sets greater than 4 GB in size. A data set of this type, called a jumbo
data set, can span more than one MPE file. The naming convention uses POSIX
extensions. For example, a dataset named SALES03 with four multiple files results in a
total of five files with the following names and filecodes:
SALES03 filecode -408
SALES03.001 filecode -409
SALES03.002 filecode -409
SALES03.003 filecode -409
SALES03.004 filecode -409