XDBF1.EXE is a program similar to XBTREE1.BAS. It operates on the same raw
data file, XDATA1.DAT, and produces similar output. The main difference is
the speed and index file size, and dBASE data file compatibility. XDBF1.EXE
uses a high-performance reindexing algorithm that is included with the QBXDBF
package.

If you want QBXDBF and dBASE compatibility then you need to order QBTree
with the QBXDBF option. See the !ORDER.FRM for ordering information.

XDBF1.EXE was compiled with BC 7.1 and linked with QBXDBFAN.LIB. The file
is not compressed in any manner--89453 is the total size (all the routines
of QBXDBF were linked into the EXE file). PKLITE or another compression
program such as LZEXE would bring the size down to around 49K.

XDBF1.EXE will create two files, CATALOG.DBF and CATALOG.DBX. Make sure
that you are not using those filenames in the current directory because
they will be erased.

To run the program make sure XDATA1.DAT is in the current directory and

        C>xdbf1

The program, like XBTREE1.BAS, processes the data file, XDATA1.DAT. This
data file is an edited portion (2048 records) of a 100,000 record data file
of marine parts. It reads the data file one record at a time and writes that
record to the CATALOG.DBF file. When all records from XDATA1.DAT have been
read and written, the index file CATALOG.DBX is built. Once built, all keys
in it are accessed, in PARTNO index order. This shows the raw access speed of
an index access. The program then displays all records in PARTNO index order
to the screen.

As in most instances, the imported file is already sorted. However, their
is no requirement that it be that way--QBXDBF will operate just as well
on unsorted data records.

Options:

For incremental indexing (like QBTree), use /i.

        C>xdbf1 /i

To test delete/pack/reindex where every other record is deleted, then the
data file 'packed', then reindexed, use /dp.

        C>xdbf1 /dp

<EOF>