Beijing Paradise BBS Backup/PARADISE.ISO/software/BBSDOORW/DDOV10.ZIP/DDOV.DOC

Text File | 1994-01-09 | 7.6 KB | 191 lines

DDOV version 1.0 - An RBBS Utility
by Ken Goosens.  Jan 8, 1994

Contents
    0.0  Licensing
    1.0  What is DDOV?
    2.0  What's New?
    3.0  The Theory Behind DDOV
    4.0  How to Set up DDOV
    5.0  Recompiling


                      0.0  Licensing

The author does not ask for any donation for use of DDOV.
The rights to the code are exclusively retained by the author.

Everyone is free to use DDOV without any cost.   Everyone is
free to distribute DDOV without cost, provided they charge nothing
for DDOV.  You cannot distribute DDOV with any shareware or
commercial product without permission of the author.   Modified
versions of DDOV can be distributed only with permission of the
author.  Enhancements or fixes to the code are welcome.   Anyone
is free to write (and charge for) an improvement over DDOV in
a language other than BASIC.  For support, the author can be reached
via the RBBS conference on RBBS-Net or on his Bulletin Board at
703-978-6360.


                1.0  What is DDOV?

DDOV is a utility to detect duplicate files and old versions of files
and give you the opportunity to delete them.   It is meant to be helpful
in making room on a hard disk for Bulletin Board Systems.   It uses
the output file FIDX.DEF in RBBS-PC, but this file can easily be created
on any BBS (using MAKEFIDX.EXE distributed with RBBS-PC).  DDOV will
perform the following kinds of functions:

o   detect duplicate files, based on the fact that they have the
    same name and size.   DDOV does not look in any way at the
    contents of a file and so will NOT detect duplicates with different
    sames, say two GIF files of the same picture.

o   detect different versions of the same file.  This inference is
    necessarily uncertain.  DDOV looks for a common stem in the front
    of the name, followed by a numeric version.

o   entire drives can be declared off limits of file deletion.  You
    will want to use this if you have CD-Roms.  DDOV will recover when
    the file cannot be omitted but this declaration will save you
    considerable time and processing.   DDOV will not even report
    dups and old versions of both are on drives that where files cannot
    be omitted.

DDOV is especially useful when a BBS expands by adding masses of files,
say from a CD-Rom or a tape of software.   The SysOp then faces the
problem of culling out duplicates and older versions.

DDOV should be distributed as DDOV10.ZIP and consists
of the following files:

DDOV.EXE    executable program
DDOV.BAS    source code
DDOV.CFG    sample configuration set up for runs
DDOV.IN     list of files to include in DDOV.LIB
DDOV.INF    information file for modules for DDOV.LIB
DDOV.DOC    documentation for DDOV
LDDOV.BAT   batch file for creating DDOV.LIB needed to recompile

                   2.0  What's New

Nothing.  This is the first release.

                   3.0  The Theory Behind DDOV

The theory on duplicate files is simple:  they have the same
file name and the same file size.   The date on the file is a
less reliable indicator and is not used.  If the files have the
same name but a different size, the version routine will flag
them as possible different versions.  A better way to detect
duplicates is by calculating a CRC for each file, but this takes
a great deal of time.

DDOV does its real work is detecting different versions of
files.   In general, versions have

o    A front part of the file name prefix that is the same

o    either the same extension, or are one of an equivalent
     class of extensions, the equivalents being ZIP, ARC, ARJ,
     PAK, and LZH in the IBM PC world.

These criteria are not certain.  A file can be named so that
the front part varies between versions and the end of the prefix
stays constant.   Files with similar names can be unrelated.

But you have to be able to distinguish versions from a series of
different files.  A series of files usually have either

o    a particular extension (GIF files are a good example), or

o    the same file date.

None of these criteria are certain.  When a series of different files
are put out over time, the version detector will wrongly treat them as
later versions.

                   4.0  How to Set up DDOV

DDOV requires a file and location index to generated prior to running
DDOV.  Many RBBS-PC SysOps will already be using the fast file search
which is based on such files.  However, since you are looking for
duplicates and old versions, you will probably want to include your
upload directory in the list of files, so you probably will want to
make a special run of MAKEFIDX for DDOV that includes it.   Be sure to
sort this file first by file name.

DDOV operates differently from most PC programs in the program
does not offer the user a simple menu of choices.  DDOV is designed
instead to get its setup from a file, do its work to detect duplicates
and old versions, and give the operator the opportunity to delete one
or both of the files.

Use your favorite full screen editor to set up a configuration for
your application.  You can name your configuration anything you want
and invoke DDOV using configuration XYZ.CFG by typing

                     DDOV XYZ.CFG

A general configuration file is provided as
DDOV.CFG.

                      5.0  How DDOV Operates

DDOV will read in your location index file, then process the file list,
looking for duplicates and old versions.  At the top of the screen a
running list of the current file being read and the number processed is
kept.   DDOV will never automatically delete any files.  When it
encounters a potential duplicate or old version, it will clear the screen,
and print the file name, location, size, and date of each on the screen.
It will note if the location is protected from deletion by "no kill ->",
which file is bigger by "larger ->", and which is newer by "newer ->".
These are meant to help you in analyzing which files to delete.   If,
however, the smaller and older file is on a drive that cannot be deleted,
DDOV will not even ask you whether you want to kill any but simply
continue processing.  When DDOV pauses for instructions about what files
to omit, your choices are

A  - omit file labelled A - which is the top entry.

B  - omit file labelled B - which is the 2nd and bottom entry

AB - omit both files displayed

Q  - cease processing. Terminate the program

J  - jump to an entry in the file list.  Very fast way to leap
     over files to start processing at a later point.

If you ask DDOV to delete a drive marked as not supporting deletion,
DDOV will simply wipe away your answer and reprompt for another answer.

DDOV will append the location/name of all files omitted to a log file,
whose default name is KILLED.FIL.   DDOV will never delete this file, so
if you want to start clean, you have to omit it.   This log file is useful
for keeping track of what files you may want to remove from your
DIR listings, but DDOV will not remove any file entries from your RBBS
directory listings - only from the physical DOS drives.

                      5.0  Recompiling

DDOV is written in Basic for MicroSoft's Professional Basic 7.1,
and uses Tom Hanlin's library PBCLONE.

First get a copy of PBCLON...ZIP.  This program is shareware so be
sure to register it.   You must extract the obj files needed.
The Hanlin shareware product LibWiz has a utility called Unlib
that makes this job easy.

Then, to make the library, just say

LDDOV

to invoke the batch file LDDOV.BAT enclosed.  This uses DDOV.IN.

For Professional Basic 7.1, the commands to recompile are:

bc DDOV,DDOV /o/x;
link DDOV,,,DOV;