home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
ftp.barnyard.co.uk
/
2015.02.ftp.barnyard.co.uk.tar
/
ftp.barnyard.co.uk
/
cpm
/
walnut-creek-CDROM
/
CPM
/
MODEMS
/
MODEM
/
RAIDSYS1.LBR
/
RAIDSYS.DZC
/
RAIDSYS.DOC
Wrap
Text File
|
2000-06-30
|
12KB
|
272 lines
RAIDSYS v1.00 - 1 April 88
The Theory and Practice of Raiding:
Everyone knows that batch transfers are much more efficient than the
usual file-here file-there methods we all use. After watching users
poke around on my system for the last three years and from my own
frustrating experiences in determining what other systems have that
I don't and what I have that they don't I finally decided to do
something about it. RAIDSYS is a method of acquisition which has
proved very effective for me in the last few months. Happily the
process can be reversed - you can then easily load up a disk or
two and send the raided systems all the software you have that they
don't.
You will need several other programs not included in this library
to use RAIDSYS. They are: LSTOOL and PASTE - read no further until
you get these libraries. You will also need a MAST.LST of your
files that uses the same format as Irv Hoff's MCAT/XCAT program, i.e.
the master list looks like:
MYFILE .LBR - 001
HISFILE .ARK - 002
THERFILE.AQM - 003
Background:
I started with a simple idea; find a program that will output the
differences between your MAST.LST and the target systems on-line
list which the DOTEMP program included in this library will produce
for you. The first program I used in this manner was CCAT26 which
was designed to output the difference between two MAST.LST files.
CCAT had several deficiencies not the least of which was that the
output file is in a multi-column format. Probably the biggest
problem with this format is that it lacks any information about
where on the target system the files are - so you still waste a
lot of time finding out where the files are, either by manually
writing in the d/u on a hard-copy of CCAT's output or by doing a
lot of FIND (or FILE) searches on the system. Plainly this would
not do - some more efficient method had to be found for this system
to work.
To this end I wrote a complete applications program in dBASE which
though awkward and slow does work. However a lot of people never
got dBASE for their CP/M machines and a lot who did never use it.
I still wanted something simple and FAST. Then I discovered yet
another program designed to compare two MAST.LST files. This is
LSTOOL written by Jim Gronek and widely available on RCP/Ms in
the country. As you will see LSTOOL fits the bill completely for
our raiding purposes.
Foreground:
The aim of RAIDSYS is to produce a "want list" from your
favorite system(s) that you can then use to batch download and
upload files with. So what makes an ideal "want list" you say?
The most effective list of this kind is of two parts as below.
ARUNZ09 .LBR D01 16k COMP28 .LBR A01 8k
ASM-SUBS.LBR C01 8k HSH13 .LBR A01 28k
BALIAS13.LBR D01 12k HSHFIX .LBR A01 20k
CLNDIR11.LBR D01 12k LDSK11 .LBR B03 12k
COMP28 .LBR A01 8k IF14 .LBR C00 16k
HSH13 .LBR A01 28k ASM-SUBS.LBR C01 8k
HSHFIX .LBR A01 20k SILENCE .LBR C01 40k
IF14 .LBR C00 16k IOPSQ .LBR C03 68k
IOPSQ .LBR C03 68k SYSFCP14.LBR C04 20k
LDR15 .LBR F05 16k ARUNZ09 .LBR D01 16k
LDSK11 .LBR B03 12k BALIAS13.LBR D01 12k
RCP/FCP .LBR F02 52k CLNDIR11.LBR D01 12k
SAK21 .LBR F01 8k SAK21 .LBR F01 8k
SILENCE .LBR C01 40k RCP/FCP .LBR F02 52k
SYSFCP14.LBR C04 20k LDR15 .LBR F05 16k
Z3INS13 .LBR H09 12k Z3LIB12 .LBR G02 96k
Z3KEY12 .LBR G03 36k Z3KEY12 .LBR G03 36k
Z3LIB12 .LBR G02 96k Z3INS13 .LBR H09 12k
Armed with lists like this for several systems you can really "make
a night of it" as far as downloads go. All you have to do once into
the operating system is systematically go from du to du and enter:
A01>KMD SB COMP28.LBR HSH13.LBR HSHFIX.LBR
If the system above used ZMD you could "wipe out" B3: by making the
command line appear as such:
A01>ZMD SB COMP28.LBR HSH13.LBR HSHFIX.LBR B3:LDSK11.LBR
Preparation for Raiding:
Get the DIRectories of the system(s) you are going to raid. They
must use SD (usually named DIR.COM) or a derivative thereof that
displays files as below:
Drive A2 Files: 5/304k Free: 1544k
EXCHLST .LBR 28k : MEX114 .LBR 136k : PCPIMP6 .LBR 32k :
IMP245 .LBR 64k : PCPHELP2.LBR 44k :
Note that the word "Drive" must start in column 10 and the
drive/user must appear in columns 16 thru 18 or DOTEMP will
output meaningless data for you. Some versions of SD have
a ":" after "Drive" - DOTEMP can handle these. Some CP/M+
systems use a form of DIR which displays like this:
EXCHLST .LBR 28k : MEX114 .LBR 136k : PCPIMP6 .LBR 32k :
IMP245 .LBR 64k : PCPHELP2.LBR 44k :
Drive A2 Files: 5/304k Free: 1544k
You must edit these and place the d/u information above the
files or DOTEMP will fail. If you get line hits during the
ASCII capture then these also must be cleaned up with your
editor. PCP users note that without XON/XOFF you will get
garbaged lines when your comm program stops input and writes
to disk. Normally this will effect only a line or two and if
the line isn't easily editable just delete it - you will still
be getting a lot of files.
Next you must run a copy of your MAST.LST through DOTEMP. The
output will look as below and will be named SYSDIR.TXT. Rename
this to MAST.LST as soon as it is output. The lack of disk
numbers will distinguish it from your normal MAST.LST.
MYFILE .LBR -
HISFILE .LBR -
THERFILE.LBR -
NOTE: There are three versions of DOTEMP in the library. They
allow you to selectively determine which kind of files you
want to acquire. DOTEMPA will search for files having an ARK
or LBR extension. DOTEMPL will search for files with the LBR
extension only, and DOTEMPC will not filter ANY extensions.
DOTEMPC will produce very large output files from some systems
as well as when it runs against your MAST.LST. which will
waste a lot of paper and time for you. If you really want to
do it this way be my guest.
Now run DOTEMP against each system directory you have acquired.
DOTEMP will operate from the command line, or failing that, it
will display all files on disk having a "DIR" extension. As it
runs it will display the current drive/user of the system being
scanned and will output a SYSDIR.TXT file that looks like this:
SILENCE .LBR C01 40k
SYSFCP14.LBR C04 20k
SAK21 .LBR F01 8k
Z3KEY12 .LBR G03 36k
Z3LIB12 .LBR G02 96k
Z3INS13 .LBR H09 12k
Rename these output lists to something that will remind you
what system you got them from such as IRV.DIR or AB17.DIR and
do no more than five or so at a time. I find that beyond five
it gets pretty confusing.
In order for LSTOOL to work properly with these lists they must
be sorted on the filename.ext. I use SUPERSORT in a batch mode
but there are a variety of free sort programs that will do the same
thing for you. Once you have them sorted you are ready for
LSTOOL to do it's work.
Using LSTOOL:
The program must be installed for your system; the documentation is
quite adequate but you don't need to read it if you are in a hurry.
Get into the program, select option 2 (compare two lists) and enter
the following:
MAST.LST and SYS.LST
where MAST.LST is the stripped copy of your MAST.LST file and SYS.LST
is the output of DOTEMP against the systems DIRectory after sorting.
LSTOOL will output a file called LIST.DIF. Using your sort program
and sort this on the drive/user columns to a second file named, say,
LIST2.DIF. Now run PASTE.COM against these two to produce your final
"want list" and you are in business. Example:
PASTE NORMSYS.DIR LIST.DIF LIST2.DIF
Now you can use LISTT16 or your editor to print out the file with a
header identifying the date and other information such as the phone
number of the system and you are ready to go.
Paybacks:
It is patently unfair to systematically raid systems in this manner
without paying them back. Occasionally you should reverse the
process and run the program to find out what you have that the
target system doesn't. Get the sysop's permission to upload a
disk or two of files and return the favor. One obstruction to
paybacks is the ubiquitous program FOR. FOR requires a description
of each file uploaded to the system and can require a lot of tedious
typing to satisfy the system. If the sysop permits it you could
make an agreement to upload to the private area or if he using ZMD
you might be allowed to upload without descriptions. You could
also write up a series of FOR files and send them ASCII as the
KMD/ZMD programs ask for each description - this is still very time
consuming. Personally I find FOR and LUX to be redundant - if you
really want to find out what the program does use LUX. When I used
FOR on my own system I found that two-thirds of the FOR descriptions
were "Look what a great programmer I am" types and the other third
were typo-ridden monstrosities. Additionally, the writing back to
disk of the FOR file can take precious minutes of your on-line
dollar.
Other favors:
Take a look at the sorted-on-filenames list and see if the system
has any duplicate files on-line. Multi-Megabyte systems have a
way of getting out of hand and often several copies of the same
program will be in several different areas. Many sysops would
appreciate your reporting these redundant files to them.
Technical Tips:
You can easily get confused as to what order all this has to
fall together and I highly recommend constructing some sort
of batch file to handle it for you. I use ZCPR3 and an ALIAS
named RAID which looks like this:
DOTEMP $1.DIR ; You do have a DIR ext?
SSORT CFILE=SYSDIR ; Sorts DOTEMP output on cols 1-12
LSTOOL ; Now to compare with MAST.LST
SSORT CFILE=DIFILE ; Sorts cols 14-17 (d/u) to LIST2.DIF
PASTE $1 LIST.DIF LIST2.DIF ; Creates final file from DIF lists
ERASE LIST.DIF,LIST2.DIF,SYSDIR.TXT ; Kills input files for us
LISTT16 $1 ; Jumps into print mode for us
Where the command line entered was:
A9>RAID AB17
And the final output file would be named AB17 with no extension.
You could easily construct a SUBMIT or ZEX (ZCPR3 only) file to
do the same thing for you depending on your preferences.
Additional advice:
The CLS.DAT file for DOTEMP is set for the clear screen sequence
of a Kaypro. Configure to your own clear screen sequence and the
program(s) are ready to run.
Copyright/Source:
DOTEMP is a relatively simple program written in MBASIC and
compiled with BASCOM. Any programmer of reasonable means
could very easily write it and perhaps even better than I.
However I have no intention of releasing the source for it
either now or in the future. Use as you desire for your
own purposes but do not sell or mutilate it.
Happy Raiding!!!
Thomas L. Ensminger
AB17 Remote Systems
[513] 879-6263
24hrs 3/12/24bps