Compression - "Down Pillow" Control Code Creation: Step 18 of 88

Author

"Down Pillow" Control Code Creation: Step 18 of 88

Plato777

2004-10-16, 3:55 pm

Creation of the "Down Pillow"
Compression Algorithm Control
Codes: Step 18 of 88

I am posting a series of steps
to comp.compression so that
there can be an external record
of my efforts to create a
sucessful copy of control codes
to use in the "Down Pillow"
compression algorithm as
discussed earlier this year.

Erik T. Evenson
====================================

Step 18 of 88

The label listing in dpc10_01.dat must
be relisted to exclude those labels used
to create the special pre-arrangement
groups in step 16. Make sure the label
listing file dpc10_01.dat and the
database file dpc17_02.dat are located
in the working directory when running
the following QBASIC program. The
program creates the new label listing
file dpc18_01.dat.

---------------------------------------
REM QBASIC program: STEP18.BAS
REM This program relists the label
' listing in dpc10_01.dat to exclude
' those labels used in the special
' group pre-arrangement in step 16.
' Make sure the files dpc10_01.dat and
' dpc17_02.dat are located in the
' working directory. The program
' creates the output file dpc18_01.dat.
OPEN "I", #1, "dpc10_01.dat"
OPEN "R", #2, "dpc17_02.dat", 3
FIELD #2, 1 AS a1$, 1 AS a2$, 1 AS a3$
OPEN "O", #3, "dpc18_01.dat"
LET d = 1
10 IF EOF(1) THEN GOTO 30
LINE INPUT #1, b$
IF d < 3145729 THEN LET e = 3
IF d < 2097153 THEN LET e = 2
IF d < 1048577 THEN LET e = 1
LET b& = VAL(b$)
GET #2, b&
IF e = 3 THEN LET f$ = a3$
IF e = 2 THEN LET f$ = a2$
IF e = 1 THEN LET f$ = a1$
IF f$ = CHR$(0) THEN GOTO 20
IF f$ = CHR$(1) THEN GOTO 25
20 PRINT #3, b$
25 LET d = d + 1
GOTO 10
30 CLOSE
END
----------------------------------------

More to follow.

Eric D. Brown

2004-10-16, 3:55 pm

Eric D. Brown

2004-10-16, 3:55 pm

Plato777 wrote:
> Creation of the "Down Pillow"
> Compression Algorithm Control
> Codes: Step 18 of 88
>
> I am posting a series of steps
> to comp.compression so that
> there can be an external record
> of my efforts to create a
> sucessful copy of control codes
> to use in the "Down Pillow"
> compression algorithm as
> discussed earlier this year.
>
> Erik T. Evenson
> ====================================
>
> Step 18 of 88
>
> The label listing in dpc10_01.dat must
> be relisted to exclude those labels used
> to create the special pre-arrangement
> groups in step 16. Make sure the label
> listing file dpc10_01.dat and the
> database file dpc17_02.dat are located
> in the working directory when running
> the following QBASIC program. The
> program creates the new label listing
> file dpc18_01.dat.
>
> ---------------------------------------
> REM QBASIC program: STEP18.BAS
> REM This program relists the label
> ' listing in dpc10_01.dat to exclude
> ' those labels used in the special
> ' group pre-arrangement in step 16.
> ' Make sure the files dpc10_01.dat and
> ' dpc17_02.dat are located in the
> ' working directory. The program
> ' creates the output file dpc18_01.dat.
> OPEN "I", #1, "dpc10_01.dat"
> OPEN "R", #2, "dpc17_02.dat", 3
> FIELD #2, 1 AS a1$, 1 AS a2$, 1 AS a3$
> OPEN "O", #3, "dpc18_01.dat"
> LET d = 1
> 10 IF EOF(1) THEN GOTO 30
> LINE INPUT #1, b$
> IF d < 3145729 THEN LET e = 3
> IF d < 2097153 THEN LET e = 2
> IF d < 1048577 THEN LET e = 1
> LET b& = VAL(b$)
> GET #2, b&
> IF e = 3 THEN LET f$ = a3$
> IF e = 2 THEN LET f$ = a2$
> IF e = 1 THEN LET f$ = a1$
> IF f$ = CHR$(0) THEN GOTO 20
> IF f$ = CHR$(1) THEN GOTO 25
> 20 PRINT #3, b$
> 25 LET d = d + 1
> GOTO 10
> 30 CLOSE
> END
> ----------------------------------------
>
> More to follow.

Also, here's a suggestion for any ideas you may have in the future...
start one thread, and reply to the one thread with added information
(instead of creating 88 different threads).

Another note, instead of 88 different steps... why don't you create ONE
program and use different Subs & Functions, opening all the files you
need at one instance, and actually assign arrays in memory to use as
working buffers. It'd not ONLY be MUCH more efficient, but it'd stop
the confusion of trying to place together 88 different pieces of a
puzzle. Then again, I think that's what you're going after... isn't it?
Confusion? Because otherwise, I don't see why you can't simply wait
to construct a single working compressor/decompressor!

Also, I know you're going for a "small sized" approach, but QBasic?
Newsflash, it's not 1987 any longer... (then again, from your funding
plea post... perhaps that's all your computer will run). What's your
expected throughput on operations of this project? 100 bytes per minute?

Plato777

2004-10-17, 3:55 am

"Eric D. Brown" <nospam@nowhere.com> wrote in message news:<10n2t269o539peb@corp.supernews.com>...
> Plato777 wrote:
>
> With all the opening and closing of files throughout your little ordeal,
> I bet your hard drive is wishing it had a few aspirins.

Hard drive? What hard drive?
Can you say R A M D I S K .

EE.

Plato777

2004-10-17, 3:55 am

"Eric D. Brown" <nospam@nowhere.com> wrote in message news:<10n2trj6pmf5657@corp.supernews.com>...

> Also, here's a suggestion for any ideas you may have in the future...
> start one thread, and reply to the one thread with added information
> (instead of creating 88 different threads).

Too confusing if someone has something to say about a certain step.
Furthermore, the thread may become very long.

>
> Another note, instead of 88 different steps... why don't you create ONE
> program and use different Subs & Functions, opening all the files you
> need at one instance, and actually assign arrays in memory to use as
> working buffers. It'd not ONLY be MUCH more efficient, but it'd stop
> the confusion of trying to place together 88 different pieces of a
> puzzle. Then again, I think that's what you're going after... isn't it?
> Confusion? Because otherwise, I don't see why you can't simply wait
> to construct a single working compressor/decompressor!

I would have to determine all steps before I can do that.
Posting step by step will keep me honest and will force
me to address any mistakes or obstacles that come my
way. I just want to remind people that Creation can be just
as messy as Destruction. Chaos must precede Order.

>
> Also, I know you're going for a "small sized" approach, but QBasic?
> Newsflash, it's not 1987 any longer... (then again, from your funding
> plea post... perhaps that's all your computer will run). What's your
> expected throughput on operations of this project? 100 bytes per minute?

Qbasic is so simple to learn and implement. If my algorithm is
successful I want everyone to make a copy if they want one.
100 Bytes/min?, perhaps. However, if my algorithm works the
compression work can be split between 1M computers--Parallel processing.
The beta-buffer is composed of 1048576 differently computable alpha-buffer parts.

EE.

Matt Mahoney

2004-10-17, 8:55 pm

"Eric D. Brown" <nospam@nowhere.com> wrote in message
news:10n2trj6pmf5657@corp.supernews.com...
> Another note, instead of 88 different steps... why don't you create ONE
> program and use different Subs & Functions, opening all the files you
> need at one instance, and actually assign arrays in memory to use as
> working buffers. It'd not ONLY be MUCH more efficient, but it'd stop
> the confusion of trying to place together 88 different pieces of a
> puzzle. Then again, I think that's what you're going after... isn't it?
> Confusion? Because otherwise, I don't see why you can't simply wait
> to construct a single working compressor/decompressor!

If he did that, it would be too obvious that his program doesn't work.

-- Matt Mahoney

Sponsored Links