Cobol - Strange result : I lost one record

Author

Strange result : I lost one record

Jerome

2006-07-21, 7:55 am

2006-07-21, 7:55 am

In article <e9qf6t$bjv$1@s1.news.oleane.net>,
Jerome <j.aubert1@free.fr> wrote:

[snip]

>Then I read the second file "fgcart" sequentially and I search in the
>first file "fgctar" all records for which primary key of fgcart match
>primary key of fgctar

Please pardon the midsentence interruption here... how, in COBOL, do you
'search the first file (for) all records' with a matching fractional key?

[snip]

>Has someone an idea ?

Yes... my idea is that you post the code which answers the question I
posed above, as a START.

DD

epc8@juno.com

2006-07-21, 6:55 pm

Jerome wrote:
> Hello,
>
> I've got a file "fgctar". When I read it sequentially, I obtain this
>
> KEY | VALUE
> ...
> 152001 | MyRecord1
> 153001 | MyRecord2
> 153153 | MyRecord3
> 154001 | MyRecord4
> 154154 | MyRecord5
> 155001 | MyRecord6
> 156001 | MyRecord7
> 156156 | MyRecord8
> ...
>
> I've got another file "fgcart", the primary key is composed by the first
> three digits of the primary key of my first file "fgctar". When I read
> it, i've got something like this :
>
> KEY | NAME
> ...
> 152 | Name1
> 153 | Name2
> 154 | Name3
> 155 | Name4
> 156 | Name5
> ...
>
> Then I read the second file "fgcart" sequentially and I search in the
> first file "fgctar" all records for which primary key of fgcart match
> primary key of fgctar and I've got this result
>
> KEY | NAME | VALUE
> ...
> 152001 | Name1 | MyRecord1
> 153001 | Name2 | MyRecord2
> 153153 | Name2 | MyRecord3
> 154001 | Name3 | MyRecord4
> 155001 | Name4 | MyRecord6
> 156001 | Name5 | MyRecord7
> 156156 | Name5 | MyRecord8
> ...
>
> So, the record "154154 MyRecord5" disappeared.
>
> Has someone an idea ?
>
> Jerome

** Editorial comments **

1=2E It would help to post your problem and program listing with copy
members expanded in one post.

2=2E It would help to give your file identifiers more meaningful names.

3=2E It is also confusing that these identifiers are *anagrams*.

** End editorial comments **

** expanded program listing inserted **

INPUT-OUTPUT SECTION.
FILE-CONTROL.
SELECT FGCTAR ASSIGN TO RANDOM "fgctar01"
ORGANIZATION INDEXED
ACCESS DYNAMIC
RECORD KEY GCTAR-CLE
STATUS ST-GCTAR.
SELECT FGCART ASSIGN TO RANDOM "fgcart01"
ORGANIZATION IS INDEXED
ACCESS IS DYNAMIC
RECORD KEY IS GCART-CLE
ALTERNATE RECORD KEY IS GCART-NOM WITH DUPLICATES
FILE STATUS IS FichierStatus.
SELECT OUTART ASSIGN TO RANDOM "out-ess.txt"
ORGANIZATION IS LINE SEQUENTIAL.
DATA DIVISION.
FILE SECTION.
* COPY "FGCART.CPY".
FD FGCART LABEL RECORD STANDARD.
01 GCART-FILE.
02 GCART-CLE.
04 GCART-NUM PIC X(12).
* COPY "FGCTAR.CPY".
FD FGCTAR LABEL RECORD STANDARD.
01 GCTAR-FILE.
02 GCTAR-CLE.
04 GCTAR-ART PIC X(12).
04 GCTAR-TAR PIC 999.
FD OUTART LABEL RECORD STANDARD.
01 OUT-ART.
02 OUT-CLE PIC X(12).
02 SEPAR2 PIC X(1).
* 02 OUT-INTITULE PIC X(30).
02 SEPAR3 PIC X(1).
02 OUT-PUHT PIC 9(6).9(2).
02 SEPAR4 PIC X(1).
* 02 OUT-PUTTC PIC 9(6).9(2).
02 SEPAR5 PIC X(1).
02 OUT-TAR.
04 OUT-CLEART PIC X(12).
04 OUT-CLETAR PIC 9(3).

WORKING-STORAGE SECTION.
01 FichierStatus PIC X(2).
01 ST-GCTAR PIC X(2).
88 RecordFound VALUE "00".
01 cpt PIC 9(3) VALUE 000.
01 tmp PIC 9(6).9(2).

01 FLAGTROUVE PIC 9 VALUE 0.

PROCEDURE DIVISION.
TRAITEMENT SECTION.
DEBUT.
OPEN INPUT FGCART.
OPEN INPUT FGCTAR.
OPEN OUTPUT OUTART.

MOVE ";" TO SEPAR2 SEPAR3 SEPAR4 SEPAR5.

PERFORM UNTIL GCART-CLE =3D HIGH-VALUES
MOVE LOW-VALUES TO GCART-CLE
START FGCART
KEY IS GREATER THAN GCART-CLE
INVALID KEY MOVE HIGH-VALUES TO GCART-CLE
END-START
PERFORM UNTIL GCART-CLE =3D HIGH-VALUES
READ FGCART
NEXT RECORD
AT END MOVE HIGH-VALUES TO GCART-CLE
NOT AT END
* Lecture du 2eme fichier
MOVE GCART-CLE TO GCTAR-ART
MOVE 000 TO GCTAR-TAR
MOVE 0 TO FLAGTROUVE
* Balayage de toutes les valeurs possibles de GCTAR-TAR
PERFORM VARYING cpt FROM 1 BY 1 UNTIL cpt =3D 999
MOVE cpt TO GCTAR-TAR
READ FGCTAR
INVALID KEY CONTINUE
END-READ
IF GCTAR-CLE =3D "154 154"
DISPLAY ST-GCTAR
END-IF
IF RecordFound
MOVE 1 TO FLAGTROUVE
MOVE GCART-CLE TO OUT-CLE
MOVE GCTAR-CLE TO OUT-TAR
MOVE GCTAR-PUHT TO OUT-PUHT
WRITE OUT-ART
END-IF
END-PERFORM
* Si on a pas trouv=E9 d'enregistrement
IF FLAGTROUVE =3D 0
MOVE GCART-CLE TO OUT-CLE
MOVE GCTAR-CLE TO OUT-TAR
MOVE GCTAR-PUHT TO OUT-PUHT
WRITE OUT-ART
END-IF

* Fin de traitement du 2eme fichier
END-READ
END-PERFORM
END-PERFORM

CLOSE FGCART.
CLOSE FGCTAR.
CLOSE OUTART.
FIN-TRAITEMENT SECTION.

SORTIE SECTION.
AFFICHAGE.
DISPLAY "Fin du traitement ", ST-GCTAR.
FIN-SORTIE SECTION.

** end expanded program listing **

Using database terminology, you are joining two tables. Let's call
GCART your name table and GCTAR your transaction table.

1=2E You are using an alternate key on your name table. There is only one
key, the name_id_number. Second, the identifier GCART-NOM does not
exist in your program. Third you are declaring it WITH DUPLICATES but
the keys in the name table are unique.

2=2E For each record in the name table you access the transaction table
999 times looking for all possible keys. So it really does not matter
why your program does not work properly.

I suggest you try one of two things.

1=2E make the key on the transaction file just the first three characters
and use WITH DUPLICATES on that key. For each record in the name file,
look up all matching records in the transaction file.

2=2E You may want to process these tables differently. It depends on the
relative sizes of the files and other factors. For each record in the
transaction table, look up the name in the name table. If it exists,
output a detail record, otherwise do something else, perhaps nothing.

-- elliot

epc8@juno.com

2006-07-22, 9:55 pm

docdwarf@panix.com wrote:
> In article <e9qf6t$bjv$1@s1.news.oleane.net>,
> Jerome <j.aubert1@free.fr> wrote:
>
> [snip]
>
>
> Please pardon the midsentence interruption here... how, in COBOL, do you
> 'search the first file (for) all records' with a matching fractional key?
>
> [snip]
>
>
> Yes... my idea is that you post the code which answers the question I
> posed above, as a START.
>
> DD

As Ernie Harwell might say, at this point the OP is "loooooooooooooong
gooooooooooone".

But it is not that hard to solve because the fractional key occupies
the leading characters of the key. In addition, the OP can simplify his
program by reading one of the indexed files sequentially.

-- elliot

2006-07-22, 9:55 pm

In article <1153584654.976174.253550@p79g2000cwp.googlegroups.com>,
<epc8@juno.com> wrote:
>docdwarf@panix.com wrote:
>
>As Ernie Harwell might say, at this point the OP is "loooooooooooooong
>gooooooooooone".

Ah well... my habits of bathing, or lack thereof, might be responsible.

DD

James J. Gavan

2006-07-22, 9:55 pm

epc8@juno.com wrote:
> docdwarf@panix.com wrote:
>
>
>
> (1) As Ernie Harwell might say, at this point the OP is "loooooooooooooong
> gooooooooooone".
>
> (2) But it is not that hard to solve because the fractional key occupies
> the leading characters of the key. In addition, the OP can simplify his
> program by reading one of the indexed files sequentially.
>

#2 - His problem

I agree with your comments. I don't recall but you may have been the
person who gave observations/corrections to his program, One thing I did
notice though was that he had the same External name for two of his
files - possibly a typo ?.

Somebody from Brazil asked, "How can you program the Help File feature
that comes with Windows ?". Response he got from M/F was, "Use Activex".
Having already designed generic support classes for file/Table handling
and GUI-ing, it didn't seem a big deal to me. (Ironically one of the M/F
examples going back to VISOC the precursor to Net Express, has this
SEARCH technique).

So produce a Help File Listbox and "setContents" using a Sorted
Collection. Now user has a text entry field to enter one or more
characters and using method "wantAllKeys" the event returns keystrokes
to-date, after each keystroke.

Assume Listbox contains -

Gorgonzola
Gouda

For following checking, turn BOTH the input text and the Listbox entry
(element) to either Uppercase or Lowercase for comparison purposes.

a) user enters "G" (search input (1:1) against Listbox entries (1:1)
then highlight Gorgonzola. (If there aren't any "Gs" - nothing happens).
b) same happens when user continues and enters "Go" - (1:2)
c) User enters "Gou" - (1:3) - now Gouda gets highlighted - if that's
the one the user wants he/she clicks on the Listbox row and the demo
program using the index position for that row displays "You selected Gouda".

Why go buy an Activex/Cmponent, when the only addition required was the
Search coding.

Using a fixed RefMod as applies to the OP's files - should be even simpler.

(Funnily enough my wife was watching TV this morning and there was a
topic about the 20-35 age group getting into text messaging as opposed
to e-mail and using abbreviations - TTYL, LOL etc., yet people quizzed
didn't recognize IMHO (In My Humble Opinion you are an XXXXXXX and this
is the way it should be done). Confession - I haven't a bloody clue what
OP really means :-) Absolute wild guess on reflection, OP = Other Person ?)

# 1 - Acknowledgment

Oh so true ! How many times a half-baked question is phrased, it is
assummed you know that they are working on big iron or a PC or perhaps
in that centrifugal tunnel at CERN. It's sometimes left to you to figure
out which compiler and version number. If you are real lucky, they are
actually writing about COBOL.

They have a problem and, "I'm getting the following sorts of errors....
So, this is what is happening.....". So not knowing the problem they
can somehow figure out what is causing it, so that they can provide us
with as litttle informaton as possible (obfuscation).

What happened to communicaton ? Some four decades ago I go into this
racket as a Systems Analyst - and that resulted in the observation that
EDP, (let's be politically correct "IT"), was the automation of paper
shuffling; and taking that process, cutting out the irrelvant bumph and
making the operation slicker and cheaper. Cheaper. Hmmm. Have we helped
reduce the paper flow ? (Note to self - "Invest some money in the BC
lumber/paper mill industry").

Some queries are from individuals - cottage industry workers; on the
other hand they quite often could be working at an installation where
there are other warm bodies also programming. A has a problem. Why
doesn't he ask B who sits three feet from him ? As a general rule, I
can't believe his co-workers would be anti to being asked questions. So
I tend to conclude it is A's self worth - doesn't want to admit he
doesn't know, so comes here in a vague way, asking for help and hopes
that if one or two of us throw some runes, we will come up with the
answer *he/she* wants.

If A asks B a quick question, off-hand he will get a suggested solution.
Two dayas later A stilll doen't have it figured out. So in the nicest
possible way he asks B to help with the problem. B's response this time,
"Show me your source". How novel. But when it comes to asking us,
short-change us where we are supposed to guess without some inkling of
what is in the source listing.

A classic was the recent query on moving a combined FirstName/LastName
field into separate First Name and Last Name fields. Answer - there is
no known generic way of doing this. Observation - the OP is using the
Net to get into clc. How many times has he signed up for something on
the Net where it just asks for "your name" - 99.999% of the time, as a
minimum you are requested to enter as separate fileds you First Name and
then your Last Name. The problem goes back to the original design and
not trying to shuffle the 'unknown' into 'known' fields. (The
alternative of course is Corporate Name - any applicaton allowing for
personal names or a corporate name probably requires re-definition of
the name fileds).

Perhaps for their sin of taking programmng up as a career, new graduates
should be forced to do a year of penance in Purgatory - i.e., shuffling
bits of paper in an office !

Of course a quick "Thank You" doesn't hurt. We've had a couple recently
- but they are preety rare.

Jimmy

epc8@juno.com

2006-07-22, 9:55 pm

James J. Gavan wrote:
> epc8@juno.com wrote:
>
> #2 - His problem
>
> I agree with your comments. I don't recall but you may have been the
> person who gave observations/corrections to his program, One thing I did
> notice though was that he had the same External name for two of his
> files - possibly a typo ?.

I found the original program so hard to follow that I ran off a program
listing and started marking it up in pencil. Having files FGCTAR and
FGCART did not help either. My suggestions were somewhat off the mark
but I did point out that searching a file repeatedly for all possible
variations of a partial key instead of using the partial key directly
was not the way to go.

[snip discussion on search techniques]

Confession - I haven't a bloody clue what
> OP really means :-) Absolute wild guess on reflection, OP = Other Person ?)

Original Poster, I think. In CLC maybe it is Original Pest? :-).

>
> # 1 - Acknowledgment
>
> Oh so true ! How many times a half-baked question is phrased, it is
> assummed you know that they are working on big iron or a PC or perhaps
> in that centrifugal tunnel at CERN. It's sometimes left to you to figure
> out which compiler and version number. If you are real lucky, they are
> actually writing about COBOL.
>
> They have a problem and, "I'm getting the following sorts of errors....
> So, this is what is happening.....". So not knowing the problem they
> can somehow figure out what is causing it, so that they can provide us
> with as litttle informaton as possible (obfuscation).
>
> What happened to communicaton ?

Yes, this I think is the key idea. Back in ye olde days computer
resources were centralized. In a typical computing center people
gathered to punch cards, wait for their jobs to run, and to look at
their print outs. So programming was partly a social activity. Asking
others for help or helping others, or sitting down with a program
listing and a pencil, was common.

Now people have enormous computing power on their desktops so they can
work or study alone. But they also have the resources of the internet.

It bothers me that people who are obviously using the internet to post
don't have the knowledge or the time or the inclination to read or do a
google groups search or a web search before posting. I'm old enough to
start blathering about instant gratificaton, etc. but I won't.

Second, it amazes me that many people are not taught how to solve
problems. They don't know how to formulate a problem succinctly. They
don't know how to gather and present relevant information to others
which makes the problem solving process easier. They don't know how to
narrow down their problem to a small test case, etc.

Instead someone posts a long program listing, some rubbish output, and
asks "What's wrong with my program?"

[flame on] Perhaps it is sun spots, better nutrition or earlier puberty
if you will, but I think part of the blame lies in the de-emphasis in
education on math and science. IMO students do not learn that some
questions do have right answers and that there are situations where you
can't get something for nothing. [flame off]

-- elliot

2006-07-22, 9:55 pm

In article <1153620715.182694.257550@h48g2000cwc.googlegroups.com>,
<epc8@juno.com> wrote:

[snip]

>I'm old enough to
>start blathering about instant gratificaton, etc. but I won't.

[snip]

>[flame on] Perhaps it is sun spots, better nutrition or earlier puberty
>if you will, but I think part of the blame lies in the de-emphasis in
>education on math and science. IMO students do not learn that some
>questions do have right answers and that there are situations where you
>can't get something for nothing. [flame off]

.... an' don't git me started 'bout what them kids're callin' 'music'
nowadays, neither... buncha durned *noise*, 'f ya ask me... an' another
thing, what about them clothes they're wearin'? Ain't they got no mothers
t' shame 'em?

.... an' th' language they use... buncha nonsense, every word... 'Daddy-o'
an' 'hey-bob-a-ree-bop' an' 'it's a real blast'... an' those fool 'sock
hops', where'd they git th' idea fer such foolishment?

.... an' another thing!... zzzzzzzzz

DD

epc8@juno.com

2006-07-23, 3:55 am

docdwarf@panix.com wrote:
> In article <1153620715.182694.257550@h48g2000cwc.googlegroups.com>,
> <epc8@juno.com> wrote:
>
> [snip]
>
>
> [snip]
>
>
> ... an' don't git me started 'bout what them kids're callin' 'music'
> nowadays, neither... buncha durned *noise*, 'f ya ask me... an' another
> thing, what about them clothes they're wearin'? Ain't they got no mothers
> t' shame 'em?
>
> ... an' th' language they use... buncha nonsense, every word... 'Daddy-o'
> an' 'hey-bob-a-ree-bop' an' 'it's a real blast'... an' those fool 'sock
> hops', where'd they git th' idea fer such foolishment?
>
> ... an' another thing!... zzzzzzzzz
>
> DD

Well, OK. I've temporarily slipped into old f*rt mode.

It just gets harder to participate in this NG when from time to time an
OP vanishes into a puff of smoke, leaving behind a program where the
best answer to "How do I fix this?" might be "Tear it up and start
over."

-- elliot

2006-07-23, 7:55 am

In article <1153627181.705171.123850@i42g2000cwa.googlegroups.com>,
<epc8@juno.com> wrote:
>
>docdwarf@panix.com wrote:

[snip]
[color=darkred]
>
>Well, OK. I've temporarily slipped into old f*rt mode.

Happens to some folks, I believe, if they live long enough... and some
others even if they don't.

>
>It just gets harder to participate in this NG when from time to time an
>OP vanishes into a puff of smoke, leaving behind a program where the
>best answer to "How do I fix this?" might be "Tear it up and start
>over."

It gets difficult to do that on the job, too, when the orders assume the
ancient 'there'll be time to do it over but there isn't time to do it
right'... part o' life, from what I've seen.

DD

Jerome

2006-07-24, 3:55 am

epc8@juno.com a écrit :

>
> I suggest you try one of two things.
>
> 1. make the key on the transaction file just the first three characters
> and use WITH DUPLICATES on that key. For each record in the name file,
> look up all matching records in the transaction file.
>
> 2. You may want to process these tables differently. It depends on the
> relative sizes of the files and other factors. For each record in the
> transaction table, look up the name in the name table. If it exists,
> output a detail record, otherwise do something else, perhaps nothing.
>
> -- elliot
>

Thanks a lot for your answer. I'm not the original programmer and i'm
not a cobol programmer so I'm lost. But your analyse will be very helpfull.

Best regards

Jerome

2006-07-24, 3:55 am

In article <ea21mk$sh8$1@s1.news.oleane.net>,
Jerome <j.aubert1@free.fr> wrote:
>epc8@juno.com a écrit :
>

[snip]
[color=darkred]
>
>Thanks a lot for your answer. I'm not the original programmer and i'm
>not a cobol programmer so I'm lost.

I suggest you try one of the following:

1) Tell your boss that if he put a carpenter to work building brick walls
then the house might fall down.

2) Realise that if your boss insists on putting a carpenter to work
building brick walls that the strength of the house that results is,
obviously, not very important.

DD

Jerome

2006-07-24, 7:55 am

epc8@juno.com a écrit :

>
> Using database terminology, you are joining two tables. Let's call
> GCART your name table and GCTAR your transaction table.
>
> 1. You are using an alternate key on your name table. There is only one
> key, the name_id_number. Second, the identifier GCART-NOM does not
> exist in your program. Third you are declaring it WITH DUPLICATES but
> the keys in the name table are unique.

>
> -- elliot
>

Hello,

I tried to delete the sentence :

ALTERNATE RECORD KEY IS GCART-NOM WITH DUPLICATES

but when i run the program, i've got this I/O error :

cobol I/ error 39,52 on fgcart file ...

I use RM/COBOL-85 Runtime version 5.36.00 by Liant Software and this
error is not referrenced in the user manual guide. I've already submit a
question on this ng and someone told me that it is a hexadecimal code.
Ok, after differents tests, i wrote this :

ALTERNATE RECORD KEY IS GCART-NOM WITH DUPLICATES

and since, i'haven't this error.

What do you think about it ?

Best regards,

Jerome

epc8@juno.com

2006-07-24, 6:55 pm

Jerome wrote:
> epc8@juno.com a =E9crit :
>
>
> Thanks a lot for your answer. I'm not the original programmer and i'm
> not a cobol programmer so I'm lost. But your analyse will be very helpful=
l=2E
>
> Best regards
>
> Jerome

See http://en.wikipedia.org/wiki/JOIN for a relevant discussion.
(Sorry. I could not find a comparable article at fr.wikipedia.org.)
While there is a "Java for Cobol Programmers" book I don't know of a
"Cobol for Some_other_language Programmers book". Thane Hubbell's
"Teach Yourself Cobol in 24 Hours" (Sams) does present an introduction
to COBOL. See the chapters on INDEXED file processing. (Murach
publishes professional level books on COBOL.)

After thinking about the problem some more, my strategy changed
somewhat. The major features are to read the name file sequentially and
look up records with a matching key in the transaction file. Since the
3 digit name-id-number occupies the first three characters of the 6
digit key in the transaction file, you can search the second file using
a query that consists of the first 3 digits followed by 3 spaces.
(Moving a shorter data item to a larger one pads right with spaces - if
you define these fields as character (X) rather than numeric (9).)

So first here is some pseudo code:

for each record in the name file
position the record pointer in the transaction file to key greater
or equal
to "nnn<space><space><space>"
while name-id-from-name-file matches first-3-characters-of-long-key
do
output detail record to output file

Here is some COBOL using a compiler comparable to the original source
code. Variable names and field lengths are my own choices. Note that my
program does no real error checking and it does no file status checking
as it really should do....

identification division.
program-id. join.
environment division.
configuration section.
source-computer. ibm-pc.
object-computer. ibm-pc.
input-output section.
file-control.
select name-file assign to disk 'name.dat'
organization is indexed
record key is name-key.
select trans-file assign to disk 'trans.dat'
organization is indexed
access mode is dynamic
record key is trans-key.
select out-file assign to disk 'out.txt'
organization is line sequential.
data division.
file section.
fd name-file.
01 name-rec.
05 name-key pic x(3).
05 name-data pic x(5).
fd trans-file.
01 trans-rec.
05 trans-key.
10 trans-name-id pic x(3).
10 trans-num pic x(3).
05 trans-data pic x(9).
fd out-file.
01 out-rec.
05 out-key pic x(6).
05 filler pic x.
05 out-name pic x(5).
05 filler pic x.
05 out-data pic x(9).
working-storage section.
01 flags.
05 name-eof pic x value 'n'.
05 trans-eof pic x.
05 key-not-found pic x.
procedure division.
main.
open input name-file
open input trans-file
open output out-file
move all '|' to out-rec
perform until name-eof is equal to 'y'
read name-file
at end
move 'y' to name-eof
not at end
perform get-match
end-read
end-perform
close name-file
close trans-file
close out-file
stop run.
get-match.
move name-key to trans-key
start trans-file key not < trans-key
invalid key
move 'y' to key-not-found
not invalid key
move 'n' to trans-eof
move 'n' to key-not-found
perform until trans-eof is equal to 'y'
or key-not-found is equal to 'y'
read trans-file next record
at end
move 'y' to trans-eof
not at end
if name-key is not equal to trans-name-id
move 'y' to key-not-found
else
move trans-key to out-key
move name-data to out-name
move trans-data to out-data
write out-rec
end-if
end-read
end-perform
end-start.

Good luck.

-- elliot

NG regulars might complain about the non-standard quote marks but at
least it is '85 standard. :-). I am sure I could also use one flag as
an exit condition on the match-records loop too.

Richard

2006-07-24, 6:55 pm

Jerome wrote:

> I tried to delete the sentence :
>
> ALTERNATE RECORD KEY IS GCART-NOM WITH DUPLICATES
>
> but when i run the program, i've got this I/O error :
>
> cobol I/ error 39,52 on fgcart file ...
>
> I use RM/COBOL-85 Runtime version 5.36.00 by Liant Software and this
> error is not referrenced in the user manual guide.

Yes, it is. It is in the block 39, 50 through 39, 5E and states that
there is a mismatched key length for keys 0 through 14. 52 refers to
key 2.

The definition of the file in the FD must match the actual file as it
was created, you can't just change the FD and expect the data and
indexes to recreated themselves by morphing the bits on the disk.

If you want a file with a different configuration of keys then you must
rebuild the data into a file of that new definition.

> I've already submit a
> question on this ng and someone told me that it is a hexadecimal code.
> Ok, after differents tests, i wrote this :
>
> ALTERNATE RECORD KEY IS GCART-NOM WITH DUPLICATES
>
> and since, i'haven't this error.

You put the definition back to what it should be for that file. You
fixed what you broke.

epc8@juno.com

2006-07-24, 6:55 pm

Richard wrote:
> Jerome wrote:
>

I am more

now than before. How does the OP get his program, as
listed, to compile?

The identification division is missing.
The primary key and the alternate key are the same.
One variable is undeclared - it does not appear a FD.

How does the program, if it compiles and runs, produce the output
given?????

The data as listed for the files does not match the FDs given. The only
common item in both files is a 3 digit number which is NOT listed as
the key. The key is a PIC X(12) to which OTHER digits are affixed to
produce the second key.
[color=darkred]
> The definition of the file in the FD must match the actual file as it
> was created, you can't just change the FD and expect the data and
> indexes to recreated themselves by morphing the bits on the disk.
>
> If you want a file with a different configuration of keys then you must
> rebuild the data into a file of that new definition.
>
>
> You put the definition back to what it should be for that file. You
> fixed what you broke.

The original is still broken in other ways.

It is still unclear what his data files really look like - or how they
were created.

What is the actual common key? Names? Three digits? The code says one
thing, the listed data something else and the explanation something
else all together.

-- elliot [resuming my search for an apiary in Sussex]

James J. Gavan

2006-07-24, 9:55 pm

epc8@juno.com wrote:
>
> -- elliot [resuming my search for an apiary in Sussex]

Given that the word apiary has Latin origins, following might narrow
your search. Try the Roman Villa site at Bignor, on the way to Bognor
Regis or Chichester :-).

If you don't find beez at least you'll be able to look at some
fascinating floor mosaics. (Apparently done a lot to the site since I
saw it in late 'fifties).

Jimmy

epc8@juno.com

2006-07-25, 3:55 am

James J. Gavan wrote:
> epc8@juno.com wrote:
>
> Given that the word apiary has Latin origins, following might narrow
> your search. Try the Roman Villa site at Bignor, on the way to Bognor
> Regis or Chichester :-).
>
> If you don't find beez at least you'll be able to look at some
> fascinating floor mosaics. (Apparently done a lot to the site since I
> saw it in late 'fifties).
>
> Jimmy

Thanks for the itinerary. Actually I'm not planning a trip to the UK in
the near future. :-). My comment was a reference to the parting words
of a well known fictional character. I'll leave that as a bit of a
mystery but this itself will provide a small clue.

-- elliot

Richard

2006-07-25, 3:55 am

James J. Gavan wrote:

> Given that the word apiary has Latin origins, following might narrow
> your search. Try the Roman Villa site at Bignor, on the way to Bognor
> Regis or Chichester :-).

And just the other side of Chichester along the same Stane Street is
Fishbourne Villa which wasn't discovered until the 60s.

Jerome

2006-07-25, 3:55 am

epc8@juno.com a écrit :

>
> I am more

now than before. How does the OP get his program, as
> listed, to compile?

> -- elliot [resuming my search for an apiary in Sussex]
>
OK, I'm going to work with your code and going to read carefully your
explanations (and the others contrib).

I thank you again for your help

Best regards

Jerome

epc8@juno.com

2006-07-25, 7:55 am

Jerome wrote:

> OK, I'm going to work with your code and going to read carefully your
> explanations (and the others contrib).

Good luck.

One point Richard made is particularly important. If you change the
program then you will have to create files containing your data that
follow that layout. For ordinary "line sequential" (text) files you can
use a standard text editor. For organization is indexed files you
generally write a program that builds these files by reading some other
type of file and writing records out to the indexed file.

Here are some sample data files:

---- name.txt ----
152Name1
153Name2
154Name3
155Name4
156Name5
---- end of file ----

---- trans.txt ----
152001MyRecord1
153001MyRecord2
153153MyRecord3
154001MyRecord4
154154MyRecord5
155001MyRecord6
156001MyRecord7
156156MyRecord8
---- end of file ----

Here is a program to read the first .txt file and create the name
table:

---- loadname.cbl ----
identification division.
program-id. loadname.
environment division.
configuration section.
source-computer. ibm-pc.
object-computer. ibm-pc.
input-output section.
file-control.
select name-text assign to disk 'name.txt'
organization is line sequential.
select name-file assign to disk 'name.dat'
organization is indexed
record key is name-key.
data division.
file section.
fd name-text.
01 text-rec.
05 text-key pic 9(3).
05 text-data pic x(5).
fd name-file.
01 name-rec.
05 name-key pic 9(3).
05 name-data pic x(5).
working-storage section.
01 flags.
05 name-eof pic x value 'n'.
procedure division.
main.
open input name-text
open output name-file
perform until name-eof is equal to 'y'
read name-text
at end
move 'y' to name-eof
not at end
move text-rec to name-rec
display name-key '|' name-data '|'
write name-rec
end-read
end-perform
close name-text
close name-file
stop run.
---- end of file ----

Here is a program to read the second text file and write the
transaction table:

---- loadtran.cbl ----
identification division.
program-id. loadtrans.
environment division.
configuration section.
source-computer. ibm-pc.
object-computer. ibm-pc.
input-output section.
file-control.
select trans-text assign to disk 'trans.txt'
organization is line sequential.
select trans-file assign to disk 'trans.dat'
organization is indexed
record key is trans-key.
data division.
file section.
fd trans-text.
01 text-rec.
05 text-key pic 9(6).
05 text-data pic x(9).
fd trans-file.
01 trans-rec.
05 trans-key pic 9(6).
05 trans-data pic x(9).
working-storage section.
01 flags.
05 trans-eof pic x value 'n'.
procedure division.
main.
open input trans-text
open output trans-file
perform until trans-eof is equal to 'y'
read trans-text
at end
move 'y' to trans-eof
not at end
move text-rec to trans-rec
display trans-key '|' trans-data '|'
write trans-rec
end-read
end-perform
close trans-text
close trans-file
stop run.
---- end of file ----

With these programs you can create test INDEXED files suitable for
running with my example join program. Then you can see how the program
that I posted works.

-- elliot

Jerome

2006-07-25, 7:55 am

epc8@juno.com a écrit :

>
> With these programs you can create test INDEXED files suitable for
> running with my example join program. Then you can see how the program
> that I posted works.
>
> -- elliot
>

Waouh, super !

My problem is that I work with data files created by another programmer
there is a long time and I haven't not enough documentation about his work.
Another problem is I'm not a specialist of cobol, but you noticed it :-)

Your codes, explanations and advices are very useful and precious (I
hope it is the good word in english). I'm going to work with the new
code that you posted.

Great thanks

Best regards

Jerome

epc8@juno.com

2006-07-25, 6:55 pm

Jerome wrote:
> epc8@juno.com a =E9crit :
>
>
> Waouh, super !
>
> My problem is that I work with data files created by another programmer
> there is a long time and I haven't not enough documentation about his wor=
k=2E
> Another problem is I'm not a specialist of cobol, but you noticed it :-)

I'm going to translate what you just said back into French and guess
that the original programmer is no longer there.

If your data files are indexed files then you may be able to convert
them back into sequential files without writing a COBOL program. Some
indexed file formats contain enough information about the file
structure to make this possible. I don't know if your compiler includes
this facility.

Here is a non COBOL example. dBase3 files are small databases. Each
contains header records which describe enough of the file's contents to
allow external utilities to extract data from the file, including field
by field access.

-- elliot

Jerome

2006-07-25, 6:55 pm

epc8@juno.com a écrit :
>
> If your data files are indexed files then you may be able to convert
> them back into sequential files without writing a COBOL program. Some
> indexed file formats contain enough information about the file
> structure to make this possible. I don't know if your compiler includes
> this facility.

> -- elliot
>

I'm reading the installation and user's guide of my compiler. 7
utilities are described. I'will s

this utilities (the original floppies).

In any case, you answered my questions in your preceding posts

Jerome

James J. Gavan

2006-07-25, 6:55 pm

epc8@juno.com wrote:
> James J. Gavan wrote:
>
>
>
> Thanks for the itinerary. Actually I'm not planning a trip to the UK in
> the near future. :-). My comment was a reference to the parting words
> of a well known fictional character. I'll leave that as a bit of a
> mystery but this itself will provide a small clue.
>
> -- elliot

> Never heard of Val Andrews. Still doing a google, "Elementary my dear
Watson".

2006-07-25, 6:55 pm

In article <YYsxg.232976$iF6.151395@pd7tw2no>,
James J. Gavan <jgavandeletethis@shaw.ca> wrote:
>epc8@juno.com wrote:

[snip]
[color=darkred]
>
> Never heard of Val Andrews. Still doing a google, "Elementary my dear
>Watson".

My memory is, admittedly, porous... but I seem to recall reading somewhere
that this is one of those 'great lines the character never really said',
along with Rick's (Humphrey Bogart) request to 'Play it again, Sam.'

DD

James J. Gavan

2006-07-25, 6:55 pm

Richard wrote:
> James J. Gavan wrote:
>
>
>
>
> And just the other side of Chichester along the same Stane Street is
> Fishbourne Villa which wasn't discovered until the 60s.
>
This is an interesting link to the Sussex sites, with photographs,
which I just checked :-

http://www.romansinsussex.co.uk/sussex/bignor.asp

We were Bognor-bound for a w

's holiday from Guildford - back in
Lambretta days. Probably would have whisked by in a car, but at a small
intersection saw a sign to our right (west): Bignor - Roman Villa.
Curious, we went and took a look.

I think what impressed me most was the baths at Bath, Somerset.
Originally built at ground level, you can now look over a bridge or
roadway to the baths below. Human garbage over the centuries has
increased the height of the road to some 20 - 30 feet.

Jimmy

epc8@juno.com

2006-07-25, 6:55 pm

docdwarf@panix.com wrote:
> In article <YYsxg.232976$iF6.151395@pd7tw2no>,
> James J. Gavan <jgavandeletethis@shaw.ca> wrote:
>
> [snip]
>
>
> My memory is, admittedly, porous... but I seem to recall reading somewhere
> that this is one of those 'great lines the character never really said',
> along with Rick's (Humphrey Bogart) request to 'Play it again, Sam.'
>
> DD

It certainly appeared on film. See
http://www.imdb.com/title/tt0037168/quotes

Anyway, there is a story that Watson once visited Holmes carrying a
large textbook on COBOL programming. "I'm puzzled," said Watson "It
says that some data items have subordinates. But what about those that
do not?" To which Holmes replies "Elementary, my dear Watson."

2006-07-25, 6:55 pm

In article <1153852133.104366.130690@m79g2000cwm.googlegroups.com>,
<epc8@juno.com> wrote:
>
>docdwarf@panix.com wrote:
>
>It certainly appeared on film. See
>http://www.imdb.com/title/tt0037168/quotes

Ummmm... it may be my imagination, or lack thereof, but on the page for
'Memorable Quotes from The Pearl of Death' (which is where the abovegiven
URL leads me) the word 'Elementary' does not appear; the word 'dear' is
found in 'My dear chap, I really must caution you against hitting
newspaper reporters in the teeth...'

It may be my memory, or lack thereof, but I do not recall a story by Sir
Arthur Conan Doyle entitled 'The Pearl of Death', either... but wasn't
there some bit of cinematic fluff about 'Sherlock Holmes versus... The
Nazis!'?

Ahhhh... http://imdb.com/title/tt0035317/ .

DD

2006-07-25, 6:55 pm

In article <6mtxg.235392$Mn5.123895@pd7tw3no>,
James J. Gavan <jgavandeletethis@shaw.ca> wrote:

[snip]

>I think what impressed me most was the baths at Bath, Somerset.
>Originally built at ground level, you can now look over a bridge or
>roadway to the baths below. Human garbage over the centuries has
>increased the height of the road to some 20 - 30 feet.

My response to this is 'a person standing on a tall pile of trash still
sees a farther horizon than one standing on a flat plain of pure, white
sand'.

DD

James J. Gavan

2006-07-25, 6:55 pm

docdwarf@panix.com wrote:
> In article <YYsxg.232976$iF6.151395@pd7tw2no>,
> James J. Gavan <jgavandeletethis@shaw.ca> wrote:
>
>
>
> [snip]
>
>
>
>
> My memory is, admittedly, porous... but I seem to recall reading somewhere
> that this is one of those 'great lines the character never really said',
> along with Rick's (Humphrey Bogart) request to 'Play it again, Sam.'

Ilsa (Ingrid Bergman) and Lazslo (Paul Henried) had already entered the
bar, unbeknown to Rick. Ilsa sees the pianist, Sam (Dooley Wilson) :-

Ilsa: Play it once, Sam. For old times' sake.
Sam: [lying] I don't know what you mean, Miss Ilsa.
Ilsa: Play it, Sam. Play "As Time Goes By."
Sam: [lying] Oh, I can't remember it, Miss Ilsa. I'm a little rusty on it.
Ilsa: I'll hum it for you. Da-dy-da-dy-da-dum, da-dy-da-dee-da-dum...
[Sam begins playing]
Ilsa: Sing it, Sam.
Sam: [singing] You must remember this / A kiss is still a kiss / A sigh
is just a sigh / The fundamental things apply / As time goes by. / And
when two lovers woo, / They still say, "I love you" / On that you can
rely / No matter what the future brings-...
Rick: [rushing up] Sam, I thought I told you never to play-...
[Sees Ilsa. Sam closes the piano and rolls it away]

Amazing how a two-bit corny movie became such an enjoyable and perennial
romantic classic.

Jimmy

Howard Brazee

2006-07-25, 6:55 pm

On Tue, 25 Jul 2006 18:12:16 +0000 (UTC), docdwarf@panix.com () wrote:

>
>My memory is, admittedly, porous... but I seem to recall reading somewhere
>that this is one of those 'great lines the character never really said',
>along with Rick's (Humphrey Bogart) request to 'Play it again, Sam.'

But Woody Allen said it again.

When you've had as many movies as Sherlock Homes has (IMDB has 257
matches on that character), we can have a large variance from the
original source. Basil Rathbone did not like having his mouth
hidden, so he started using anachronistic the calabash pipe that we
associate with Holmes today.

Posted Via Usenet.com Premium Usenet Newsgroup Services
----------------------------------------------------------
** SPEED ** RETENTION ** COMPLETION ** ANONYMITY **
----------------------------------------------------------
http://www.usenet.com

Howard Brazee

2006-07-25, 6:55 pm

On Tue, 25 Jul 2006 18:46:19 +0000 (UTC), docdwarf@panix.com () wrote:

>My response to this is 'a person standing on a tall pile of trash still
>sees a farther horizon than one standing on a flat plain of pure, white
>sand'.

Sort of like "I can see farther because I stand on the shoulders of
midgets".

Posted Via Usenet.com Premium Usenet Newsgroup Services
----------------------------------------------------------
** SPEED ** RETENTION ** COMPLETION ** ANONYMITY **
----------------------------------------------------------
http://www.usenet.com

2006-07-25, 6:55 pm

In article <AKtxg.235516$Mn5.32552@pd7tw3no>,
James J. Gavan <jgavandeletethis@shaw.ca> wrote:
>docdwarf@panix.com wrote:
>
>Ilsa (Ingrid Bergman) and Lazslo (Paul Henried) had already entered the
>bar, unbeknown to Rick. Ilsa sees the pianist, Sam (Dooley Wilson) :-
>
>Ilsa: Play it once, Sam. For old times' sake.
>Sam: [lying] I don't know what you mean, Miss Ilsa.
>Ilsa: Play it, Sam. Play "As Time Goes By."
>Sam: [lying] Oh, I can't remember it, Miss Ilsa. I'm a little rusty on it.
>Ilsa: I'll hum it for you. Da-dy-da-dy-da-dum, da-dy-da-dee-da-dum...
>[Sam begins playing]
>Ilsa: Sing it, Sam.
>Sam: [singing] You must remember this / A kiss is still a kiss / A sigh
>is just a sigh / The fundamental things apply / As time goes by. / And
>when two lovers woo, / They still say, "I love you" / On that you can
>rely / No matter what the future brings-...
>Rick: [rushing up] Sam, I thought I told you never to play-...
>[Sees Ilsa. Sam closes the piano and rolls it away]

From http://imdb.com/title/tt0034583/quotes:

--begin quoted text:

Rick: You know what I want to hear.
Sam: [lying] No, I don't.
Rick: You played it for her, you can play it for me!
Sam: [lying] Well, I don't think I can remember...
Rick: If she can stand it, I can! Play it!

--end quoted text

.... and nary a 'Play it again, Sam' in sight.

>
>Amazing how a two-bit corny movie became such an enjoyable and perennial
>romantic classic.

As I recall 'The Wizard of Oz' opened to lukewarm reviews... and now it
isn't in Kansas any more.

DD

2006-07-26, 3:55 am

In article <cqpcc2lskeh3u01gsvg6tc03j95vq7170o@4ax.com>,
Howard Brazee <howard@brazee.net> wrote:
>On Tue, 25 Jul 2006 18:12:16 +0000 (UTC), docdwarf@panix.com () wrote:
>
>
>But Woody Allen said it again.

Well, what else what might expect from a Brooklyn-boy?

>
>When you've had as many movies as Sherlock Homes has (IMDB has 257
>matches on that character), we can have a large variance from the
>original source. Basil Rathbone did not like having his mouth
>hidden, so he started using anachronistic the calabash pipe that we
>associate with Holmes today.

Well... I recall references to Holmes' smoking an 'oily, black clay'...
checking http://www.gutenberg.org/dirs/etext99/advsh12.txt (only thirteen
of Conan Doyle's stories... but a fitting number, nevertheless) there are
also references to a briar and a 'long cherrywood'. A bit of
web-sleuthing indicates that one Gary Schrier has recently published 'The
History of the Calabash Pipe' recently and it is being pitched as having
'plenty of direct Sherlockian interest'
(http://www.sherlock-holmes.org.uk/pdf/DM263.pdf)... but what's even more
interesting, to my eye, is this exchange from
http://forum.pipes.org/discus/messa...html?1139537539 which
mentions 'a gentleman writing the first book on the history of the
calabash but it is not due out until May', referring, I believe, to the
book that came out recently.

DD

2006-07-26, 3:55 am

In article <n5qcc2ljeh4d6po1e3ubcucerjl3cr73pd@4ax.com>,
Howard Brazee <howard@brazee.net> wrote:
>On Tue, 25 Jul 2006 18:46:19 +0000 (UTC), docdwarf@panix.com () wrote:
>
>
>Sort of like "I can see farther because I stand on the shoulders of
>midgets".

Sort of... excepts midgets are usually not defined as being 'tall', as the
pile of trash was... and midgets are human beings, composed of cells with,
for the most part, 23 pairs of chromosomes, and trash, quite frequently,
isn't...

.... other than that, *most* 'sort of'... kinda almost like, maybe.

DD

James J. Gavan

2006-07-27, 3:55 am

Jerome wrote:
> Hello,
>
> I've got a file "fgctar". When I read it sequentially, I obtain this
>
> KEY | VALUE
> ...
> 152001 | MyRecord1
> 153001 | MyRecord2
> 153153 | MyRecord3
> 154001 | MyRecord4
> 154154 | MyRecord5
> 155001 | MyRecord6
> 156001 | MyRecord7
> 156156 | MyRecord8
> ...
>
> I've got another file "fgcart", the primary key is composed by the first
> three digits of the primary key of my first file "fgctar". When I read
> it, i've got something like this :
>
> KEY | NAME
> ...
> 152 | Name1
> 153 | Name2
> 154 | Name3
> 155 | Name4
> 156 | Name5
> ...
>
> Then I read the second file "fgcart" sequentially and I search in the
> first file "fgctar" all records for which primary key of fgcart match
> primary key of fgctar and I've got this result
>
> KEY | NAME | VALUE
> ...
> 152001 | Name1 | MyRecord1
> 153001 | Name2 | MyRecord2
> 153153 | Name2 | MyRecord3
> 154001 | Name3 | MyRecord4
> 155001 | Name4 | MyRecord6
> 156001 | Name5 | MyRecord7
> 156156 | Name5 | MyRecord8
> ...
>
> So, the record "154154 MyRecord5" disappeared.
>
> Has someone an idea ?
>
> Jerome

Any luck so far - you don't seem to be getting anywhere fast on this
one. You are reading ISAMs sequentially.

Let's cheat a bit - hopefully you can do this with RM/COBOL. (It's the
technique Micro Focus provides to get RM/COBOL ISAM data into Micro
Focus format). Don't know your compiler but I am *assuming* RM/COBOL
still has two files for each ISAM, e.g. fgctar01.dat and fgctar01.idx.
Routine below ignores the "idx" or "inx" file.

Take your existing file definitions and in a *separate* program do this :-

INPUT-OUTPUT SECTION.
FILE-CONTROL.
SELECT FGCTAR ASSIGN TO RANDOM "fgctar01"
ORGANIZATION INDEXED
ACCESS DYNAMIC
RECORD KEY GCTAR-CLE
STATUS ST-GCTAR.
SELECT FGCART ASSIGN TO RANDOM "fgcart01"
ORGANIZATION IS INDEXED
ACCESS IS DYNAMIC
RECORD KEY IS GCART-CLE
ALTERNATE RECORD KEY IS GCART-NOM WITH DUPLICATES
FILE STATUS IS FichierStatus.
SELECT OUTART ASSIGN TO RANDOM "out-ess.txt"
ORGANIZATION IS LINE SEQUENTIAL.

change above :-

*> Select Data-File
*> assign Data-filename
*> organization SEQUENTIAL
*> access sequential
*> file status ws-FileStatus.

SELECT FGCTAR ASSIGN TO "fgctar01"
ORGANIZATION SEQUENTIAL
ACCESS SEQUENTIAL
FILE-STATUS FGCTAR-Status
SELECT FGCART ASSIGN TO "fgcart01"
ORGANIZATION SEQUENTIAL
ACCESS SEQUENTIAL
FILE-STATUS FGCART-Status

Leave your record formats unchanged.

Separately for the two files above :-

1) Open as INPUT
2) READ NEXT until EOF
3) WRITE each record read to a temporary file LINE SEQUENTIAL
4) Then CLOSE inputs and outputs
5) As you step through each file, for each "action" ( OPEN, READ, WRITE
or CLOSE) ensure you check the files-status code :-

IF FILE-STATUS <> ?????
PERFORM file-error

Now when you bring up the two LINE SEQUENTIALS in a text editor do you
get *exactly* the same set of records per file you showed in your very
first message ?

Record sequencing wont necessarily be the same :-

1) READing an ISAM Sequentially (READ NEXT) will give records in
ascending sequence by key.

2) Re-classifying ISAM files as SEQUENTIAL will give you records in the
sequence that they were added to the file.

Have you got just the one set of test data; if possible try it against
another test set and see if again you "lose" records.

One other alternative - use the RM/COBOL ISAM Rebuild routines. I have a
very old copy of RM and it had the nasty habit of silently dropping
problematic records. Hopefully their routines are more slick now.

Whichever route you go, do a record count per file.

Above doesn't solve your problem, if it works - but it should help point
you in the right direction with your "real" program.

Jimmy

Jerome

2006-07-27, 3:55 am

James J. Gavan a écrit :

> Above doesn't solve your problem, if it works - but it should help point
> you in the right direction with your "real" program.
>
>
> Jimmy

Thank you for your help,

I haven't index file, my compiler is a RM/COBOL-85 Liant Software v5.36.

My problem has been already solved but I will read your code attentively
because it's a luck for me to learn cobol. So, I still thank you for
your assistance.

Best regards

Jerome

2006-07-27, 3:55 am

In article <ea9tkv$ik5$1@s1.news.oleane.net>,
Jerome <j.aubert1@free.fr> wrote:
>James J. Gavan a écrit :
>
>
>Thank you for your help,
>
>I haven't index file, my compiler is a RM/COBOL-85 Liant Software v5.36.
>
>My problem has been already solved but I will read your code attentively
>because it's a luck for me to learn cobol. So, I still thank you for
>your assistance.

Congratulations on getting your problem solved... can you post the code
that did it?

DD

Jerome

2006-07-27, 3:55 am

docdwarf@panix.com a écrit :

> Congratulations on getting your problem solved... can you post the code
> that did it?
>
> DD
>
Yes, of course, in fact it is the code of Eliott that solved my problem :

IDENTIFICATION DIVISION.
PROGRAM-ID. JOIN.
ENVIRONMENT DIVISION.
CONFIGURATION SECTION.
SOURCE-COMPUTER. IBM-PC.
OBJECT-COMPUTER. IBM-PC.
INPUT-OUTPUT SECTION.
FILE-CONTROL.
SELECT NAME-FILE ASSIGN TO DISK "fgcart01"
ORGANIZATION IS INDEXED
RECORD KEY IS GCART-NUM
ALTERNATE RECORD KEY IS GCART-NOM WITH DUPLICATES.
SELECT TRANS-FILE ASSIGN TO DISK "fgctar01"
ORGANIZATION IS INDEXED
RECORD KEY IS GCTAR-CLE.
SELECT OUT-FILE ASSIGN TO DISK "out.txt"
ORGANIZATION IS LINE SEQUENTIAL.
DATA DIVISION.
FILE SECTION.
COPY "FGCART.CPY".
COPY "FGCTAR.CPY".
FD OUT-FILE.
01 OUT-REC.
05 OUT-KEY PIC X(6).
05 FILLER PIC X.
05 OUT-NAME PIC X(5).
05 FILLER PIC X.
05 OUT-DATA PIC 9(6).9(2).
WORKING-STORAGE SECTION.
01 FLAGS.
05 NAME-EOF PIC X VALUE 'n'.
05 TRANS-EOF PIC X.
05 KEY-NOT-FOUND PIC X.
PROCEDURE DIVISION.
MAIN.
OPEN INPUT NAME-FILE
OPEN INPUT TRANS-FILE
OPEN OUTPUT OUT-FILE
MOVE ALL '|' TO OUT-REC
PERFORM UNTIL NAME-EOF IS EQUAL TO 'y'
READ NAME-FILE
AT END
MOVE 'y' TO NAME-EOF
NOT AT END
PERFORM GET-MATCH
END-READ
END-PERFORM
CLOSE NAME-FILE
CLOSE TRANS-FILE
CLOSE OUT-FILE
STOP RUN.
GET-MATCH.

MOVE GCART-NUM TO GCTAR-ART

START TRANS-FILE KEY NOT < GCTAR-ART
INVALID KEY
MOVE 'y' TO KEY-NOT-FOUND
NOT INVALID KEY
MOVE 'n' TO TRANS-EOF
MOVE 'n' TO KEY-NOT-FOUND
PERFORM UNTIL TRANS-EOF IS EQUAL TO 'y'
OR KEY-NOT-FOUND IS EQUAL TO 'y'

READ TRANS-FILE NEXT RECORD
AT END
MOVE 'y' TO TRANS-EOF
NOT AT END
IF GCART-NUM IS NOT EQUAL TO GCTAR-ART
MOVE 'y' TO KEY-NOT-FOUND
ELSE
MOVE GCTAR-CLE TO OUT-KEY
MOVE GCART-NOM TO OUT-NAME
MOVE GCTAR-PUHT TO OUT-DATA
WRITE OUT-REC
END-IF
END-READ
END-PERFORM
END-START.

Michael Wojcik

2006-07-27, 9:55 pm

In article <ea5q0h$bbn$1@reader2.panix.com>, docdwarf@panix.com () writes:
>
> As I recall 'The Wizard of Oz' opened to lukewarm reviews... and now it
> isn't in Kansas any more.

Some of us still think it's a feeble distortion of a fine novel.
(Salman Rushdie disagrees, but - much as I like Rushdie - he's wrong.)

The film turns Baum's self-confident, capable Dorothy into a
trembling, nearly helpless wimp, thereby completely missing the point
of the narrative (which is Yet Another version of how the vital rural
underclass will rehabilitate the effete urban aristocracy), and
ruining one of the best female characters in children's literature;
fills Oz with saccharine clowning and patronizing moralizing
(particularly in its version of the Wizard's gifts), thereby robbing
the narrative of its connection to the original European marchen
tradition; and cheats at the end, thereby robbing itself of any claim
to decent plotting.

It has some technical merit, particularly in the famous transition
from sepia-tone (the Kansas segments are not actually black-and-white)
to full (indeed over-full) color, and some good scenes, such as the
Wicked Witch's Emerald City flyover. But that's not enough to make
it a good film, much less a great one.

Not that I care, or anything.

--
Michael Wojcik michael.wojcik@microfocus.com

Some there are, brave, high-souled fellows, who could borrow the world to
play at ball, and never feel the responsibility, whereas others are uneasy
and not themselves with a single shilling that does not belong to them.
-- Arthur Ransome

2006-07-27, 9:55 pm

In article <eaam2n022br@news1.newsguy.com>,
Michael Wojcik <mwojcik@newsguy.com> wrote:
>
>In article <ea5q0h$bbn$1@reader2.panix.com>, docdwarf@panix.com () writes:
>
>Some of us still think it's a feeble distortion of a fine novel.

Welcome to Hollywood, Mr Wojcik... such things happen there, at times...
or so I've been told. Feeble or not the film has managed to work its way
into American popular culture, to the point where an intoned 'I'll get
you... and your little dog, too!' seems to be recognised across the length
and breadth of This Great Nation as an embodiment of 'wickedness'.

[snip]

>Not that I care, or anything.

You don't care? It may not be apathy, Mr Wojcik - oh, and you left out
the part about the Silver Standard, too - it may be simple... fear.
Perhaps things might be different were you to have a bit more bravery...
after all...

What makes a king out of a slave? Courage!

What makes the flag on the mast to wave? Courage!

What makes the elephant charge his tusk, in the misty mist or the dusty
dust? What makes the muskrat guard his musk? Courage!

What makes the Sphinx the seventh wonder? Courage!

What makes the dawn come up like thunder? Courage!

What makes the Hottentot so hot?

What puts the 'ape' in ap-ri-cot?

What do they got that you ain't got?

DD

Sponsored Links