Re: Two string instruction questions

[M. Ray Mullins]

Very late to the game, but could you use one of the new vector string 
instructions? Possibly a combination of VECTOR FIND ELEMENT EQUAL and 
VECTOR STRING RANGE COMPARE. I've done nothing with this (yet), so I 
can't provide any answers or advice.


You would have to be z13 or better to open.

On 2018-03-14 08:51, Charles Mills wrote:

1.   Is there a machine instruction that will find one string within
another? That given "Now is the time" and "is" would find the "is" and
return a pointer to it? A machine instruction analog of Rexx POS?

2.   Searching the PoOp for such an instruction led me to CUSE. It does
not seem that CUSE could be used for this - is that correct? If I am reading
CUSE correctly, then given "Now is the time", "All is well" and 2 or 3 would
return the position of "is". Is my reading correct? What would that be good
for? What would be a reasonable real-world use?

  


Thanks,

  


Charles

  


--
M. Ray Mullins
Roseville, CA, USA
http://www.catherdersoftware.com/
http://www.z390.org/

German is essentially a form of assembly language consisting entirely of far 
calls heavily accented with throaty guttural sounds. ---ilvi
French is essentially German with messed-up pronunciation and spelling.  
--Robert B Wilson
English is essentially French converted to 7-bit ASCII.  ---Christophe Pierret 
[for Alain LaBonté]

Re: Two string instruction questions

[Charles Mills]

Lol that I what I typed. Your email joined the lines. 
Find "foo [new line] bar"

CharlesSent from a mobile; please excuse the brevity.
 Original message From: Paul Gilmartin 
<0014e0e4a59b-dmarc-requ...@listserv.uga.edu> Date: 3/15/18  10:18 AM  
(GMT-08:00) To: ASSEMBLER-LIST@LISTSERV.UGA.EDU Subject: Re: Two string 
instruction questions 
On 2018-03-15, at 11:02:17, Charles Mills wrote:

>  Your points are good but FWIW ; is a command separator and at a "higher 
>level" than quoted string parsing.
> Find "foo;bar" is for better or worse exactly the same as
> Find "foobar"
>  
I believe not.  Find "foo;bar" is the same as
    Find "foo
    bar"

... both of which are reported as syntax errors.

Very bad design.  A "higher level" executive has no business performing
a bottom-up parse of commands it passes to a lower level processor.  At
least it should support an escape convention, such as:
    Find "foo\;bar"
... with the "\" protecting the ";".  The command passed to the lower level
would then simply be:
    Find "foo;bar"

But that couldn't be done with a simple TRT.  Oh, my gosh!  Think of the
performance implications of doing it right!

Feedback from a parser to its lexical analyzer is generally deemed
harmful (but consider PL/I!)  OTOH, the lexical analyzer should not
impose limitations on the parser's syntax.

> printf( "foo;bar\n" );  /* in C  */
>  
C's lexical analyzer knows enough to recognize that the second ";", but
not the first, is a token separator.

-- gil

Re: Two string instruction questions

[Paul Gilmartin]

On 2018-03-15, at 11:02:17, Charles Mills wrote:

>  Your points are good but FWIW ; is a command separator and at a "higher 
> level" than quoted string parsing.
> Find "foo;bar" is for better or worse exactly the same as
> Find "foobar"
>  
I believe not.  Find "foo;bar" is the same as
Find "foo
bar"

... both of which are reported as syntax errors.

Very bad design.  A "higher level" executive has no business performing
a bottom-up parse of commands it passes to a lower level processor.  At
least it should support an escape convention, such as:
Find "foo\;bar"
... with the "\" protecting the ";".  The command passed to the lower level
would then simply be:
Find "foo;bar"

But that couldn't be done with a simple TRT.  Oh, my gosh!  Think of the
performance implications of doing it right!

Feedback from a parser to its lexical analyzer is generally deemed
harmful (but consider PL/I!)  OTOH, the lexical analyzer should not
impose limitations on the parser's syntax.

> printf( "foo;bar\n" );  /* in C  */
>  
C's lexical analyzer knows enough to recognize that the second ";", but
not the first, is a token separator.

-- gil

Re: Two string instruction questions

[Paul Gilmartin]

On 2018-03-15, at 01:21:23, robi...@dodo.com.au wrote:

> Use a TR as well. Works wonders.
>  
Maybe

> - Original Message -
> Sent:Wed, 14 Mar 2018 21:06:36 -0600
> 
> Ok. Now make it case-insensitive, which is a common convention. That
> can be done with Boyer-Moore.
>  
When Boyer-Moore was first announced (in CACM?) about 40 years ago,
it fascinated me.  I coded an enhanced variant such that majuscule
characters in the pattern would match only majuscule characters in
the subject; minuscule characters in the pattern would match either
case.  I thought it might be useful that "Robin" would match either
"Robin" or "ROBIN", but not "robin".  Don't know how useful it was,
but it was fun to code.  I don't believe I could have done it with
"a TR as well".

-- gil

Re: Two string instruction questions

[Charles Mills]

 Your points are good but FWIW ; is a command separator and at a "higher level" 
than quoted string parsing.
Find "foo;bar" is for better or worse exactly the same as
Find "foobar"
CharlesSent from a mobile; please excuse the brevity.
 Original message From: Paul Gilmartin 
<0014e0e4a59b-dmarc-requ...@listserv.uga.edu> Date: 3/15/18  9:45 AM  
(GMT-08:00) To: ASSEMBLER-LIST@LISTSERV.UGA.EDU Subject: Re: Two string 
instruction questions 
On 2018-03-15, at 08:27:57, Charles Mills wrote:.
> 
> 2. TRT is a single op code but that does not make it "fast." 
>  
The peculiar evil of TRT is that "Everything looks like a nail."
You need only hammer on it enough with TRT.

In ISPF, the command:
    FIND "foo;bar"

... fails for "Unterminated delimited string".  But:
    SAY "foo;bar";  /* in Rexx  */
    echo "foo;bar"; # in POSIX shell
    printf( "foo;bar\n" );  /* in C  */

In the all 4 languages ";" is a command separator.  Only ISPF fails
to understand that in a quoted string it should not have that special
meaning.  I suspect that ISPF does a misguided bottom-up search for
";" with TRT and deems it an unconditional command separator.

(Yes, I know that I can choose to sacrifice some other character as a
command separator, and avoid that in strings.  At times I've used "¾".
But why should such mickeymouse be necessary?)

(Yes, I know that I can specify my search target in hex.  Ugh!  It
would be better if I could use sporadic hex escapes in otherwise
plain text.)

-- gil

Re: Two string instruction questions

[Paul Gilmartin]

On 2018-03-15, at 08:27:57, Charles Mills wrote:.
> 
> 2. TRT is a single op code but that does not make it "fast." 
>  
The peculiar evil of TRT is that "Everything looks like a nail."
You need only hammer on it enough with TRT.

In ISPF, the command:
FIND "foo;bar"

... fails for "Unterminated delimited string".  But:
SAY "foo;bar";  /* in Rexx  */
echo "foo;bar"; # in POSIX shell
printf( "foo;bar\n" );  /* in C  */

In the all 4 languages ";" is a command separator.  Only ISPF fails
to understand that in a quoted string it should not have that special
meaning.  I suspect that ISPF does a misguided bottom-up search for
";" with TRT and deems it an unconditional command separator.

(Yes, I know that I can choose to sacrifice some other character as a
command separator, and avoid that in strings.  At times I've used "¾".
But why should such mickeymouse be necessary?)

(Yes, I know that I can specify my search target in hex.  Ugh!  It
would be better if I could use sporadic hex escapes in otherwise
plain text.)

-- gil

Re: Two string instruction questions

[Alan Atkinson]

Yeah  - mea culpa. It's been years since I looked at that code. We escaped our 
search strings and then looked for the opening character.
You might still do better to do an SRST / CLC loop.


On 3/15/18, 11:04 AM, "IBM Mainframe Assembler List on behalf of Charles 
Mills" <ASSEMBLER-LIST@LISTSERV.UGA.EDU on behalf of charl...@mcn.org> wrote:

Nothing, so far as I know. :-)

But it finds only a single character and not a substring in a string; a 
needle in a haystack, to use the Rexx idiom.

For finding a single character in a string, SRST is almost undoubtedly 
faster than TRT. For finding either of two characters, hard to say, but I am 
going to guess SRST is still faster.

Charles


-Original Message-
From: IBM Mainframe Assembler List [mailto:ASSEMBLER-LIST@LISTSERV.UGA.EDU] 
On Behalf Of Alan Atkinson
Sent: Thursday, March 15, 2018 7:32 AM
To: ASSEMBLER-LIST@LISTSERV.UGA.EDU
    Subject: Re: Two string instruction questions

Am I missing something? What's wrong with SRST?


On 3/15/18, 10:28 AM, "IBM Mainframe Assembler List on behalf of Charles 
Mills" <ASSEMBLER-LIST@LISTSERV.UGA.EDU on behalf of charl...@mcn.org> wrote:

Traditional, but not anywhere near optimally quick.

1. Read the Wikipedia article @Gil linked to. There are faster ways 
than the
obvious way.

2. TRT is a single op code but that does not make it "fast." Picture an
assembler loop to do what TRT does. Not very fast, right? Now picture 
the
same loop in millicode. Still not very fast. That's TRT. Millicode is 
not
magic; it's more or less machine code running under the covers. SRST is
probably considerably faster (although still not magically fast, and not
case-insensitive as TRT potentially is). On the other hand, "it 
depends";
and cache misses are what is slow -- instructions themselves hardly 
matter.

Charles


-Original Message-
From: IBM Mainframe Assembler List 
[mailto:ASSEMBLER-LIST@LISTSERV.UGA.EDU]
On Behalf Of Robin Vowels
Sent: Wednesday, March 14, 2018 7:32 PM
To: ASSEMBLER-LIST@LISTSERV.UGA.EDU
    Subject: Re: Two string instruction questions

From: "Charles Mills" <charl...@mcn.org>
Sent: Thursday, March 15, 2018 2:51 AM


> 1.   Is there a machine instruction that will find one string 
within
> another? That given "Now is the time" and "is" would find the "is" 
and 
> return a pointer to it? A machine instruction analog of Rexx POS?
> 
> 2.   Searching the PoOp for such an instruction led me to CUSE. It
does
> not seem that CUSE could be used for this - is that correct? If I am 
> reading CUSE correctly, then given "Now is the time", "All is well" 
> and 2 or 3 would return the position of "is". Is my reading correct? 
> What would that be good for? What would be a reasonable real-world 
use?

The traditional TRT to search for the first letter, followed by CLC for 
the
word (for which the search can commence at the second letter, because 
the
first letter has already been found), will likely be the quickest.

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

Re: Two string instruction questions

[Charles Mills]

Nothing, so far as I know. :-)

But it finds only a single character and not a substring in a string; a needle 
in a haystack, to use the Rexx idiom.

For finding a single character in a string, SRST is almost undoubtedly faster 
than TRT. For finding either of two characters, hard to say, but I am going to 
guess SRST is still faster.

Charles


-Original Message-
From: IBM Mainframe Assembler List [mailto:ASSEMBLER-LIST@LISTSERV.UGA.EDU] On 
Behalf Of Alan Atkinson
Sent: Thursday, March 15, 2018 7:32 AM
To: ASSEMBLER-LIST@LISTSERV.UGA.EDU
Subject: Re: Two string instruction questions

Am I missing something? What's wrong with SRST?


On 3/15/18, 10:28 AM, "IBM Mainframe Assembler List on behalf of Charles 
Mills" <ASSEMBLER-LIST@LISTSERV.UGA.EDU on behalf of charl...@mcn.org> wrote:

Traditional, but not anywhere near optimally quick.

1. Read the Wikipedia article @Gil linked to. There are faster ways than the
obvious way.

2. TRT is a single op code but that does not make it "fast." Picture an
assembler loop to do what TRT does. Not very fast, right? Now picture the
same loop in millicode. Still not very fast. That's TRT. Millicode is not
magic; it's more or less machine code running under the covers. SRST is
probably considerably faster (although still not magically fast, and not
case-insensitive as TRT potentially is). On the other hand, "it depends";
and cache misses are what is slow -- instructions themselves hardly matter.

Charles


-Original Message-
From: IBM Mainframe Assembler List [mailto:ASSEMBLER-LIST@LISTSERV.UGA.EDU]
On Behalf Of Robin Vowels
Sent: Wednesday, March 14, 2018 7:32 PM
To: ASSEMBLER-LIST@LISTSERV.UGA.EDU
    Subject: Re: Two string instruction questions

From: "Charles Mills" <charl...@mcn.org>
Sent: Thursday, March 15, 2018 2:51 AM


> 1.   Is there a machine instruction that will find one string within
> another? That given "Now is the time" and "is" would find the "is" and 
> return a pointer to it? A machine instruction analog of Rexx POS?
> 
> 2.   Searching the PoOp for such an instruction led me to CUSE. It
does
> not seem that CUSE could be used for this - is that correct? If I am 
> reading CUSE correctly, then given "Now is the time", "All is well" 
> and 2 or 3 would return the position of "is". Is my reading correct? 
> What would that be good for? What would be a reasonable real-world use?

The traditional TRT to search for the first letter, followed by CLC for the
word (for which the search can commence at the second letter, because the
first letter has already been found), will likely be the quickest.

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

Re: Two string instruction questions

[Alan Atkinson]

Am I missing something? What's wrong with SRST?


On 3/15/18, 10:28 AM, "IBM Mainframe Assembler List on behalf of Charles 
Mills" <ASSEMBLER-LIST@LISTSERV.UGA.EDU on behalf of charl...@mcn.org> wrote:

Traditional, but not anywhere near optimally quick.

1. Read the Wikipedia article @Gil linked to. There are faster ways than the
obvious way.

2. TRT is a single op code but that does not make it "fast." Picture an
assembler loop to do what TRT does. Not very fast, right? Now picture the
same loop in millicode. Still not very fast. That's TRT. Millicode is not
magic; it's more or less machine code running under the covers. SRST is
probably considerably faster (although still not magically fast, and not
case-insensitive as TRT potentially is). On the other hand, "it depends";
and cache misses are what is slow -- instructions themselves hardly matter.

Charles


-Original Message-
From: IBM Mainframe Assembler List [mailto:ASSEMBLER-LIST@LISTSERV.UGA.EDU]
On Behalf Of Robin Vowels
Sent: Wednesday, March 14, 2018 7:32 PM
To: ASSEMBLER-LIST@LISTSERV.UGA.EDU
    Subject: Re: Two string instruction questions

From: "Charles Mills" <charl...@mcn.org>
Sent: Thursday, March 15, 2018 2:51 AM


> 1.   Is there a machine instruction that will find one string within
> another? That given "Now is the time" and "is" would find the "is" and 
> return a pointer to it? A machine instruction analog of Rexx POS?
> 
> 2.   Searching the PoOp for such an instruction led me to CUSE. It
does
> not seem that CUSE could be used for this - is that correct? If I am 
> reading CUSE correctly, then given "Now is the time", "All is well" 
> and 2 or 3 would return the position of "is". Is my reading correct? 
> What would that be good for? What would be a reasonable real-world use?

The traditional TRT to search for the first letter, followed by CLC for the
word (for which the search can commence at the second letter, because the
first letter has already been found), will likely be the quickest.

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

Re: Two string instruction questions

[Charles Mills]

Works wonders for slowing things down! TR modifies storage, one of the most 
expensive acts of all.

Charles


-Original Message-
From: IBM Mainframe Assembler List [mailto:ASSEMBLER-LIST@LISTSERV.UGA.EDU] On 
Behalf Of robi...@dodo.com.au
Sent: Thursday, March 15, 2018 12:21 AM
To: ASSEMBLER-LIST@LISTSERV.UGA.EDU
Subject: Re: Two string instruction questions

Use a TR as well. Works wonders.

Re: Two string instruction questions

[Charles Mills]

Traditional, but not anywhere near optimally quick.

1. Read the Wikipedia article @Gil linked to. There are faster ways than the
obvious way.

2. TRT is a single op code but that does not make it "fast." Picture an
assembler loop to do what TRT does. Not very fast, right? Now picture the
same loop in millicode. Still not very fast. That's TRT. Millicode is not
magic; it's more or less machine code running under the covers. SRST is
probably considerably faster (although still not magically fast, and not
case-insensitive as TRT potentially is). On the other hand, "it depends";
and cache misses are what is slow -- instructions themselves hardly matter.

Charles


-Original Message-
From: IBM Mainframe Assembler List [mailto:ASSEMBLER-LIST@LISTSERV.UGA.EDU]
On Behalf Of Robin Vowels
Sent: Wednesday, March 14, 2018 7:32 PM
To: ASSEMBLER-LIST@LISTSERV.UGA.EDU
Subject: Re: Two string instruction questions

From: "Charles Mills" <charl...@mcn.org>
Sent: Thursday, March 15, 2018 2:51 AM


> 1.   Is there a machine instruction that will find one string within
> another? That given "Now is the time" and "is" would find the "is" and 
> return a pointer to it? A machine instruction analog of Rexx POS?
> 
> 2.   Searching the PoOp for such an instruction led me to CUSE. It
does
> not seem that CUSE could be used for this - is that correct? If I am 
> reading CUSE correctly, then given "Now is the time", "All is well" 
> and 2 or 3 would return the position of "is". Is my reading correct? 
> What would that be good for? What would be a reasonable real-world use?

The traditional TRT to search for the first letter, followed by CLC for the
word (for which the search can commence at the second letter, because the
first letter has already been found), will likely be the quickest.

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

Re: Two string instruction questions

[robin51]

Use a TR as well. Works wonders.

- Original Message -
From: "IBM Mainframe Assembler List" 
To:
Cc:
Sent:Wed, 14 Mar 2018 21:06:36 -0600
Subject:Re: Two string instruction questions

 On 2018-03-14, at 20:32:18, Robin Vowels wrote:

 > From: "Charles Mills" 
 > Sent: Thursday, March 15, 2018 2:51 AM
 > 
 >> 1. Is there a machine instruction that will find one string within
 >> another? That given "Now is the time" and "is" would find the "is"
and
 >> return a pointer to it? A machine instruction analog of Rexx POS?
 >> 2. Searching the PoOp for such an instruction led me to CUSE. It
does
 >> not seem that CUSE could be used for this - is that correct? If I
am reading
 >> CUSE correctly, then given "Now is the time", "All is well" and 2
or 3 would
 >> return the position of "is". Is my reading correct? What would
that be good
 >> for? What would be a reasonable real-world use?
 > 
 > The traditional TRT to search for the first letter,
 > followed by CLC for the word (for which the search can commence
 > at the second letter, because the first letter has already been
found),
 > will likely be the quickest.
 > 
 Ok. Now make it case-insensitive, which is a common convention. That
 can be done with Boyer-Moore.

 -- gil

Re: Two string instruction questions

[Paul Gilmartin]

On 2018-03-14, at 20:32:18, Robin Vowels wrote:

> From: "Charles Mills" 
> Sent: Thursday, March 15, 2018 2:51 AM
> 
>> 1.   Is there a machine instruction that will find one string within
>> another? That given "Now is the time" and "is" would find the "is" and
>> return a pointer to it? A machine instruction analog of Rexx POS?
>> 2.   Searching the PoOp for such an instruction led me to CUSE. It does
>> not seem that CUSE could be used for this - is that correct? If I am reading
>> CUSE correctly, then given "Now is the time", "All is well" and 2 or 3 would
>> return the position of "is". Is my reading correct? What would that be good
>> for? What would be a reasonable real-world use?
> 
> The traditional TRT to search for the first letter,
> followed by CLC for the word (for which the search can commence
> at the second letter, because the first letter has already been found),
> will likely be the quickest.
>  
Ok.  Now make it case-insensitive, which is a common convention.  That
can be done with Boyer-Moore.

-- gil

Re: Two string instruction questions

[Charles Mills]

Thanks @Jonathan.

Charles


-Original Message-
From: IBM Mainframe Assembler List [mailto:ASSEMBLER-LIST@LISTSERV.UGA.EDU]
On Behalf Of Jonathan Scott
Sent: Wednesday, March 14, 2018 9:56 AM
To: ASSEMBLER-LIST@LISTSERV.UGA.EDU
Subject: Re: Two string instruction questions

Ref:  Your note of 14 March 2018, 08:51:22 -0700
 
Charles Mills wrote:
> 1.   Is there a machine instruction that will find one string within
> another? That given "Now is the time" and "is" would find the "is" and 
> return a pointer to it? A machine instruction analog of Rexx POS?
 
I'm not aware of any such instruction.  You can use SRST to find a single
character, and that can be used as part of a search algorithm.
 
> 2.   Searching the PoOp for such an instruction led me to CUSE. It
does
> not seem that CUSE could be used for this - is that correct? If I am 
> reading CUSE correctly, then given "Now is the time", "All is well" 
> and 2 or 3 would return the position of "is". Is my reading correct? 
> What would that be good for? What would be a reasonable real-world use?
 
CUSE is very useful for creating an efficient representation of the
differences between two versions of a record, for example to create a log
record showing which bytes changed. It finds the end of a block of changed
data, ignoring unchanged sequences which are too small to be worth
representing as a separate section. To skip to the end of the unchanged
data, CLCL or CLCLE can be used.
 
Jonathan Scott
HLASM, IBM Hursley, UK

Re: Two string instruction questions

[MELVYN MALTZ]

Hi Peter,

Partially right
In a VSAM file (KSDS only I think) you can request the Index to have 
compressed keys

Only worth while if the keys were large (max length is 127 bytes)

The Index entry would be compared to the previous and both the beginning and 
ending key bytes would be compared and whatever was the same got a length 
instead of the bytes, the final index entry looks horrible, but it saved 
valuable disk space back in the time when dinosaurs (like me) ruled the 
Earth :-) The RBA would not have been there for a KSDS


Have I got the very last "VSAM Logic" Manual on the planet ?

Melvyn Maltz

- Original Message - 
From: "Farley, Peter x23353" <peter.far...@broadridge.com>

To: <ASSEMBLER-LIST@LISTSERV.UGA.EDU>
Sent: Wednesday, March 14, 2018 10:18 PM
Subject: Re: Two string instruction questions


I think I read somewhere that is what keyed VSAM Index records are, aren't 
they?  A count of equal key bytes and then the remaining non-equal bytes, 
followed by the RBA in the data component?  Or is that a fib I was told?


Peter

-Original Message-
From: IBM Mainframe Assembler List [mailto:ASSEMBLER-LIST@LISTSERV.UGA.EDU] 
On Behalf Of Charles Mills

Sent: Wednesday, March 14, 2018 5:49 PM
To: ASSEMBLER-LIST@LISTSERV.UGA.EDU
Subject: Re: Two string instruction questions

That's interesting. Thanks!

I did think of what CUSE would be perfect for: what I know as "vertical 
compression" but Google does not seem to know the term. Think of standard 
run length compression as "horizontal." Picture something like that, but 
where a code indicates "the next 'n' bytes of this record are identical to 
the bytes in the same position in the previous record." Only works for 
sequential files, because you need the previous record to decode this 
records. But works well where there are a lot of repeating fields.


--


This message and any attachments are intended only for the use of the 
addressee and may contain information that is privileged and confidential. 
If the reader of the message is not the intended recipient or an authorized 
representative of the intended recipient, you are hereby notified that any 
dissemination of this communication is strictly prohibited. If you have 
received this communication in error, please notify us immediately by e-mail 
and delete the message and any attachments from your system. 

Re: Two string instruction questions

[Farley, Peter x23353]

I think I read somewhere that is what keyed VSAM Index records are, aren't 
they?  A count of equal key bytes and then the remaining non-equal bytes, 
followed by the RBA in the data component?  Or is that a fib I was told?

Peter

-Original Message-
From: IBM Mainframe Assembler List [mailto:ASSEMBLER-LIST@LISTSERV.UGA.EDU] On 
Behalf Of Charles Mills
Sent: Wednesday, March 14, 2018 5:49 PM
To: ASSEMBLER-LIST@LISTSERV.UGA.EDU
Subject: Re: Two string instruction questions

That's interesting. Thanks!

I did think of what CUSE would be perfect for: what I know as "vertical 
compression" but Google does not seem to know the term. Think of standard run 
length compression as "horizontal." Picture something like that, but where a 
code indicates "the next 'n' bytes of this record are identical to the bytes in 
the same position in the previous record." Only works for sequential files, 
because you need the previous record to decode this records. But works well 
where there are a lot of repeating fields.

--


This message and any attachments are intended only for the use of the addressee 
and may contain information that is privileged and confidential. If the reader 
of the message is not the intended recipient or an authorized representative of 
the intended recipient, you are hereby notified that any dissemination of this 
communication is strictly prohibited. If you have received this communication 
in error, please notify us immediately by e-mail and delete the message and any 
attachments from your system.

Re: Two string instruction questions

[Charles Mills]

That's interesting. Thanks!

I did think of what CUSE would be perfect for: what I know as "vertical
compression" but Google does not seem to know the term. Think of standard
run length compression as "horizontal." Picture something like that, but
where a code indicates "the next 'n' bytes of this record are identical to
the bytes in the same position in the previous record." Only works for
sequential files, because you need the previous record to decode this
records. But works well where there are a lot of repeating fields.

Charles


-Original Message-
From: IBM Mainframe Assembler List [mailto:ASSEMBLER-LIST@LISTSERV.UGA.EDU]
On Behalf Of Paul Gilmartin
Sent: Wednesday, March 14, 2018 10:56 AM
To: ASSEMBLER-LIST@LISTSERV.UGA.EDU
Subject: Re: Two string instruction questions

On 2018-03-14, at 09:51:22, Charles Mills wrote:

> 1.   Is there a machine instruction that will find one string within
> another? That given "Now is the time" and "is" would find the "is" and 
> return a pointer to it? A machine instruction analog of Rexx POS?
> 
> 2.   Searching the PoOp for such an instruction led me to CUSE. It
does
> not seem that CUSE could be used for this - is that correct? If I am 
> reading CUSE correctly, then given "Now is the time", "All is well" 
> and 2 or 3 would return the position of "is". Is my reading correct? 
> What would that be good for? What would be a reasonable real-world use?
>  
BTW, the classic technique is Boyer-Moore.  But see:
https://en.wikipedia.org/wiki/String_searching_algorithm

This might easily be done in microcode.  And I once coded a case-insensitive
variant of Boyer-Moore.

-- gil

Re: Two string instruction questions

[Paul Gilmartin]

On 2018-03-14, at 09:51:22, Charles Mills wrote:

> 1.   Is there a machine instruction that will find one string within
> another? That given "Now is the time" and "is" would find the "is" and
> return a pointer to it? A machine instruction analog of Rexx POS?
> 
> 2.   Searching the PoOp for such an instruction led me to CUSE. It does
> not seem that CUSE could be used for this - is that correct? If I am reading
> CUSE correctly, then given "Now is the time", "All is well" and 2 or 3 would
> return the position of "is". Is my reading correct? What would that be good
> for? What would be a reasonable real-world use?
>  
BTW, the classic technique is Boyer-Moore.  But see:
https://en.wikipedia.org/wiki/String_searching_algorithm

This might easily be done in microcode.  And I once coded a case-insensitive
variant of Boyer-Moore.

-- gil

Re: Two string instruction questions

[Jonathan Scott]

Ref:  Your note of 14 March 2018, 08:51:22 -0700
 
Charles Mills wrote:
> 1.   Is there a machine instruction that will find one string within
> another? That given "Now is the time" and "is" would find the "is" and
> return a pointer to it? A machine instruction analog of Rexx POS?
 
I'm not aware of any such instruction.  You can use SRST to find a
single character, and that can be used as part of a search algorithm.
 
> 2.   Searching the PoOp for such an instruction led me to CUSE. It does
> not seem that CUSE could be used for this - is that correct? If I am reading
> CUSE correctly, then given "Now is the time", "All is well" and 2 or 3 would
> return the position of "is". Is my reading correct? What would that be good
> for? What would be a reasonable real-world use?
 
CUSE is very useful for creating an efficient representation of the
differences between two versions of a record, for example to create a
log record showing which bytes changed. It finds the end of a block of
changed data, ignoring unchanged sequences which are too small to be
worth representing as a separate section. To skip to the end of the
unchanged data, CLCL or CLCLE can be used.
 
Jonathan Scott
HLASM, IBM Hursley, UK

Re: Two string instruction questions

[Tony Harminc]

On 14 March 2018 at 11:51, Charles Mills  wrote:
> 1.   Is there a machine instruction that will find one string within
> another? That given "Now is the time" and "is" would find the "is" and
> return a pointer to it? A machine instruction analog of Rexx POS?

I am almost certain that there is not.

> 2.   Searching the PoOp for such an instruction led me to CUSE. It does
> not seem that CUSE could be used for this - is that correct? If I am reading
> CUSE correctly, then given "Now is the time", "All is well" and 2 or 3 would
> return the position of "is". Is my reading correct? What would that be good
> for? What would be a reasonable real-world use?

The usefulness or otherwise of CUSE has been discussed here a couple
of times over the years. IIRC DB2 (sorry, Db2) rows were mentioned.
Certainly CUSE by itself is not a Rexx-style POS. I found the
description in the POP unclear, and so wrote and tested some little
programs to see what it does. It finds a substring match *at the same
offset* in both strings, so it cannot do what is needed for POS.

To implement POS I ended up using SRST to find the first character,
followed by CLCL for the rest, and it turns out that the register
setup is not too bad for this use if you choose carefully. I couldn't
see a useful way to exploit CUSE after SRST for this, but it's quite
possible I missed the trick. And it's certainly possible that an
old-school approach using CLC in a loop might be faster than my two
step process.

To do POS my way you have to remember a couple of  (obvious but easy
to forget) things:

- if your SRST finds the first character but the CLCL doesn't find the
whole string, then you have to loop back and try for the first
character again. For example if you searched a string "Now is the time
it seems" for "it".

- you have to keep track of the length remaining in the string you are
searching in when you do the CLCL. Unlike SRST, which uses an
end-pointer that should stay valid, CLCL needs a current length for
each operand. Somehow I wanted CLCL's padding to be useful here, but
it isn't.

And a more subtle one:
- SRST does not change the pointer to the string (the second register)
unless you get CC3. For this instruction CC3 can occur only if the
string is > 256 bytes. So if you know in advance that this will be the
case, you may be able to avoid having to fix up that register if you
are going to need it to calculate a length for the CLCL. OTOH you may
be setting up a code maintainer for a nightmare if you assume this and
conditions change later. It's also almost impossible to keep track in
one's head of which of the "CPU determined number of bytes"
instructions guaranty at least how many bytes. Some are 256, some are
8, and many are 1.

I'm not a Compiler Guy, but I am fascinated that this kind of subtlety
can be encoded in instruction description tables that compiler code
generation can use.

Tony H.

Re: Two string instruction questions

[Seymour J Metz]

I don't recall seeing "at same offset" in the PoOps. I'll double check and, if 
appropriate, send an RCF.


--
Shmuel (Seymour J.) Metz
http://mason.gmu.edu/~smetz3


From: IBM Mainframe Assembler List <ASSEMBLER-LIST@listserv.uga.edu> on behalf 
of Charles Mills <charl...@mcn.org>
Sent: Wednesday, March 14, 2018 12:33 PM
To: ASSEMBLER-LIST@listserv.uga.edu
Subject: Re: Two string instruction questions

I don't read it that way but I am less than certain of my interpretation.
Some CUSE examples in Appendix A would be nice, eh?

I tried Googling for that and got this which seems to support my
interpretation of CUSE:
http://secure-web.cisco.com/1WeuX5IJkM54cWmI1-Adusi3_Nq8j51SDdDLcRonfztzRjOX28BXK1MAPxpEsFAMZIDKNnI4j3eAapp7ZCXFFLowlGR4QELy_XfzSMfnlTu91jY4TregKNiwgho-lesmjaQqWG3f-10JrcjjEhUy4sgyaunS6nf0_sPzuA1WLqpSUejPNNOTm05OLGFLvF3EViDTtGBGZRzeZS8Hf-JJ9xKB5dlcgZZ3HoYOwGqpYufVEVEeZQNPKp4VnqEV46JslqlGRaYcoY7uCpeEaUjexO-rNwAIjyzYyYqJGm-Wsm68PaokfRTsCL0b-AqME36O1RMFmzC9VltFF3hEQh7u4Ynbiyl_wX5py-E07slzlwfFjp94Ij-l0UxD8izoO7Q6VIOfS6NpOSSJKIXBSTlEV9yQ73f6Et3Qn2_rc69Gqa0cNTpcfQizBt2kFl2inBN7B/http%3A%2F%2Fibmmainframes.com%2Fabout23525.html

It also shares my "well then, what IS CUSE good for?"

FWIW, both of my stings are anticipated to be < 256 bytes.

Charles

-Original Message-
From: IBM Mainframe Assembler List [mailto:ASSEMBLER-LIST@LISTSERV.UGA.EDU]
On Behalf Of Seymour J Metz
Sent: Wednesday, March 14, 2018 9:22 AM
To: ASSEMBLER-LIST@LISTSERV.UGA.EDU
Subject: Re: Two string instruction questions

If your search string is less than 256 bytes then CUSE should work, if I am
reading th PoOps correctly. Set R0 to the length of the search string.


--
Shmuel (Seymour J.) Metz
http://mason.gmu.edu/~smetz3


From: IBM Mainframe Assembler List <ASSEMBLER-LIST@listserv.uga.edu> on
behalf of Charles Mills <charl...@mcn.org>
Sent: Wednesday, March 14, 2018 11:51 AM
To: ASSEMBLER-LIST@listserv.uga.edu
Subject: Two string instruction questions

1.   Is there a machine instruction that will find one string within
another? That given "Now is the time" and "is" would find the "is" and
return a pointer to it? A machine instruction analog of Rexx POS?

2.   Searching the PoOp for such an instruction led me to CUSE. It does
not seem that CUSE could be used for this - is that correct? If I am reading
CUSE correctly, then given "Now is the time", "All is well" and 2 or 3 would
return the position of "is". Is my reading correct? What would that be good
for? What would be a reasonable real-world use?

Re: Two string instruction questions

[retired mainframer]

According to John Ehrman's "Assembler Language Programming for IBM System zT
Servers Version 2.00," the CUSE instruction searches only for matches at the
same offset.  In the case you describe, it would not find a match unless the
second string was "is" so that the word you are looking for is at offset
4 in both strings.

> -Original Message-
> From: IBM Mainframe Assembler List 
> On Behalf Of Charles Mills
> Sent: Wednesday, March 14, 2018 8:51 AM
> To: ASSEMBLER-LIST@LISTSERV.UGA.EDU
> Subject: Two string instruction questions
> 
> 1.   Is there a machine instruction that will find one string within
> another? That given "Now is the time" and "is" would find the "is" and
> return a pointer to it? A machine instruction analog of Rexx POS?
> 
> 2.   Searching the PoOp for such an instruction led me to CUSE. It
does
> not seem that CUSE could be used for this - is that correct? If I am
reading
> CUSE correctly, then given "Now is the time", "All is well" and 2 or 3
would
> return the position of "is". Is my reading correct? What would that be
good
> for? What would be a reasonable real-world use?

Re: Two string instruction questions

[Charles Mills]

I don't read it that way but I am less than certain of my interpretation.
Some CUSE examples in Appendix A would be nice, eh?

I tried Googling for that and got this which seems to support my
interpretation of CUSE:
http://ibmmainframes.com/about23525.html 

It also shares my "well then, what IS CUSE good for?"

FWIW, both of my stings are anticipated to be < 256 bytes.

Charles

-Original Message-
From: IBM Mainframe Assembler List [mailto:ASSEMBLER-LIST@LISTSERV.UGA.EDU]
On Behalf Of Seymour J Metz
Sent: Wednesday, March 14, 2018 9:22 AM
To: ASSEMBLER-LIST@LISTSERV.UGA.EDU
Subject: Re: Two string instruction questions

If your search string is less than 256 bytes then CUSE should work, if I am
reading th PoOps correctly. Set R0 to the length of the search string.


--
Shmuel (Seymour J.) Metz
http://mason.gmu.edu/~smetz3


From: IBM Mainframe Assembler List <ASSEMBLER-LIST@listserv.uga.edu> on
behalf of Charles Mills <charl...@mcn.org>
Sent: Wednesday, March 14, 2018 11:51 AM
To: ASSEMBLER-LIST@listserv.uga.edu
Subject: Two string instruction questions

1.   Is there a machine instruction that will find one string within
another? That given "Now is the time" and "is" would find the "is" and
return a pointer to it? A machine instruction analog of Rexx POS?

2.   Searching the PoOp for such an instruction led me to CUSE. It does
not seem that CUSE could be used for this - is that correct? If I am reading
CUSE correctly, then given "Now is the time", "All is well" and 2 or 3 would
return the position of "is". Is my reading correct? What would that be good
for? What would be a reasonable real-world use?

Re: Two string instruction questions

[Richard Kuebbing]

It appears if r0 also equals length of second operand, you should get desired 
result.

Richard

-Original Message-
From: IBM Mainframe Assembler List [mailto:ASSEMBLER-LIST@LISTSERV.UGA.EDU] On 
Behalf Of Seymour J Metz
Sent: Wednesday, March 14, 2018 12:22 PM
To: ASSEMBLER-LIST@LISTSERV.UGA.EDU
Subject: Re: Two string instruction questions

If your search string is less than 256 bytes then CUSE should work, if I am 
reading th PoOps correctly. Set R0 to the length of the search string.


--
Shmuel (Seymour J.) Metz
http://mason.gmu.edu/~smetz3


From: IBM Mainframe Assembler List <ASSEMBLER-LIST@listserv.uga.edu> on behalf 
of Charles Mills <charl...@mcn.org>
Sent: Wednesday, March 14, 2018 11:51 AM
To: ASSEMBLER-LIST@listserv.uga.edu
Subject: Two string instruction questions

1.   Is there a machine instruction that will find one string within
another? That given "Now is the time" and "is" would find the "is" and return a 
pointer to it? A machine instruction analog of Rexx POS?

2.   Searching the PoOp for such an instruction led me to CUSE. It does
not seem that CUSE could be used for this - is that correct? If I am reading 
CUSE correctly, then given "Now is the time", "All is well" and 2 or 3 would 
return the position of "is". Is my reading correct? What would that be good 
for? What would be a reasonable real-world use?



Thanks,



Charles


- The information contained in this 
communication (including any attachments hereto) is confidential and is 
intended solely for the personal and confidential use of the individual or 
entity to whom it is addressed. The information may also constitute a legally 
privileged confidential communication. If the reader of this message is not the 
intended recipient or an agent responsible for delivering it to the intended 
recipient, you are hereby notified that you have received this communication in 
error and that any review, dissemination, copying, or unauthorized use of this 
information, or the taking of any action in reliance on the contents of this 
information is strictly prohibited. If you have received this communication in 
error, please notify us immediately by e-mail, and delete the original message. 
Thank you

Re: Two string instruction questions

[Seymour J Metz]

If your search string is less than 256 bytes then CUSE should work, if I am 
reading th PoOps correctly. Set R0 to the length of the search string.


--
Shmuel (Seymour J.) Metz
http://mason.gmu.edu/~smetz3


From: IBM Mainframe Assembler List  on behalf 
of Charles Mills 
Sent: Wednesday, March 14, 2018 11:51 AM
To: ASSEMBLER-LIST@listserv.uga.edu
Subject: Two string instruction questions

1.   Is there a machine instruction that will find one string within
another? That given "Now is the time" and "is" would find the "is" and
return a pointer to it? A machine instruction analog of Rexx POS?

2.   Searching the PoOp for such an instruction led me to CUSE. It does
not seem that CUSE could be used for this - is that correct? If I am reading
CUSE correctly, then given "Now is the time", "All is well" and 2 or 3 would
return the position of "is". Is my reading correct? What would that be good
for? What would be a reasonable real-world use?



Thanks,



Charles