Which is Better Tag Language or Old Fashion Panel
Lizette Koehler
We are still using the original way to setup panels. My boss has asked me to convert them to Dialog Tag Language.
Is there a preference to use the Tag language now or is it still what ever you are comfortable with?
Also, is there a conversion utility to take old fashion panel and convert to Tab lang?
Is a compiled panel better than a source?
More questions later.
Thanks bunches.
Lizette
Jim Moore
Concentrated Logic
Lizette Koehler
I really have no problem either way. I am use to writing the old fashion stuff because it works well. However, I rarely get a chance to use the new DTL to set things up. Not everyone in my universe can do DTL. In fact, I maybe the only one, so I may have to convert it back due to support issues.
Lizette
>
>I can't be of much help but if you don't mind my asking, why does your boss want this done?
>
>Concentrated Logic
>
>-
>
Jim Moore
Concentrated Logic
----- Original Message -----
From: "Lizette Koehler" <star...@MINDSPRING.COM>
To: ISP...@LISTSERV.ND.EDU
Baldon, David
It is my impression that most folks still code panels by hand (i.e. the old fashioned way). There have no updates to DTL for quite a few releases. If people were using it I believe IBM would be enhancing (or fixing) it.
I tend to use DTL for production level stuff but for quick and dirty I still code them by hand. Unfortunately the implementation of DTL makes it difficult to get things laid out exactly as you want them. It get real frustrating real fast.
I do most of my development on my workstation and all the source is stored in a non-mainframe database which means when a panel is xfer'd to the mainframe it goes through a ASCII to EBCDIC conversion and some characters get messed up. It is usually the attribute characters that get message up since you usually use non-typical characters. This can be overcome by using hex values for the attribute characters but it is not easy to specify them in "edit" (especially on the workstation). Using DTL alleviates this issue by making it easy to specify hex values for the attributes. To me that is a big deal and helps make the effort worth it.
I don't know of a conversion utility.
Compiled panels are better in some ways but I think it depends on the use. I'm not convinced that there is enough performance gain to make the extra step worth it.
...David
Baldon, David
Does anyone write anything to run in Workstation Mode? I tried it a couple of times but it wasn't a good experience for me.
...David
-----Original Message-----
From: ISPF discussion list [mailto:ISP...@LISTSERV.ND.EDU] On Behalf Of Lizette Koehler
Amy Schneider
IBM does have a utility for converting old fashioned panels to DTL. Search on ISPDTLC.
It is worthwhile knowing how to code DTL and the features are sometimes nice to have. I hope you're being given a lot of time for testing.
Also, several years ago, someone ran a bunch of tests for comparing the speed of loading compiled panels vs. non-compiled ones. They found no difference in time. There are issues in compiling panels, such as a user is prevented from modifying the panel if they're supposed to be able to and centering panel title doesn't work.
Amy Schneider
CA
--- On Tue, 9/1/09, Lizette Koehler <star...@MINDSPRING.COM> wrote:
Dave Salt
I gave up on DTL many years ago. As one person put it "It's a solution in search of a problem". I couldn't agree more.
Some people seem to have the impression that drop-down menus (etc) are a feature of DTL, but this isn't true. Anything you can code a DTL panel to do you can code a regular panel to do. It's just much more difficult to code it using DTL, so why bother?
IBM also seems to have given up on DTL as there hasn't been any enhancements to it in a very long time. Panels supplied by IBM continue to be supplied as regular rather than DTL, which also says a lot.
BTW, compiled versus non-compiled has nothing to do with regular versus DTL; i.e. both can be compiled.
HTH,
Dave Salt
SimpList(tm) - try it; you'll get it!
http://www.mackinney.com/products/SIM/simplist.htm
----------------------------------------
> Date: Tue, 1 Sep 2009 13:21:51 -0400
> From: star...@MINDSPRING.COM
> Subject: Which is Better Tag Language or Old Fashion Panel
> To: ISP...@LISTSERV.ND.EDU
>
_________________________________________________________________
Click less, chat more: Messenger on MSN.ca
http://go.microsoft.com/?linkid=9677404
Rob Zenuk
someone put it best on this list when they stated DTL is a solution looking
for a problem.
Aside from KEYLIST's, I don't think DTL adds anything to the ISPF product
other than confusion and obfuscation. I fiddled with DTL several years
ago. I actually made some early mods to the DTL "compiler" ISPDTLC which
reads your DTL and produces panel source (and other stuff). By the way,
ISPDTLC is 42K of really ugly REXX code. Originally (circa 1993?), it just
produced compiled panels. Before IBM documented many of the 3270 "GUI" controls
(now there's an oxymoron), I found in the code where the panel source was
generated and wrote it out to a dataset before it got passed to ISPPREP.
After fighting through writing an application using DTL, I decided the
required effort and limitations were not worth the value. I was just as happy
coding panels the original way. I got tired of trying to coerce DTL into
doing what I wanted. I was more productive using native ISPF panels.
Since I decided I did not need/like KEYLIST's, never developed ISPF
workstation applications and did not have the PC upload issue David described; I
gave up on DTL.
My two cents...
Rob
In a message dated 9/1/2009 11:46:14 A.M. US Mountain Standard Time,
amys...@YAHOO.COM writes:
When I had my ISPF training, the instructor said that if he was writing
new panels, he'd do it in DTL, but for existing panels, he'd keep them as
they were written.
IBM does have a utility for converting old fashioned panels to DTL.
Search on ISPDTLC.
It is worthwhile knowing how to code DTL and the features are sometimes
nice to have. I hope you're being given a lot of time for testing.
Also, several years ago, someone ran a bunch of tests for comparing the
speed of loading compiled panels vs. non-compiled ones. They found no
difference in time. There are issues in compiling panels, such as a user is
prevented from modifying the panel if they're supposed to be able to and
centering panel title doesn't work.
Amy Schneider
CA
--- On Tue, 9/1/09, Lizette Koehler <star...@MINDSPRING.COM> wrote:
> From: Lizette Koehler <star...@MINDSPRING.COM>
> Subject: Which is Better Tag Language or Old Fashion Panel
> To: ISP...@LISTSERV.ND.EDU
> Date: Tuesday, September 1, 2009, 1:21 PM
Pinnacle
My responses below. My personal recommendation is not to use DTL unless you
really want keylist support.
Regards,
Tom Conley
----- Original Message -----
From: "Lizette Koehler" <star...@MINDSPRING.COM>
Newsgroups: bit.listserv.ispf-l
Sent: Tuesday, September 01, 2009 2:08 PM
Subject: Which is Better Tag Language or Old Fashion Panel
> Listers -
>
> We are still using the original way to setup panels. My boss has asked me
> to convert them to Dialog Tag Language.
>
> Is there a preference to use the Tag language now or is it still what ever
> you are comfortable with?
>
The only reason to use DTL is if you want to implement keylist support for
specific panels. Per ISPF development, DTL is the only way to do keylists
correctly.
> Also, is there a conversion utility to take old fashion panel and convert
> to Tab lang?
>
No. You have to add the tags yourself.
> Is a compiled panel better than a source?
>
You need to draw a distinction here. There is DTL source, which is
compiled, the output of which is panel source. Do not confuse that with the
"preprocessed panel", which is the compressed, uneditable, execution-rady
form of a panel, created from panel source.
DTL source -> compiler -> panel source -> preprocessor -> preprocessed panel
John P Kalinich
on 09/01/2009 01:30:52 PM:
> My boss would like the new-fangle stuff. Thinks it is Cooler. Plus
> the use of ABC and dynamic setups could be nicer.
You can do Action Bar Choices in the old panel language quite easily. And
it does modernize the look of your panels. I have plenty of examples if
you want them.
Regards,
John K
Dave Salt
You've received a couple of responses saying not to use DTL unless you want keylist support. However, regular ISPF panels fully support the use of keylists.
This confirms what I said in my earlier email about the perception that DTL panels support features that regular panels don't. Many years ago I started off with this same perception; i.e. if DTL was so much harder to code than regular panels then there had to be SOME benefit to using it; i.e. features that aren't supported in regular panels? But once I realized that all panel features are equally available in both DTL and regular panels, I immediately gave up on DTL and went back to coding regular panels.
Dave Salt
SimpList(tm) - try it; you'll get it!
http://www.mackinney.com/products/SIM/simplist.htm
----------------------------------------
> Date: Tue, 1 Sep 2009 15:23:09 -0400
> From: pinn...@ROCHESTER.RR.COM
> Subject: Re: Which is Better Tag Language or Old Fashion Panel
> To: ISP...@LISTSERV.ND.EDU
Baldon, David
...David
-----Original Message-----
From: ISPF discussion list [mailto:ISP...@LISTSERV.ND.EDU] On Behalf Of Dave Salt
Sent: Tuesday, September 01, 2009 2:01 PM
To: ISP...@LISTSERV.ND.EDU
Baldon, David
...David
-----Original Message-----
From: ISPF discussion list [mailto:ISP...@LISTSERV.ND.EDU] On Behalf Of Dave Salt
Lizette Koehler
I have taken your comments to my boss and we are going to discuss whether or not to use DTL. My main concern would be that I would be the only one that could support it.
So you have - as always - been AWESOME!
Lizette
Baldon, David
Mosley, George
Following PL/I compiles, we run a TSO batch step to pack the listing
data and to apply ISPF stats to the listing PDS member.
The code is below.
The problem I'm looking for help with is that this code can be executing
simultaneously for many programs at a time, all trying to update the
same listing PDS. Occasionally, we get messages (from the ISPF Log) and
the TSO step fails with a RC of 990:
Start of ISPF Log - - - - Session # 1
-------------------------------------------------------
TSO - Command - - %WZUSTAT PLI.TEST.LISTING ABC001 MOI9 PACK
TSO - Command - - %WZEPACK
- A RESERVE failed when attempting to save ().
Edit will retry a maximum 4
times.'
- A RESERVE failed when attempting to save ().
Edit will retry a maximum 4
times.'
- A RESERVE failed when attempting to save ().
Edit will retry a maximum 4
times.'
- A RESERVE failed when attempting to save ().
Edit will retry a maximum 4
times.'
- A RESERVE failed when attempting to save ().
Edit will retry a maximum 4
times.'
The REXX Code
/* REXX */
Arg statpds statmem userid packopt
Address ISPEXEC
"CONTROL ERRORS RETURN"
member = Strip(statmem)
"LMINIT DATAID(did1) DATASET('"statpds"') ENQ(SHRW)"
retc = rc
If retc = 0 Then Do
"EDIT DATAID("did1") MEMBER("member") MACRO(WZEPACK)"
If rc > 4 Then
Say 'Error editing 'packpds'('member')'
"LMMSTATS DATAID("did1") MEMBER("member") USER("userid")"
If rc > 0 Then
Say 'STATs for 'statpds'('member') not saved. RC' retc
End
"LMFREE DATAID("did1")"
Exit 0
The EDIT Macro
ISREDIT MACRO
ISREDIT PACK ON
ISREDIT BUILTIN END
ISREDIT MEND
EXIT
Any suggestions, anyone?
Thanks.
George
Steve Coalbran
/S
> Date: Tue, 1 Sep 2009 13:21:51 -0400
> From: star...@MINDSPRING.COM
> Subject: Which is Better Tag Language or Old Fashion Panel
> To: ISP...@LISTSERV.ND.EDU
_________________________________________________________________
Access your other email accounts and manage all your email from one place.
http://clk.atdmt.com/UKM/go/167688463/direct/01/
Steve Coalbran
RTFM time again ?!?!?!?
/S
> Date: Tue, 1 Sep 2009 11:45:53 -0700
> From: amys...@YAHOO.COM
> Subject: Re: Which is Better Tag Language or Old Fashion Panel
> To: ISP...@LISTSERV.ND.EDU
>
> When I had my ISPF training, the instructor said that if he was writing new panels, he'd do it in DTL, but for existing panels, he'd keep them as they were written.
>
> IBM does have a utility for converting old fashioned panels to DTL. Search on ISPDTLC.
>
> It is worthwhile knowing how to code DTL and the features are sometimes nice to have. I hope you're being given a lot of time for testing.
>
> Also, several years ago, someone ran a bunch of tests for comparing the speed of loading compiled panels vs. non-compiled ones. They found no difference in time. There are issues in compiling panels, such as a user is prevented from modifying the panel if they're supposed to be able to and centering panel title doesn't work.
>
> Amy Schneider
> CA
>
>
> --- On Tue, 9/1/09, Lizette Koehler <star...@MINDSPRING.COM> wrote:
>
> > From: Lizette Koehler <star...@MINDSPRING.COM>
> > Subject: Which is Better Tag Language or Old Fashion Panel
> > To: ISP...@LISTSERV.ND.EDU
> > Date: Tuesday, September 1, 2009, 1:21 PM
> > Listers -
> >
> > We are still using the original way to setup panels.
> > My boss has asked me to convert them to Dialog Tag
> > Language.
> >
> > Is there a preference to use the Tag language now or is it
> > still what ever you are comfortable with?
> >
> > Also, is there a conversion utility to take old fashion
> > panel and convert to Tab lang?
> >
> > Is a compiled panel better than a source?
> >
> > More questions later.
> >
> > Thanks bunches.
> >
> > Lizette
> >
>
>
>
_________________________________________________________________
Save time by using Hotmail to access your other email accounts.
http://clk.atdmt.com/UKM/go/167688463/direct/01/
Arthur Gutowski
<star...@MINDSPRING.COM> wrote:
>Thank you every one.
>
>I have taken your comments to my boss and we are going to discuss
whether or not to use DTL. My main concern would be that I would be the
only one that could support it.
>
>So you have - as always - been AWESOME!
>
>Lizette
FWIW, I don't think DTL is inherently any more difficult (or easy) to
understand than panel source. It's just another language. And, my
experience the development team in Perth (I think that's where they are) is
more than accomodating with enhancement requests and bug fixes.
We (still) have a number of copies of ISR@PRIM floating around, where
someone thought they could whack the panel source around without touching
the panel source. Had they known about PNS fields and how to manipulate
the statements associated with this, it would have been fine, but they didn't,
and clicking on "Workplace" took you to RACF, and clicking on RACF did
nothing at all. Without starting a second "holy war", I found it easier
to "return to the source" and create a USERMOD around our modifications. I
know it's extra work to compile the panel and maintaine DTL and panel source
for the updates, but at least, like any decent applications programmer, I
have "source/object" release management. I'm still trying to rid myself of the
various "children out of wedlock" however (takes some time here).
In fact, for new panels (tables, too), I start with DTL source, using existing
panels (mine or IBM's) as a starting point.
I've only lodged two substantial complaints with DTL to date. One is in getting
custom INIT, PROC, REINIT logic placed correctly in the compiled panel
source. In fact, DTL doesn't easily support panel logic in it's "native tongue",
so I'm relegated to coding <source> tags with embedded panel source. Yuck.
The other appears to have been addressed - a couple of years ago I was not
able to successfully build ISPF command tables using DTL source. As I recall,
even the DTL IBM provided for ISPCMDS did not compile cleanly. Either I got
smarter (dumb luck being more likely) or IBM fixed up their code (or a little of
both). It seems to work OK now.
In the end, everyone who said "it depends" was right. For me, DTL is a good
way to retain and enhance panel functionality provided by IBM, even though it
is a little more work to maintain.
Regards,
Art Gutowski
Ford Motor Company
Rob Scott
Rob Scott
Developer
Rocket Software
275 Grove Street * Newton, MA 02466-2272 * USA
Tel: +1.617.614.2305
Email: rsc...@rs.com
Web: www.rocketsoftware.com
-----Original Message-----
From: ISPF discussion list [mailto:ISP...@LISTSERV.ND.EDU] On Behalf Of Arthur Gutowski
Sent: 02 September 2009 14:40
To: ISP...@LISTSERV.ND.EDU
Subject: Re: Which is Better Tag Language or Old Fashion Panel
Baldon, David
<!doctype dm system [
<!entity AttrDefs system>
<!entity CopyRite system>
]>
Place this where you want the include to occur;
&AttrDefs; <!-- case must match embedded file name exactly -->
...David
Rob Zenuk
panels (and stuff). If there is a way to convert native panels to DTL, it has
eluded me...
From the z/OS 1.10 ISPF DTL Guide:
1.1.5 What is the ISPF conversion utility?
ISPDTLC is the ISPF conversion tool that converts Dialog Tag Language (DTL)
source files to ISPF panel language source format or executable
preprocessed ISPF format. ISPF provides you with an invocation panel that allows you
to specify a number of options for the conversion, or you can use
conversion utility command syntax from the command line of your terminal. _Chapter
10, "Using the conversion utility" in topic 1.10_
(http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/ISPZDT70/1.10?SHELF=ISPZPM70&DT=20080610202
255#HDRCONV) provides a complete description of both methods.
Rob
In a message dated 9/2/2009 5:06:31 A.M. US Mountain Standard Time,
coal...@HOTMAIL.COM writes:
Wow! It exists already?!
RTFM time again ?!?!?!?
/S
> Date: Tue, 1 Sep 2009 11:45:53 -0700
> From: amys...@YAHOO.COM
> Subject: Re: Which is Better Tag Language or Old Fashion Panel
Pedro Vera
Not only does it exist, but ENTITY is the standard tag for such things. At
least, it is used in XML. Probably, the confusion was in wanting to use
cobol terminology when working with a markup language.
==========
My two cents: it is easier to use DTL for simple panels. For example, I
use it for a large number of help panels. You do not have to worry so
much about formatting the text to look nice. After all, that is one of
the strengths of a tag language.
I also use the imbed capability quite heavily. I use for the action bar:
if you have hundreds of panels, you would not believe how hard it is to
implement a 'small' design change. But with an imbed capability, you make
your change once, then re-compile them all. I have also used it to imbed
a standard copyright statement.
Yes, you can do everything using the native panel language, but the more
you do of action bars, point and shoot, field level help, keylists,
(probably others I have forgotten), the more DTL proves its usefulness.
I like DTL, but I am not a purist... sometimes the compiler will not
cooperate in making exactly what I want. In those cases, I start with DTL
then edit the resulting panel.
side note: I edit HTML files from the ISPF editor and rely on its DTL
highlighting to it nicely.
Pedro Vera
Pedro Vera
but better integrity for your dataset.
How about adding a DD statement to your JCL with DISP=OLD?
Pedro Vera
Mosley, George
and user expectation.
I'd be satisfied with a way to end gracefully and ignore the "error".
The occasional unpacked listing would not be unacceptable.
George (x7327)
-----Original Message-----
From: ISPF discussion list [mailto:ISP...@LISTSERV.ND.EDU] On Behalf Of
Pedro Vera
Sent: Wednesday, September 02, 2009 9:24 AM
To: ISP...@LISTSERV.ND.EDU
Pedro Vera
------------------------------------------------------------
This email and any attachments are intended only for the named
recipient and may contain confidential and/or privileged material.
Any unauthorized copying, dissemination or other use by a person
other than the named recipient of this communication is prohibited.
If you received this in error or are not named as a recipient,
please notify the sender and destroy all copies of this email
immediately.
Rob Scott
Thanks for that - it appears to be exactly what I need.
Pedro Vera
editor macro, it does an implied SAVE.
How about changing the macro to explicitly issue the SAVE and then
checking the return code before exiting the macro. If it did not work,
then sleep for a few seconds and try again or end gracefully.
(you seem to show clist syntax... rexx is recommended now-a-days).
Pedro Vera
Mosley, George
That's an idea. Unfortunately, testing is difficult since I haven't been
able, so far, to recreate the problem at will.
The other idea I had was to set ZISPFRC to 0 in the invoking Rexx code
before exiting.
Not sure what you mean by the "clist syntax"...
George (x7327)
-----Original Message-----
From: ISPF discussion list [mailto:ISP...@LISTSERV.ND.EDU] On Behalf Of
Pedro Vera
Sent: Wednesday, September 02, 2009 10:07 AM
To: ISP...@LISTSERV.ND.EDU
Pedro Vera
Pedro Vera
I think that will mask other errors and may have unintended consequences.
About clist: the sample program you provide was not complete, but it did
not look like rexx.
Pedro Vera
Scott T. Harder
you have a *huge* ISPF application and are worried about performance there,
is that you can consistently create more "IBM-looking" panels if everyone
uses DTL. Now, you can do this, also, with the old panel language, but
across a group of people I think that it's easier to do consistently with
DTL.
I learned DTL easily and I had never worked with a tag language before that.
The manuals were great (as I recall) and so it was real easy to learn. It
was *fun* too, by way of being different. But, I would suggest that once
you decide to go DTL that you stick with it across the board and convert the
old panel language panels as work is required to be done on them. You'll
probably end up with panels that never get converted that way, but it
doesn't make sense to me to go and convert panels that will never need to
get updated, just to get the "new-fangled stuff". That would be a complete
waste of time.
Just my $0.02,
Scott T. Harder
-----Original Message-----
From: ISPF discussion list [mailto:ISP...@LISTSERV.ND.EDU] On Behalf Of
Lizette Koehler
Sent: Tuesday, September 01, 2009 2:31 PM
To: ISP...@LISTSERV.ND.EDU
Subject: Re: Which is Better Tag Language or Old Fashion Panel
My boss would like the new-fangle stuff. Thinks it is Cooler. Plus the use
of ABC and dynamic setups could be nicer.
I really have no problem either way. I am use to writing the old fashion
stuff because it works well. However, I rarely get a chance to use the new
DTL to set things up. Not everyone in my universe can do DTL. In fact, I
maybe the only one, so I may have to convert it back due to support issues.
Lizette
>
>I can't be of much help but if you don't mind my asking, why does your boss
want this done?
>
>Concentrated Logic
>
>-
>
Dave Salt
DO I = 1 TO 1000 UNTIL MESSAGE_SINKS_IN
Say "Regular ISPF panels *CAN* be compiled. They *CAN* use drop down menus. They *CAN* use keylists. From a user perspective, there is no new fangled stuff that a DTL panel can do over and above what a regular ISPF panel can do."
END
:-)
Dave Salt
SimpList(tm) - try it; you'll get it!
http://www.mackinney.com/products/SIM/simplist.htm
----------------------------------------
> Date: Wed, 2 Sep 2009 17:19:01 -0400
> From: scottyt...@GMAIL.COM
> Subject: Re: Which is Better Tag Language or Old Fashion Panel
> To: ISP...@LISTSERV.ND.EDU
George.William
-----Original Message-----
From: ISPF discussion list [mailto:ISP...@LISTSERV.ND.EDU] On Behalf Of
Dave Salt
Sent: Wednesday, September 02, 2009 3:20 PM
To: ISP...@LISTSERV.ND.EDU
Subject: Re: Which is Better Tag Language or Old Fashion Panel
/* REXX to promote the use of regular ISPF panels versus DTL */
DO I = 1 TO 1000 UNTIL MESSAGE_SINKS_IN
Say "Regular ISPF panels *CAN* be compiled. They *CAN* use drop down
menus. They *CAN* use keylists. From a user perspective, there is no new
fangled stuff that a DTL panel can do over and above what a regular ISPF
panel can do."
END
:-)
Dave Salt
SimpList(tm) - try it; you'll get it!
http://www.mackinney.com/products/SIM/simplist.htm
______________________________________________________________________
CONFIDENTIALITY NOTICE: This email from the State of California is for the sole use of the intended recipient and may contain confidential and privileged information. Any unauthorized review or use, including disclosure or distribution, is prohibited. If you are not the intended recipient, please contact the sender and destroy all copies of this email.
Pedro Vera
That is a good thing! </humor>
But seriously, some of our products occasionally get customer requirements
because they have inconsistent user interfaces. That is, the customers do
not want to have to know that you have to do this in application A, have
to do that in application B and have to do something different in
application C. There is a lot of goodwill to be gained with consistency.
Pedro Vera
Baldon, David
One thing that old style panels cannot do is what the entity statement can do. Specifically including common panel source from a common member into multiple panels. However, with z/OS 1.11 you should be able to do it via the new panel source statement. It'll take some effort but I think it can be done. I never did understand why an )INCLUDE statement was never done for panels...oh well...it is what it is.
...David
Pedro Vera
I think we disagree about terminology.
- DTL files can be 'compiled' using the ISPDTLC utilitiy, to create
regular ISPF panels.
- ISPF panels can be 'preprocessed' using the ISPPREP dialog.
At least, that is the terminology used in the manuals.
Pedro Vera
Dave Salt
Dave Salt
SimpList(tm) - try it; you'll get it!
http://www.mackinney.com/products/SIM/simplist.htm
----------------------------------------
> Date: Wed, 2 Sep 2009 16:45:09 -0600
> From: David_...@BMC.COM
_________________________________________________________________
New! Open Messenger faster on the MSN homepage
http://go.microsoft.com/?linkid=9677405
Steve Comstock
You can mix and match, to a degree. In our course "Developing Dialog
Manager Applications in z/OS" we use mostly panel language. But near
the end we use DTL for keylists.
I once created a whole course on DTL, but no one ever bought it.
Six months of time, total waste.
YMMV.
Kind regards,
-Steve Comstock
The Trainer's Friend, Inc.
_www.trainersfriend.com_ (http://www.trainersfriend.com)
In a message dated 9/1/2009 12:30:56 P.M. Mountain Daylight Time,
star...@MINDSPRING.COM writes:
My boss would like the new-fangle stuff. Thinks it is Cooler. Plus the
use of ABC and dynamic setups could be nicer.
I really have no problem either way. I am use to writing the old fashion
stuff because it works well. However, I rarely get a chance to use the new
DTL to set things up. Not everyone in my universe can do DTL. In fact, I
maybe the only one, so I may have to convert it back due to support issues.
Lizette
>
>I can't be of much help but if you don't mind my asking, why does your
boss want this done?
>
>Concentrated Logic
>
>-
>
>Listers -
>
>We are still using the original way to setup panels. My boss has asked
me to convert them to Dialog Tag Language.
>
>Is there a preference to use the Tag language now or is it still what
ever you are comfortable with?
>
>Also, is there a conversion utility to take old fashion panel and convert
to Tab lang?
>
>Is a compiled panel better than a source?
>
>More questions later.
>
>Thanks bunches.
>
>Lizette
**************A Good Credit Score is 700 or Above. See yours in just 2 easy
steps!
(http://pr.atwola.com/promoclk/100126575x1222846709x1201493018/aol?redir=http://www.freecreditreport.com/pm/default.aspx?sc=668072&hmpgID=115&bcd
=JulystepsfooterNO115)
Dave Salt
> I think we disagree about terminology.
>
> - DTL files can be 'compiled' using the ISPDTLC utilitiy, to create
> regular ISPF panels.
We must be looking at different manuals. The ISPF DTL Guide says ISPDTLC is the "ISPF Dialog Tag Language Conversion Utility". It converts (not compiles) DTL to create regular ISPF panels.
> - ISPF panels can be 'preprocessed' using the ISPPREP dialog.
I agree; regular panels can be preprocessed into something that's analogous to compiled programs; i.e. uneditable gobbeldygook that runs more efficiently than the original interpreted source code. Several people have referred to panels as being 'compiled' and I didn't see any need to correct them.
Dave Salt
SimpList(tm) - try it; you'll get it!
http://www.mackinney.com/products/SIM/simplist.htm
> From: pe...@US.IBM.COM
> I think we disagree about terminology.
>
> - DTL files can be 'compiled' using the ISPDTLC utilitiy, to create
> regular ISPF panels.
>
> - ISPF panels can be 'preprocessed' using the ISPPREP dialog.
>
> At least, that is the terminology used in the manuals.
>
>
> Pedro Vera
_________________________________________________________________
New: Messenger sign-in on the MSN homepage
http://go.microsoft.com/?linkid=9677403
Pedro Vera
You did not search far enough, from pg 3:
v What is the ISPF Conversion Utility? This section provides a description
of the conversion utility, the compiler you use to convert your DTL source
files for use by ISPF.
There are numerous other references to 'compiler' in the book.
Pedro Vera
Dave Salt
<OFA1262AA4.27FED4D4-ON882576...@us.ibm.com>
<BLU149-W2682D67B9...@phx.gbl>
<OF1706A9AC.7C319937-ON882576...@us.ibm.com>
Content-Type: text/plain; charset="iso-8859-1"
MIME-Version: 1.0
X-OriginalArrivalTime: 03 Sep 2009 00:06:00.0584 (UTC) FILETIME=[51C8C080:01CA2C2A]
X-Source-IP: 65.55.116.16
X-Spam: No
X-Spam-Score: 0.00%
X-ND-MTA-Date: Wed, 02 Sep 2009 20:06:28 EDT
X-ND-CM-Score: 0.00%
X-ND-CT-Class: Not-spam
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by lists-p1.cc.nd.edu id n83062rD020984
> From: pe...@US.IBM.COM
> You did not search far enough, from pg 3:
>
> v What is the ISPF Conversion Utility? This section provides a description
> of the conversion utility, the compiler you use to convert your DTL source
> files for use by ISPF.
I went to page 3 and clicked the link in the section you referenced above and it took me to this:
"ISPDTLC is the ISPF conversion tool that converts Dialog Tag Language (DTL) source files to ISPF panel language source format or executable pre-processed ISPF format. ISPF provides you with an invocation panel that allows you to specify a number of options for the conversion, or you can use conversion utility command syntax from the command line of your terminal. Chapter 10, "Using the conversion utility" in topic 1.10 provides a complete description of both methods."
As you can see it mentions conversion 5 times. It doesn't mention the word 'compiled' even once.
> There are numerous other references to 'compiler' in the book.
I searched the entire manual for 'compiler' and there are 10 hits. I searched it for 'convert' and there are 35 hits. But all of this is beside the point. When people on this list have mentioned 'compiled panels', I've taken it to mean 'preprocessed'. In other words, panels that have been optimized for use by the system and are unreadable to us humans. If your definition of a compiled panel is the output from ISPDTLC, i.e. regular ISPF panel source code, then you and I have a very different understanding of what a compiled panel is.
Dave Salt
SimpList(tm) - try it; you'll get it!
http://www.mackinney.com/products/SIM/simplist.htm
_________________________________________________________________
New! Open Messenger faster on the MSN homepage
http://go.microsoft.com/?linkid=9677405
Pinnacle
From: "Dave Salt" <ds...@HOTMAIL.COM>
Newsgroups: bit.listserv.ispf-l
Sent: Wednesday, September 02, 2009 6:21 PM
Subject: Re: Which is Better Tag Language or Old Fashion Panel
> /* REXX to promote the use of regular ISPF panels versus DTL */
>
> DO I = 1 TO 1000 UNTIL MESSAGE_SINKS_IN
>
> Say "Regular ISPF panels *CAN* be compiled. They *CAN* use drop down
> menus. They *CAN* use keylists. From a user perspective, there is no new
> fangled stuff that a DTL panel can do over and above what a regular ISPF
> panel can do."
>
> END
>
Dave,
When Matt Seabold, Doug Nadel, Marv Knight, Peter Van Dyke, etc. all say
that keylists can only be properly implemented with DTL, I tend to believe
them. I don't recall the specific reasons, but they've all presented on DTL
at SHARE in the past, and they've all made that statement. Here's the
bullet from the latest presentation:
.DTL provides the best way to define the ISPF tables that define keylists
used by an application.
Many of us have heard this before, which is why many of us have made the
statement that DTL should be used if you want to define keylists to your
app.
Regards,
Tom Conley
Dave Salt
I think you're getting confused between how keylists are created and how they are used. Keylists are ISPF tables, and one way to create them is by using DTL. Certainly, that's the way I've always created them. However, NONE of the panels I've written are written in DTL, and they ALL use keylists.
So, I will repeat this statement one last time:
REGULAR ISPF PANELS *CAN* USE KEYLISTS.
Are you sure?
YES!
Completely?
YES!!
There can be no mistake?
NONE!!!
Are you absolutely completely sure?
ARRGGGHHHH!!!! (runs away screaming)...
;-)
Regards,
Dave Salt
SimpList(tm) - try it; you'll get it!
http://www.mackinney.com/products/SIM/simplist.htm
----------------------------------------
> Date: Wed, 2 Sep 2009 21:32:19 -0400
> From: pinn...@ROCHESTER.RR.COM
> Subject: Re: Which is Better Tag Language or Old Fashion Panel
> To: ISP...@LISTSERV.ND.EDU
Pedro Vera
Sorry, I did not realize you alone were the final authority on this
subject. I was relying on the manuals. And clearly, the manuals do not
refer to the preprocessor as a compiler.
>... you and I have a very different understanding of what a compiled...
From Wikipedia: "A compiler is a computer program ... that transforms
source code written in a computer language (the source language) into
another computer language ..."
I think the DTL compiler matches the definition. But I grant you that the
ISPF preprocessor is also covered by the definition.
Pedro Vera
Dave Salt
>> I've taken it to mean ...
>
> Sorry, I did not realize you alone were the final authority on this
> subject.
So if I take something to mean something, it somehow means I'm stating that I'm the final authority? Is that what you've taken what I said to mean?
> I was relying on the manuals. And clearly, the manuals do not
> refer to the preprocessor as a compiler.
I was relying on common sense. And clearly, when people refer to compiled panels, they're not refering to the source panel that comes out of ISPDTLC.
> From Wikipedia: "A compiler is a computer program ... that transforms
> source code written in a computer language (the source language) into
> another computer language ..."
"...the target language, often having a binary form known as object code."
Strange, you somehow seem to have omitted the last half of that sentence, and it's quite important. What comes out of ISPDTLC is source code. What comes out of the preprocessor is binary.
> I think the DTL compiler matches the definition.
Of the first half of the sentence you posted? Okay.
> But I grant you that the
> ISPF preprocessor is also covered by the definition.
Well, that's something.
James Campbell
/* REXX */
Arg statpds statmem userid packopt
Address ISPEXEC
"CONTROL ERRORS RETURN"
call bpxwdyn("alloc da("statpds") old rtddn(ddn)")
do 60 while result = <don't know what you get if there's a lock - try it and see>
call syscalls 'ON'
address syscall "sleep" 1 /*sleep for a second */
call syscalls "off"
call bpxwdyn("alloc da("statpds") old rtddn(ddn)")
end
if result = <don't know> then
exit 16
member = Strip(statmem)
"LMINIT DATAID(did1) DDNAME("ddn") ENQ(SHRW)"
....
"LMFREE ..."
address tso "free da("ddn")"
exit 0
What I've done with some jobs that I expect to run quickly is to add a step to the end
//DUMMY EXEC PGM=IEFBR14,COND=(0,LE)
//DUMMY DD DISP=OLD,DSN=A.DUMMY.DATASET
This ensures that only one job can actually run at a time. Might not be suitable for you
2) As others mentioned, you should explicitly SAVE at the end of a batch edit macro
James Campbell
> ------------------------------
>
> Date: Tue, 1 Sep 2009 13:54:30 -0700
> From: "Mosley, George" <George...@ICBC.COM>
> Subject: Edit Macro Reserve Failure
>
> Hello List.
>
> Following PL/I compiles, we run a TSO batch step to pack the listing
> data and to apply ISPF stats to the listing PDS member.
>
> The code is below.
>
> The problem I'm looking for help with is that this code can be executing
> simultaneously for many programs at a time, all trying to update the
> same listing PDS. Occasionally, we get messages (from the ISPF Log) and
> the TSO step fails with a RC of 990:
>
> Start of ISPF Log - - - - Session # 1
> -------------------------------------------------------
> TSO - Command - - %WZUSTAT PLI.TEST.LISTING ABC001 MOI9 PACK
>
> TSO - Command - - %WZEPACK
>
> - A RESERVE failed when attempting to save ().
> Edit will retry a maximum 4
> times.'
>
> - A RESERVE failed when attempting to save ().
> Edit will retry a maximum 4
> times.'
>
> - A RESERVE failed when attempting to save ().
> Edit will retry a maximum 4
> times.'
>
> - A RESERVE failed when attempting to save ().
> Edit will retry a maximum 4
> times.'
>
> - A RESERVE failed when attempting to save ().
> Edit will retry a maximum 4
> times.'
>
>
> The REXX Code
>
> /* REXX */
> Arg statpds statmem userid packopt
> Address ISPEXEC
> "CONTROL ERRORS RETURN"
> member = Strip(statmem)
> "LMINIT DATAID(did1) DATASET('"statpds"') ENQ(SHRW)"
> retc = rc
> If retc = 0 Then Do
> "EDIT DATAID("did1") MEMBER("member") MACRO(WZEPACK)"
> If rc > 4 Then
> Say 'Error editing 'packpds'('member')'
> "LMMSTATS DATAID("did1") MEMBER("member") USER("userid")"
> If rc > 0 Then
> Say 'STATs for 'statpds'('member') not saved. RC' retc
> End
> "LMFREE DATAID("did1")"
> Exit 0
>
> The EDIT Macro
>
> ISREDIT MACRO
> ISREDIT PACK ON
> ISREDIT BUILTIN END
> ISREDIT MEND
> EXIT
>
> Any suggestions, anyone?
>
> Thanks.
>
> George
>
Paul Gilmartin
>
> call bpxwdyn("alloc da("statpds") old rtddn(ddn)")
> do 60 while result = <don't know what you get if there's a lock -
> try it and see>
> call syscalls 'ON'
> address syscall "sleep" 1 /*sleep for a second */
> call syscalls "off"
> call bpxwdyn("alloc da("statpds") old rtddn(ddn)")
> end
>
(Even LEAVE would be better.)
What's the point of SYSCALLS OFF? It has been discussed on
MVS-OE and an expert IBM employee has advised against it because
of some unpleasant side effects.
-- gil
Scott T. Harder
From: ISPF discussion list [mailto:ISP...@LISTSERV.ND.EDU] On Behalf Of
Pedro Vera
Sent: Wednesday, September 02, 2009 6:42 PM
To: ISP...@LISTSERV.ND.EDU
Subject: Re: Which is Better Tag Language or Old Fashion Panel
>...you can consistently create more "IBM-looking" panels...
That is a good thing! </humor>
;-)
But seriously, some of our products occasionally get customer requirements
because they have inconsistent user interfaces. That is, the customers do
not want to have to know that you have to do this in application A, have
to do that in application B and have to do something different in
application C. There is a lot of goodwill to be gained with consistency.
In both look and feel as well in the area of functionality. I know
people who have struggled with Microsoft MMC. Evidently, it ain't that
easy, but they do it so that their applications basically function in the
same general way as the system (and other) tools that the user is already
familiar with. It does make a huge difference; especially up front when a
user is learning the app and especially on that platform, where a good
manual is hard to find.
All the best,
Scott T. Harder
Steve Coalbran
MESSAGE_SINKS_IN = 1
/Steve
> Date: Wed, 2 Sep 2009 18:19:46 -0400
> From: ds...@HOTMAIL.COM
_________________________________________________________________
View your other email accounts from your Hotmail inbox. Add them now.
http://clk.atdmt.com/UKM/go/167688463/direct/01/
Dave Salt
> I agree. :-)
> MESSAGE_SINKS_IN = 1
LOL; my work is done! :-)
Dave Salt
_________________________________________________________________
New: Messenger sign-in on the MSN homepage
http://go.microsoft.com/?linkid=9677403
Mosley, George
For ages, I've had the following on a PFKey (which I think I got from
someone on this list):
tso tsoexec logoff;=x
Invoking it would exit me from ISPF and complete the TSO Logoff process
automatically.
For some reason, it's stopped working. Now when I invoke it, I get the
following message:
IKJ79105I INVALID COMMAND NAME SYNTAX, LOGOFF;=X
... and I remain wherever I was in ISPF.
Any suggestions, anyone?
Thanks.
George
------------------------------------------------------------
This email and any attachments are intended only for the named
recipient and may contain confidential and/or privileged material.
Any unauthorized copying, dissemination or other use by a person
other than the named recipient of this communication is prohibited.
If you received this in error or are not named as a recipient,
please notify the sender and destroy all copies of this email
immediately.
Pedro Vera
semicolon.
Pedro Vera
IMS Tools
Mark Zelden
wrote:
>Hello List.
>
>For ages, I've had the following on a PFKey (which I think I got from
>someone on this list):
>
>tso tsoexec logoff;=x
>
>Invoking it would exit me from ISPF and complete the TSO Logoff process
>automatically.
>
>For some reason, it's stopped working. Now when I invoke it, I get the
>following message:
>
>IKJ79105I INVALID COMMAND NAME SYNTAX, LOGOFF;=X
>
>... and I remain wherever I was in ISPF.
>
>Any suggestions, anyone?
>
My first thought would be to check your command delimiter setting
under OPTIONS and to verify it is set to a semicolon.
Mark
--
Mark Zelden
Sr. Software and Systems Architect - z/OS Team Lead
Zurich North America / Farmers Insurance Group - ZFUS G-ITO
mailto:mark....@zurichna.com
z/OS Systems Programming expert at http://expertanswercenter.techtarget.com/
Mark's MVS Utilities: http://home.flash.net/~mzelden/mvsutil.html
Mosley, George
That's exactly it. Had to change it for an edit session the other day
and totally forgot about it.
Thanks again.
George (x7327)
-----Original Message-----
From: ISPF discussion list [mailto:ISP...@LISTSERV.ND.EDU] On Behalf Of
Pedro Vera
Sent: Friday, November 20, 2009 8:39 AM
To: ISP...@LISTSERV.ND.EDU
Pedro Vera
IMS Tools
Edward Jaffe
> tso tsoexec logoff;=x
>
> Invoking it would exit me from ISPF and complete the TSO Logoff process
> automatically.
>
Thanks for showing me this. Old dogs love learning new tricks... :-)
--
Edward E Jaffe
Phoenix Software International, Inc
5200 W Century Blvd, Suite 800
Los Angeles, CA 90045
310-338-0400 x318
edj...@phoenixsoftware.com
http://www.phoenixsoftware.com/
Laumann, James A. , Jim
Since this is invoked via a PF key - I have to assume that ISPF is in
charge....
So that says the string should be "=x;TSO TSOEXEC LOGOFF"
I did try it (pf key assignment)- as originally posted - to see the
effects...but I was way to deep in ISPF - only got taken back to
ISR@PRIM - went thru 4 key lists before I got back to where I made my
change.
Can one of you explain the why/what is going on?
Many thanks
Jim Laumann
Data Center Infrastructure Services
Phone: 507-284-5510
E-mail: lauman...@mayo.edu
__________________________
Mayo Clinic
200 First Street S.W.
Rochester, MN 55905
www.mayoclinic.org
-----Original Message-----
From: ISPF discussion list [mailto:ISP...@LISTSERV.ND.EDU] On Behalf Of
Edward Jaffe
Sent: Friday, November 20, 2009 11:12 AM
To: ISP...@LISTSERV.ND.EDU
Subject: Re: TSO LOGOFF from ISPF
Rob Zenuk
only removes one session. Does anyone know of a way to close all splits in
one fell swoop? Now that would be the path to a fast logoff. For brevity
and ease of typing purposes I think I would call it
"GETTHEHECKOUTOFDODGE'... Hmmmm... Guess I should not respond on Friday's after the Margarita's
are already flowing...
Rob
In a message dated 11/20/2009 10:12:05 A.M. US Mountain Standard Tim,
Pedro Vera
I admire your benefits package.
Pedro Vera
IMS Tools
Mark Zelden
<edj...@PHOENIXSOFTWARE.COM> wrote:
>Mosley, George wrote:
>> tso tsoexec logoff;=x
>>
>> Invoking it would exit me from ISPF and complete the TSO Logoff process
>> automatically.
>>
>
>Thanks for showing me this. Old dogs love learning new tricks... :-)
>
>--
For anyone that uses Gilbert Saint-Flour's FASTPATH (CBT file 183),
you can just use "LOGOFF" from anywhere. There is also "CRASH"
that just terminates ISPF (regardless of how many logical screens
or "stacked" screens are running).
Höglund Lars
=x;logoff
-----Ursprungligt meddelande-----
Fr�n: ISPF discussion list [mailto:ISP...@LISTSERV.ND.EDU] F�r Mosley, George
Skickat: den 20 november 2009 17:34
Till: ISP...@LISTSERV.ND.EDU
�mne: TSO LOGOFF from ISPF
Paul Gilmartin
> Isn't viceversa
>
> =x;logoff
>
Nope. After "=x" ISPF isn't around to execute further commands.
> -----Ursprungligt meddelande-----
> Fr�n: ISPF discussion list [mailto:ISP...@LISTSERV.ND.EDU] F�r
> Mosley, George
> Skickat: den 20 november 2009 17:34
>
> For ages, I've had the following on a PFKey (which I think I got
> from someone on this list):
>
> tso tsoexec logoff;=x
-- gil