Charles,
Do you get fed up with having to specify the pesky tr1 namespace? It
makes it difficult to share code between z/OS and distributed. You can
work around that with a useful namespace declaration.
Also, if you are using LANGLVL(EXTENDED0X) you can significantly cut
down on the verbosity
On 9/04/2018 8:51 AM, Andrew Rowley wrote:
On 9/04/2018 4:09 AM, Charles Mills wrote:
#include
#define __IBMCPP_TR1__ 1
#include
class myRegex
{
public:
std::tr1::regex regexObject;
};
int main(int argc, char* argv[])
{
printf("RegEx test 4/6/2018\n");
On 9/04/2018 4:09 AM, Charles Mills wrote:
#include
#define __IBMCPP_TR1__ 1
#include
class myRegex
{
public:
std::tr1::regex regexObject;
};
int main(int argc, char* argv[])
{
printf("RegEx test 4/6/2018\n");
std::tr1::regex::flag_type flags = std::tr1::regex::extended;
Edward,
Here's the basic process:
1. Order and receive physical DVD media containing the software products,
via postal mail or courier.
2. Insert the DVD into the drive of any machine that your designated z/OS
LPAR can reach. (Not necessarily *every* z/OS LPAR.) That machine might be
the HMC
On 9/04/2018 3:18 AM, Jack J. Woehr wrote:
On 4/8/2018 1:04 PM, Charles Mills wrote:
It do. The error message is ABEND=S0C4 U REASON=0004.
Try initializing the regex in the declaration instead of doing it in
main() via pointer-to-reference;
That's not a pointer-to-reference. A
Works ok for me. The code is sound so I would open a PMR.
On 9/04/2018 2:09 AM, Charles Mills wrote:
#include
#define __IBMCPP_TR1__ 1
#include
class myRegex
{
public:
std::tr1::regex regexObject;
};
int main(int argc, char* argv[])
{
printf("RegEx test 4/6/2018\n");
On 4/8/2018 2:58 PM, Charles Mills wrote:
The pattern is supplied at run time -- as is typical of regex implementations
-- so no number of subclasses solves the problem.
Put the flags in the same scope as the regex and naybe it doesn't blow up.
Either forget about your RegexObject and just
On 4/8/2018 2:58 PM, Charles Mills wrote:
I firmly believe this is a bug, and I do have a workaround, l
I'll betcha it's not a bug, it's something about the flags being out of
scope before the dtor is called for the object.
C++ is, to put it mildly, design-heavy.
You can prove it if you
I'm not "messing" -- .assign() is a documented, public method.
http://www.cplusplus.com/reference/regex/basic_regex/assign/
The pattern is supplied at run time -- as is typical of regex implementations
-- so no number of subclasses solves the problem.
It would conceivably be possible to pass
On 4/8/2018 1:43 PM, Charles Mills wrote:
Thanks, but won't work well in the real code.
Besides, the C++ way to do that is to have subclasses with increasingly
specific ctors for your needs, not mess with a template's internal data
in main().
Betcha if you work on the class design a little
On 4/8/2018 1:43 PM, Charles Mills wrote:
Thanks, but won't work well in the real code.
I just meant as a debugging step. Make it work, then make it break again
and you're home.
--
Jack J. Woehr # Science is more than a body of knowledge. It's a way of
www.well.com/~jax # thinking, a
Thanks, but won't work well in the real code.
I have multiple instantiations of the same class, and don't know the pattern
until well after the instantiation. Pattern may change after instantiation.
It doesn't take much to make it work. The std::tr1::regex::flag_type flags =
On 4/8/2018 1:04 PM, Charles Mills wrote:
It do. The error message is ABEND=S0C4 U REASON=0004.
Try initializing the regex in the declaration instead of doing it in
main() via pointer-to-reference;
--
Jack J. Woehr # Science is more than a body of knowledge. It's a way of
It do. The error message is ABEND=S0C4 U REASON=0004.
Traceback:
DSA Entry E Offset Statement Load Mod Program Unit
Service Status
1
On 4/8/2018 12:09 PM, Charles Mills wrote:
Believe it or not, it appears to be a bug in the C++ runtime. Anyone who
wants to prove me wrong is welcome to try the below. I have tested only on
z/OS V2R2.
No convenient access, but let me guess: blows up in the delete?
Got an error message?
--
Doing more with less
By Trevor Eddolls
Sunday, April 08, 2018
There’s been a mantra in management circles that mainframes don’t need that
many staff. The heady days of the 1980s seem like a different world when a
dozen operators would come to work for each shift and there were rooms full
of
Believe it or not, it appears to be a bug in the C++ runtime. Anyone who
wants to prove me wrong is welcome to try the below. I have tested only on
z/OS V2R2.
#include
#define __IBMCPP_TR1__ 1
#include
class myRegex
{
public:
std::tr1::regex regexObject;
};
int main(int argc, char*
Unfortunately mostly in german:
https://www.youtube.com/watch?v=TERZ2DIp4hY
-Ursprüngliche Nachricht-
Von: IBM Mainframe Discussion List Im Auftrag von
William Donzelli
Gesendet: 08 April, 2018 17:16
An: IBM-MAIN@LISTSERV.UA.EDU
Betreff: Re: Mainframe videos?
For the break:
https://www.youtube.com/watch?v=RowwNXKEt4k
A good video to close with:
https://www.youtube.com/watch?v=63ot5h08rKg
--
Will
On Sun, Apr 8, 2018 at 8:01 AM, Mike Beer wrote:
> Hi,
>
> TAPES:
> UNIVAC TAPE: https://www.youtube.com/watch?v=-KuoZ6cades
> Overland:
Look through some of Marc Verdiell's videos. He's got a bunch on the 1401 that
the Computer History Museum runs.
https://www.youtube.com/user/mverdiell
From: Mike Beer
Sent: Sunday, 8 April 2018 7:01 am
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: AW:
The environments in which (mis)use of userkey common occurs make tracking
via generic tracker generally not possible.
It has been possible to reject these situations for well over 10 years.
And much of z/OS testing does exactly that.
The PER IF SLIP trap approach that is provided is a very
Hi,
TAPES:
UNIVAC TAPE: https://www.youtube.com/watch?v=-KuoZ6cades
Overland: https://www.youtube.com/watch?v=eCBxNhEzIfc
https://www.youtube.com/watch?v=CEL8wnW5uvs
1401: https://www.youtube.com/watch?v=_O-UaHAWVKU
After 4:30: https://www.youtube.com/watch?v=wIjgZhAjQS4
PUNCHED CARDS:
Hello
I am currently tasked with presenting a series of talks on the general
subject of 'mainframes' from a historical perspective.
I can find plenty of pictures and movies of 360 front panels, various disk
drive models and printers, but try as I might, I have been unable to track
down any
Should be easier. I used Omegamon to display that. We have just one lonely
user of it, that we are working to retire.
_
Dave Jousma
Manager Mainframe Engineering, Assistant Vice President
david.jou...@53.com
1830 East Paris,
24 matches
Mail list logo