Multiple Porgrams Comipilation

[Thomas.Schneider.Wien]

Oh, sorry, I did forget to say:
The problem described in my *previous message* does of course *only
occur* when
*there is **ANY Error** * in the batch of needed Compilations.

*When the is NO error at all* :: Evrything is ok, of course.

*But when there are > 0 errors in any of the files to be compiled* ::

NetRexxC says for the modues, which are ok: successful compilation !

**BUT does ***NOT*** compile those modules***

... and that is my (1) complaint...

Tom.

--
Thomas Schneider Projects ReyC & LOGOS on www.KENAI.com
_______________________________________________
Ibm-netrexx mailing list
[hidden email]

[Mike Cowlishaw]

As I pointed out a long time ago, if you have a set of inter-dependent
classes then if one fails to compile then the other classes must be deleted
because after fixing the one that failed the others *must* be recompiled.

If they are not inter-dependent, then compile them separately, or a bunch
at a time.

Perhaps the message should say 'Compilation successful so far...' :-).

Mike

 

> -----Original Message-----
> From: [hidden email]
> [mailto:[hidden email]] On Behalf Of
> Thomas Schneider
> Sent: 27 June 2010 03:45
> To: IBM Netrexx
> Subject: [Ibm-netrexx] Multiple Porgrams Comipilation
>
> Oh, sorry, I did forget to say:
> The problem described in my *previous message* does of course *only
> occur* when
> *there is **ANY Error** * in the batch of needed Compilations.
>
> *When the is NO error at all* :: Evrything is ok, of course.
>
> *But when there are > 0 errors in any of the files to be compiled* ::
>
> NetRexxC says for the modues, which are ok: successful compilation !
>
> **BUT does ***NOT*** compile those modules***
>
> ... and that is my (1) complaint...
>
> Tom.
>
> --
> Thomas Schneider Projects ReyC & LOGOS on www.KENAI.com
> _______________________________________________
> Ibm-netrexx mailing list
> [hidden email]
>

_______________________________________________
Ibm-netrexx mailing list
[hidden email]

[kenner]

I'm running 4.01. Is that the latest???

Kenneth Klein
Systems Specialist
502-868-3644
859-750-5179 (Cell)
502-868-2298 (Fax)
[hidden email]
_______________________________________________
Ibm-netrexx mailing list
[hidden email]

[Thomas.Schneider.Wien]

The current *official* release is on my www.Rexx2Nrx.com home-page.

I'm currently working very hard to make the *next* release (which is now totally object oriented, and in NetRexx)
available.....

The code-name is: ReyC: The Rey Compiler.

But I'm still *not thru*, sorry to say....

ReyC will attempt to compile classic Rexx, ooRexx, and NetRexx *at once*...

But ReyC does also compile COBOL and PL/I ......

The planned commercial release date is now 1.9.2010.

When you would like to act as a beta tester, give me a mail....

Thomas Schneider.

PS: ReyC will go open source on www.kenai.com, project ReyC.

When you would like to be kept informed of the prgress, please subscribe this project ... :-)


=============================================================================
Am 29.06.2010 17:00, schrieb [hidden email]

I'm running 4.01. Is that the latest???

Kenneth Klein
Systems Specialist
502-868-3644
859-750-5179 (Cell)
502-868-2298 (Fax)
[hidden email]

_______________________________________________
Ibm-netrexx mailing list
[hidden email]

  

_______________________________________________
Ibm-netrexx mailing list
[hidden email]

[kenner]

Am I reading the doc correctly? It seems that netrexx runs on the PC and rexx2nrc runs on the PC, but rexx2nrc does not run on OMVS. What am I missing here??

YIA

Kenneth Klein
Systems Specialist
502-868-3644
859-750-5179 (Cell)
502-868-2298 (Fax)
[hidden email]

[hidden email]
Sent by: [hidden email]

06/29/2010 11:28 AM

Please respond to
IBM Netrexx <[hidden email]>

To	IBM Netrexx <[hidden email]>
cc
Subject	[Ibm-netrexx] How is the rexx2nrx conversion utility coming along?

I'm running 4.01. Is that the latest???

Kenneth Klein
Systems Specialist
502-868-3644
859-750-5179 (Cell)
502-868-2298 (Fax)
[hidden email]_______________________________________________
Ibm-netrexx mailing list
[hidden email]



_______________________________________________
Ibm-netrexx mailing list
[hidden email]

[Thomas.Schneider.Wien]

Hello Mike,
    sorry to say:

I, personally, totally *dis-agree*.

As all of my samples show, what you are doing is 100% ok, except:

Please leave the compiled classes there ...

This will aloow a step wise approach....
Otherwise, the whole approach is *frustrating* ....

*My personal point of view of course* :-)

What do the other members of this group say ??

Tom.
========================================================================

Am 29.06.2010 09:28, schrieb Mike Cowlishaw:

> As I pointed out a long time ago, if you have a set of inter-dependent
> classes then if one fails to compile then the other classes must be deleted
> because after fixing the one that failed the others *must* be recompiled.
>
> If they are not inter-dependent, then compile them separately, or a bunch
> at a time.
>
> Perhaps the message should say 'Compilation successful so far...' :-).
>
> Mike
>
>
>
>    
>> -----Original Message-----
>> From: [hidden email]
>> [mailto:[hidden email]] On Behalf Of
>> Thomas Schneider
>> Sent: 27 June 2010 03:45
>> To: IBM Netrexx
>> Subject: [Ibm-netrexx] Multiple Porgrams Comipilation
>>
>> Oh, sorry, I did forget to say:
>> The problem described in my *previous message* does of course *only
>> occur* when
>> *there is **ANY Error** * in the batch of needed Compilations.
>>
>> *When the is NO error at all* :: Evrything is ok, of course.
>>
>> *But when there are>  0 errors in any of the files to be compiled* ::
>>
>> NetRexxC says for the modues, which are ok: successful compilation !
>>
>> **BUT does ***NOT*** compile those modules***
>>
>> ... and that is my (1) complaint...
>>
>> Tom.
>>
>> --
>> Thomas Schneider Projects ReyC&  LOGOS on www.KENAI.com
>> _______________________________________________
>> Ibm-netrexx mailing list
>> [hidden email]
>>
>>      
>
>    

--
Thomas Schneider Projects ReyC & LOGOS on www.KENAI.com
_______________________________________________
Ibm-netrexx mailing list
[hidden email]

[Thomas.Schneider.Wien]

As far as I know (although I did *never* try it) :

As Rexx2Nrx is delivered as a set of Java Classes (in the Rexx2Nrx.jar) it *should* run on OMVS :-) :-)

Long long time ago, a student working for IBM used it on Unix, wothout *no problems* :-)

One thing I've changed are the work file names:

They are currently (on www.rexx2Nrx.com) containing a '$' in the temporary File Tpes, which has
another meaning on Unix Systems (didn't know, years ago ..)

But that's the only open issue I am aware of of the *official release* (on www.Rexx2Nrx.com)

Tom.
=========================================================================
Am 29.06.2010 21:40, schrieb [hidden email]

Am I reading the doc correctly? It seems that netrexx runs on the PC and rexx2nrc runs on the PC, but rexx2nrc does not run on OMVS. What am I missing here??

YIA

Kenneth Klein
Systems Specialist
502-868-3644
859-750-5179 (Cell)
502-868-2298 (Fax)
[hidden email]

[hidden email]
Sent by: [hidden email]

06/29/2010 11:28 AM

Please respond to
IBM Netrexx [hidden email]

To	IBM Netrexx [hidden email]
cc
Subject	[Ibm-netrexx] How is the rexx2nrx conversion utility coming along?

I'm running 4.01. Is that the latest???

Kenneth Klein
Systems Specialist
502-868-3644
859-750-5179 (Cell)
502-868-2298 (Fax)
[hidden email]_______________________________________________
Ibm-netrexx mailing list
[hidden email]

_______________________________________________
Ibm-netrexx mailing list
[hidden email]

  

_______________________________________________
Ibm-netrexx mailing list
[hidden email]

[Fernando Cassia-2]

On Tue, Jun 29, 2010 at 4:39 PM, Thomas Schneider <[hidden email]> wrote:

ReyC will attempt to compile classic Rexx, ooRexx, and NetRexx *at once*...

But ReyC does also compile COBOL and PL/I ......

The planned commercial release date is now 1.9.2010.

Why not open source it now, as is, and then keep improving it until you reach your "milestone" release?.

Lots of open source project started being half-finished projects, ie Mozilla "milestone" releases then versions 0.x
until finally 1.0 was reached.

Just my $0.02...
FC

_______________________________________________
Ibm-netrexx mailing list
[hidden email]

[Thomas.Schneider.Wien]

May I ask the ibm-netrexx admin to pass this message forward, please....

Should give some answers to the questions raised below ...
Tom.
=======================================================================
Am 29.06.2010 21:39, schrieb Thomas Schneider:

The current *official* release is on my www.Rexx2Nrx.com home-page.

I'm currently working very hard to make the *next* release (which is now totally object oriented, and in NetRexx)
available.....

The code-name is: ReyC: The Rey Compiler.

But I'm still *not thru*, sorry to say....

ReyC will attempt to compile classic Rexx, ooRexx, and NetRexx *at once*...

But ReyC does also compile COBOL and PL/I ......

The planned commercial release date is now 1.9.2010.

When you would like to act as a beta tester, give me a mail....

Thomas Schneider.

PS: ReyC will go open source on www.kenai.com, project ReyC.

When you would like to be kept informed of the prgress, please subscribe this project ... :-)


=============================================================================
Am 29.06.2010 17:00, schrieb [hidden email]

I'm running 4.01. Is that the latest???

Kenneth Klein
Systems Specialist
502-868-3644
859-750-5179 (Cell)
502-868-2298 (Fax)
[hidden email]

_______________________________________________
Ibm-netrexx mailing list
[hidden email]

  

_______________________________________________
Ibm-netrexx mailing list
[hidden email]

  

_______________________________________________
Ibm-netrexx mailing list
[hidden email]

[Robert L Hamilton]

Jan 9th or sept 1st

?????

BobH

Sent from my iPhone

Enjoy your day

Bob Hamilton

On Jun 29, 2010, at 2:50 PM, Fernando Cassia <[hidden email]> wrote:

On Tue, Jun 29, 2010 at 4:39 PM, Thomas Schneider <[hidden email]> wrote:

ReyC will attempt to compile classic Rexx, ooRexx, and NetRexx *at once*...

But ReyC does also compile COBOL and PL/I ......

The planned commercial release date is now 1.9.2010.

Why not open source it now, as is, and then keep improving it until you reach your "milestone" release?.

Lots of open source project started being half-finished projects, ie Mozilla "milestone" releases then versions 0.x
until finally 1.0 was reached.

Just my $0.02...
FC

_______________________________________________
Ibm-netrexx mailing list
[hidden email]

_______________________________________________
Ibm-netrexx mailing list
[hidden email]

[Robert L Hamilton]

Scratch that

Bobh

Sent from my iPhone

Enjoy your day

Bob Hamilton

On Jun 29, 2010, at 2:50 PM, Fernando Cassia <[hidden email]> wrote:

On Tue, Jun 29, 2010 at 4:39 PM, Thomas Schneider <[hidden email]> wrote:

ReyC will attempt to compile classic Rexx, ooRexx, and NetRexx *at once*...

But ReyC does also compile COBOL and PL/I ......

The planned commercial release date is now 1.9.2010.

Why not open source it now, as is, and then keep improving it until you reach your "milestone" release?.

Lots of open source project started being half-finished projects, ie Mozilla "milestone" releases then versions 0.x
until finally 1.0 was reached.

Just my $0.02...
FC

_______________________________________________
Ibm-netrexx mailing list
[hidden email]

_______________________________________________
Ibm-netrexx mailing list
[hidden email]

[Aviatrexx]

Tom,

I think it boils down to the relationship (if any) between all the
NetRexx routines one is compiling.  Either you have a lot of
inter-dependent routines, or they are all separate entities and are
being batch-compiled mainly for convenience.  In my experience, the
latter is rarely the case.

You could design a "build" process around NetRexxC that is smart
enough to know the specific interdependencies of all the routines,
e.g. RoutineX is needed by RoutineA and RoutineC, but not RoutineB.
With that information, you could safely let the build process decide
which successfully-compiled class files could be saved, and which
should be re-compiled due to an unsuccessful compilation of RoutineX.

Alternatively, you could throw together a simple Rexx program to
sequentially compile and save all of your routines independently, then
recompile only those with errors (or their dependents) by hand.

It seems to me that NetRexxC, given the information available to it,
is doing the right thing by not leaving dependent class files lying
around after an unsuccessful compilation.

Now if you want to write LNKEDIT and implement WXTRNs, have at it...

-Chip-

On 6/29/10 19:43 Thomas Schneider said:

> Hello Mike,
>    sorry to say:
>
> I, personally, totally *dis-agree*.
>
> As all of my samples show, what you are doing is 100% ok, except:
>
> Please leave the compiled classes there ...
>
> This will aloow a step wise approach....
> Otherwise, the whole approach is *frustrating* ....
>
> *My personal point of view of course* :-)
>
> What do the other members of this group say ??
>
> Tom.
> ========================================================================
>
> Am 29.06.2010 09:28, schrieb Mike Cowlishaw:
>> As I pointed out a long time ago, if you have a set of inter-dependent
>> classes then if one fails to compile then the other classes must be
>> deleted
>> because after fixing the one that failed the others *must* be recompiled.
>>
>> If they are not inter-dependent, then compile them separately, or a bunch
>> at a time.
>>
>> Perhaps the message should say 'Compilation successful so far...' :-).
>>
>> Mike
>>
>>
>>
>>  
>>> -----Original Message-----
>>> From: [hidden email]
>>> [mailto:[hidden email]] On Behalf Of
>>> Thomas Schneider
>>> Sent: 27 June 2010 03:45
>>> To: IBM Netrexx
>>> Subject: [Ibm-netrexx] Multiple Porgrams Comipilation
>>>
>>> Oh, sorry, I did forget to say:
>>> The problem described in my *previous message* does of course *only
>>> occur* when
>>> *there is **ANY Error** * in the batch of needed Compilations.
>>>
>>> *When the is NO error at all* :: Evrything is ok, of course.
>>>
>>> *But when there are>  0 errors in any of the files to be compiled* ::
>>>
>>> NetRexxC says for the modues, which are ok: successful compilation !
>>>
>>> **BUT does ***NOT*** compile those modules***
>>>
>>> ... and that is my (1) complaint...
>>>
>>> Tom.
>>>
>>> --
>>> Thomas Schneider Projects ReyC&  LOGOS on www.KENAI.com
>>> _______________________________________________
>>> Ibm-netrexx mailing list
>>> [hidden email]
>>>
>>>      
>>
>>    
>
>

_______________________________________________
Ibm-netrexx mailing list
[hidden email]

[Thomas.Schneider.Wien]

Hello Chip and Mike,
    *I*, personally, can *live*with the way NetRexxC is currently behaving !

I do have, however, another *standpoint* (meaning) than Mike does have.

When the *compilation is successful*, all the internal information to
*make* it *successful* was and *is available*.

Hence, no matter to *NOT save the classes* just successfully compiled :-)

Full Stop (from my point).

I do *not* like to repeat those different points of view between Mike
and myself ...

He did such a *great job* :-)

Thanks, Mike, again :-)
Tom.
=========================================================================

Am 29.06.2010 22:46, schrieb Chip Davis:

> Tom,
>
> I think it boils down to the relationship (if any) between all the
> NetRexx routines one is compiling.  Either you have a lot of
> inter-dependent routines, or they are all separate entities and are
> being batch-compiled mainly for convenience.  In my experience, the
> latter is rarely the case.
>
> You could design a "build" process around NetRexxC that is smart
> enough to know the specific interdependencies of all the routines,
> e.g. RoutineX is needed by RoutineA and RoutineC, but not RoutineB.
> With that information, you could safely let the build process decide
> which successfully-compiled class files could be saved, and which
> should be re-compiled due to an unsuccessful compilation of RoutineX.
>
> Alternatively, you could throw together a simple Rexx program to
> sequentially compile and save all of your routines independently, then
> recompile only those with errors (or their dependents) by hand.
>
> It seems to me that NetRexxC, given the information available to it,
> is doing the right thing by not leaving dependent class files lying
> around after an unsuccessful compilation.
>
> Now if you want to write LNKEDIT and implement WXTRNs, have at it...
>
> -Chip-
>
> On 6/29/10 19:43 Thomas Schneider said:
>> Hello Mike,
>>    sorry to say:
>>
>> I, personally, totally *dis-agree*.
>>
>> As all of my samples show, what you are doing is 100% ok, except:
>>
>> Please leave the compiled classes there ...
>>
>> This will aloow a step wise approach....
>> Otherwise, the whole approach is *frustrating* ....
>>
>> *My personal point of view of course* :-)
>>
>> What do the other members of this group say ??
>>
>> Tom.
>> ========================================================================
>>
>> Am 29.06.2010 09:28, schrieb Mike Cowlishaw:
>>> As I pointed out a long time ago, if you have a set of inter-dependent
>>> classes then if one fails to compile then the other classes must be
>>> deleted
>>> because after fixing the one that failed the others *must* be
>>> recompiled.
>>>
>>> If they are not inter-dependent, then compile them separately, or a
>>> bunch
>>> at a time.
>>>
>>> Perhaps the message should say 'Compilation successful so far...' :-).
>>>
>>> Mike
>>>
>>>
>>>
>>>> -----Original Message-----
>>>> From: [hidden email]
>>>> [mailto:[hidden email]] On Behalf Of
>>>> Thomas Schneider
>>>> Sent: 27 June 2010 03:45
>>>> To: IBM Netrexx
>>>> Subject: [Ibm-netrexx] Multiple Porgrams Comipilation
>>>>
>>>> Oh, sorry, I did forget to say:
>>>> The problem described in my *previous message* does of course *only
>>>> occur* when
>>>> *there is **ANY Error** * in the batch of needed Compilations.
>>>>
>>>> *When the is NO error at all* :: Evrything is ok, of course.
>>>>
>>>> *But when there are>  0 errors in any of the files to be compiled* ::
>>>>
>>>> NetRexxC says for the modues, which are ok: successful compilation !
>>>>
>>>> **BUT does ***NOT*** compile those modules***
>>>>
>>>> ... and that is my (1) complaint...
>>>>
>>>> Tom.
>>>>
>>>> --
>>>> Thomas Schneider Projects ReyC&  LOGOS on www.KENAI.com
>>>> _______________________________________________
>>>> Ibm-netrexx mailing list
>>>> [hidden email]
>>>>
>>>
>>
>>
> _______________________________________________
> Ibm-netrexx mailing list
> [hidden email]
>
>

--
Thomas Schneider Projects ReyC & LOGOS on www.KENAI.com
_______________________________________________
Ibm-netrexx mailing list
[hidden email]

[Thomas.Schneider.Wien]

ReyC is still not ready for deployment, as there are still some *bugs* in my Parser....

Sorry to say,
Thomas.

PS: Be sure, when I did solve all the *bugs*, I'll notify all of you, for sure...

Tom.
=================================================================
Am 29.06.2010 22:00, schrieb [hidden email]

Jan 9th or sept 1st

?????

BobH

Sent from my iPhone

Enjoy your day

Bob Hamilton

On Jun 29, 2010, at 2:50 PM, Fernando Cassia <[hidden email]> wrote:

On Tue, Jun 29, 2010 at 4:39 PM, Thomas Schneider <[hidden email]> wrote:

ReyC will attempt to compile classic Rexx, ooRexx, and NetRexx *at once*...

But ReyC does also compile COBOL and PL/I ......

The planned commercial release date is now 1.9.2010.

Why not open source it now, as is, and then keep improving it until you reach your "milestone" release?.

Lots of open source project started being half-finished projects, ie Mozilla "milestone" releases then versions 0.x
until finally 1.0 was reached.

Just my $0.02...
FC

_______________________________________________
Ibm-netrexx mailing list
[hidden email]

_______________________________________________
Ibm-netrexx mailing list
[hidden email]

  

_______________________________________________
Ibm-netrexx mailing list
[hidden email]

[Thomas.Schneider.Wien]

Hello Chip,

     I don't yet know why I currently do have 461 incoming mails..... ???

I did already resolve my NetRexx problems I did have some time ago, and
are currently trying to finish my art (ReyC)
Frankly speaking, I'm still bothered with a couple of points by porting
Rexx2Nrx so many times ...

Does have nothing to do with NetRexx, and/or Java, but obviously with
simply my *ignorance*.

I'm, however, confindent, that I can resolve the *open issues*, as I  
call them,
in a reasonable time....

Tom.
==========================================================================================


Am 29.06.2010 22:46, schrieb Chip Davis:

> Tom,
>
> I think it boils down to the relationship (if any) between all the
> NetRexx routines one is compiling.  Either you have a lot of
> inter-dependent routines, or they are all separate entities and are
> being batch-compiled mainly for convenience.  In my experience, the
> latter is rarely the case.
>
> You could design a "build" process around NetRexxC that is smart
> enough to know the specific interdependencies of all the routines,
> e.g. RoutineX is needed by RoutineA and RoutineC, but not RoutineB.
> With that information, you could safely let the build process decide
> which successfully-compiled class files could be saved, and which
> should be re-compiled due to an unsuccessful compilation of RoutineX.
>
> Alternatively, you could throw together a simple Rexx program to
> sequentially compile and save all of your routines independently, then
> recompile only those with errors (or their dependents) by hand.
>
> It seems to me that NetRexxC, given the information available to it,
> is doing the right thing by not leaving dependent class files lying
> around after an unsuccessful compilation.
>
> Now if you want to write LNKEDIT and implement WXTRNs, have at it...
>
> -Chip-
>
> On 6/29/10 19:43 Thomas Schneider said:
>> Hello Mike,
>>    sorry to say:
>>
>> I, personally, totally *dis-agree*.
>>
>> As all of my samples show, what you are doing is 100% ok, except:
>>
>> Please leave the compiled classes there ...
>>
>> This will aloow a step wise approach....
>> Otherwise, the whole approach is *frustrating* ....
>>
>> *My personal point of view of course* :-)
>>
>> What do the other members of this group say ??
>>
>> Tom.
>> ========================================================================
>>
>> Am 29.06.2010 09:28, schrieb Mike Cowlishaw:
>>> As I pointed out a long time ago, if you have a set of inter-dependent
>>> classes then if one fails to compile then the other classes must be
>>> deleted
>>> because after fixing the one that failed the others *must* be
>>> recompiled.
>>>
>>> If they are not inter-dependent, then compile them separately, or a
>>> bunch
>>> at a time.
>>>
>>> Perhaps the message should say 'Compilation successful so far...' :-).
>>>
>>> Mike
>>>
>>>
>>>
>>>> -----Original Message-----
>>>> From: [hidden email]
>>>> [mailto:[hidden email]] On Behalf Of
>>>> Thomas Schneider
>>>> Sent: 27 June 2010 03:45
>>>> To: IBM Netrexx
>>>> Subject: [Ibm-netrexx] Multiple Porgrams Comipilation
>>>>
>>>> Oh, sorry, I did forget to say:
>>>> The problem described in my *previous message* does of course *only
>>>> occur* when
>>>> *there is **ANY Error** * in the batch of needed Compilations.
>>>>
>>>> *When the is NO error at all* :: Evrything is ok, of course.
>>>>
>>>> *But when there are>  0 errors in any of the files to be compiled* ::
>>>>
>>>> NetRexxC says for the modues, which are ok: successful compilation !
>>>>
>>>> **BUT does ***NOT*** compile those modules***
>>>>
>>>> ... and that is my (1) complaint...
>>>>
>>>> Tom.
>>>>
>>>> --
>>>> Thomas Schneider Projects ReyC&  LOGOS on www.KENAI.com
>>>> _______________________________________________
>>>> Ibm-netrexx mailing list
>>>> [hidden email]
>>>>
>>>
>>
>>
> _______________________________________________
> Ibm-netrexx mailing list
> [hidden email]
>
>

--
Thomas Schneider Projects ReyC & LOGOS on www.KENAI.com
_______________________________________________
Ibm-netrexx mailing list
[hidden email]