ibm-netrexx

First rework

Classic

List

Threaded

6 messages Options

Patric Bechtel

First rework

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi all,

I just uploaded a first rework of the NetRexx sources I got last week:
http://kenai.com/jira/browse/NETREXX-10

I invested quite some time to rework the code in a first step to fix
some issues with the code, which is:

- - strictcase, strictargs and strictsignal IMHO is required for
production code (done)
- - make API more "java", using camel case functions, ALL_CAPS for
constants and such (started, TBD)
- - increase compiler performance e.g. using StringBuilder for code
generation, more caching of classes and such (TBD)
- - decoupling the code, properly separating status keeping, parsing,
dependency resolution and translation process from each other. It would
be nice if we could put these into separate packages even. (TBD)
- - create external API to enable efficient ant target (maybe integrate
one into NetRexxC.jar?), reusing caches etc (TBD)
- - start implementing some low hanging targets (detailed discussion about
this should start *after* being done with the above issues), like loop
over collections, annotations, transparent BigDecimal handling etc (TBD)

If we can agree upon this, I could create the topic above into separate
issues, so detail discussions can go on in JIRA, then.

Comments?

- --
cu, Patric
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: GnuPT 2.5.2

iEYEARECAAYFAk3/mm0ACgkQfGgGu8y7ypDaBQCeKlbRV+r0O+V01UbBQTcB6cnQ
Zj0AoJtLk9j8a4NU5tqWg/LvmJLIpNvH
=nSlk
-----END PGP SIGNATURE-----
_______________________________________________
Ibm-netrexx mailing list
[hidden email]
Online Archive : http://ibm-netrexx.215625.n3.nabble.com/

Kermit Kiser

Re: First rework

Hi Patric -

It sounds like you have done some excellent work.

I am not sure what you want to agree on, but I think it would be a very
good idea to separate your list into individual issues so that we can
discuss them and vote on them separately. (I realize that is more work
for you :-( ) I would like to know more about some of the things you
are working on.

One thing that I would vote against is changing the defaults for
existing compile options as that could break some compiles that people
currently use. It might be better to custom set an environment variable
at install time to modify compiler defaults.

From your issue post on Kenai, it sounds like You and Bill and I have
all created different solutions to the Ant compile problem so there is
already a plethora of ideas to choose from there!

-- Kermit

On 6/20/2011 12:07 PM, Patric Bechtel wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hi all,
>
> I just uploaded a first rework of the NetRexx sources I got last week:
> http://kenai.com/jira/browse/NETREXX-10
>
> I invested quite some time to rework the code in a first step to fix
> some issues with the code, which is:
>
> - - strictcase, strictargs and strictsignal IMHO is required for
> production code (done)
> - - make API more "java", using camel case functions, ALL_CAPS for
> constants and such (started, TBD)
> - - increase compiler performance e.g. using StringBuilder for code
> generation, more caching of classes and such (TBD)
> - - decoupling the code, properly separating status keeping, parsing,
> dependency resolution and translation process from each other. It would
> be nice if we could put these into separate packages even. (TBD)
> - - create external API to enable efficient ant target (maybe integrate
> one into NetRexxC.jar?), reusing caches etc (TBD)
> - - start implementing some low hanging targets (detailed discussion about
> this should start *after* being done with the above issues), like loop
> over collections, annotations, transparent BigDecimal handling etc (TBD)
>
> If we can agree upon this, I could create the topic above into separate
> issues, so detail discussions can go on in JIRA, then.
>
> Comments?
>
> - --
> cu, Patric
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.10 (GNU/Linux)
> Comment: GnuPT 2.5.2
>
> iEYEARECAAYFAk3/mm0ACgkQfGgGu8y7ypDaBQCeKlbRV+r0O+V01UbBQTcB6cnQ
> Zj0AoJtLk9j8a4NU5tqWg/LvmJLIpNvH
> =nSlk
> -----END PGP SIGNATURE-----
> _______________________________________________
> Ibm-netrexx mailing list
> [hidden email]
> Online Archive : http://ibm-netrexx.215625.n3.nabble.com/
>
>
>

_______________________________________________
Ibm-netrexx mailing list
[hidden email]
Online Archive : http://ibm-netrexx.215625.n3.nabble.com/

rvjansen

Re: First rework

In reply to this post by Patric Bechtel

Hi Patric,

ok, cool - thanks for all this work. it is a good thing you are on the ARB, which still must convene. I'll send out invites this week so we can make a start with this; we still must discuss and take important decisions around which language level is the lower threshold etc - this might have consequences for some of these fixes. If we can agree in the ARB, it might be better to have you as a committer so you can slowly work these in trunk; also, let's continue this at the developer list, so we can have ibm-netrexx as a user list?

strict~everything: you will have no discussion with me about this
camel case etc - I was planning to keep MFC style throughout the codebase, but I am also fond of (lower)camelcase. but I also put spaces around equal signs and aligh these up
stringbuilder - that is 1.5; not sure if everyone will agree on that
decoupling - no qualms with me
external api - this is where I insist we design first and implement after ARB decision
low hanging fruit: yes, but ditto

good to have you on board! I was wondering already,

For the ARB, I would like to have a biweekly or monthly meeting over video, I have a webex for this. Mike, who was voted Architect ad interim in my proposal to the RexxLA board, might not be able to be there every time, but I would like to run things by him as much as possible.

so these are my first comments, let's move over to 'developer'

best regards,

René Jansen.

On 20 jun 2011, at 21:07, Patric Bechtel wrote:

_______________________________________________
Ibm-netrexx mailing list
[hidden email]
Online Archive : http://ibm-netrexx.215625.n3.nabble.com/

measel

RE: Mint Version

Asking again that there please be a "mint" version with no additional cheese whiz.

I've used this for over 10 years with no complaints and would like to continue with same.

-----Original Message-----
From: [hidden email] [mailto:[hidden email]] On Behalf Of René Jansen
Sent: Monday, June 20, 2011 3:56 PM
To: [hidden email]; IBM Netrexx
Subject: Re: [Ibm-netrexx] First rework

Hi Patric,

ok, cool - thanks for all this work. it is a good thing you are on the ARB, which still must convene. I'll send out invites this week so we can make a start with this; we still must discuss and take important decisions around which language level is the lower threshold etc - this might have consequences for some of these fixes. If we can agree in the ARB, it might be better to have you as a committer so you can slowly work these in trunk; also, let's continue this at the developer list, so we can have ibm-netrexx as a user list?

strict~everything: you will have no discussion with me about this
camel case etc - I was planning to keep MFC style throughout the codebase, but I am also fond of (lower)camelcase. but I also put spaces around equal signs and aligh these up
stringbuilder - that is 1.5; not sure if everyone will agree on that
decoupling - no qualms with me
external api - this is where I insist we design first and implement after ARB decision
low hanging fruit: yes, but ditto

good to have you on board! I was wondering already,

For the ARB, I would like to have a biweekly or monthly meeting over video, I have a webex for this. Mike, who was voted Architect ad interim in my proposal to the RexxLA board, might not be able to be there every time, but I would like to run things by him as much as possible.

so these are my first comments, let's move over to 'developer'

best regards,

René Jansen.

On 20 jun 2011, at 21:07, Patric Bechtel wrote:

_______________________________________________
Ibm-netrexx mailing list
[hidden email]
Online Archive : http://ibm-netrexx.215625.n3.nabble.com/

_______________________________________________
Ibm-netrexx mailing list
[hidden email]
Online Archive : http://ibm-netrexx.215625.n3.nabble.com/

Patric Bechtel

Re: Mint Version

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Measel, Mike schrieb am 20.06.2011 23:46:
> Asking again that there please be a "mint" version with no additional cheese whiz.
>
> I've used this for over 10 years with no complaints and would like to continue with same.

I've been using it even longer, and believe me, I'm not going to add
cheese whiz ;-)

As this is the user list, may I ask,
- - which JVM(s) you're using currently?
- - which are the JVM(s) you're going to use a future version of NetRexx
upon?

Just want to know wether a Java5 version of NetRexx might be possible.
Maybe even Java6...

There's quite some improvements made to the JVM in later versions, which
would make NetRexx code smaller and faster; speaking of a 'mint'
version, fresher ;-)

- --
cu, Patric
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: GnuPT 2.5.2

iEYEARECAAYFAk4AaTEACgkQfGgGu8y7ypDymACfeGQJs6MEQU8w9UeifHlcXuLu
ZqEAoIqBaYrLrBv8t+rr7bMEkWvTOBYw
=D9ll
-----END PGP SIGNATURE-----
_______________________________________________
Ibm-netrexx mailing list
[hidden email]
Online Archive : http://ibm-netrexx.215625.n3.nabble.com/

rvjansen

Re: Mint Version

In reply to this post by measel

Mike,

exactly, this is our aim - for everyone who is happy with NetRexx and how it has worked for 15 years, nothing should change (except everyone being happier still).
This is the reason we should move this type of discussion to the developers list.

best regards,

René.

On Jun 20, 2011, at 11:46 PM, Measel, Mike wrote:

> Asking again that there please be a "mint" version with no additional cheese whiz.
>
> I've used this for over 10 years with no complaints and would like to continue with same.
>
> -----Original Message-----
> From: [hidden email] [mailto:[hidden email]] On Behalf Of René Jansen
> Sent: Monday, June 20, 2011 3:56 PM
> To: [hidden email]; IBM Netrexx
> Subject: Re: [Ibm-netrexx] First rework
>
> Hi Patric,
>
> ok, cool - thanks for all this work. it is a good thing you are on the ARB, which still must convene. I'll send out invites this week so we can make a start with this; we still must discuss and take important decisions around which language level is the lower threshold etc - this might have consequences for some of these fixes. If we can agree in the ARB, it might be better to have you as a committer so you can slowly work these in trunk; also, let's continue this at the developer list, so we can have ibm-netrexx as a user list?
>
> strict~everything: you will have no discussion with me about this
> camel case etc - I was planning to keep MFC style throughout the codebase, but I am also fond of (lower)camelcase. but I also put spaces around equal signs and aligh these up
> stringbuilder - that is 1.5; not sure if everyone will agree on that
> decoupling - no qualms with me
> external api - this is where I insist we design first and implement after ARB decision
> low hanging fruit: yes, but ditto
>
> good to have you on board! I was wondering already,
>
> For the ARB, I would like to have a biweekly or monthly meeting over video, I have a webex for this. Mike, who was voted Architect ad interim in my proposal to the RexxLA board, might not be able to be there every time, but I would like to run things by him as much as possible.
>
> so these are my first comments, let's move over to 'developer'
>
> best regards,
>
> René Jansen.
>
> On 20 jun 2011, at 21:07, Patric Bechtel wrote:
>
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA1
>>
>> Hi all,
>>
>> I just uploaded a first rework of the NetRexx sources I got last week:
>> http://kenai.com/jira/browse/NETREXX-10
>>
>> I invested quite some time to rework the code in a first step to fix
>> some issues with the code, which is:
>>
>> - - strictcase, strictargs and strictsignal IMHO is required for
>> production code (done)
>> - - make API more "java", using camel case functions, ALL_CAPS for
>> constants and such (started, TBD)
>> - - increase compiler performance e.g. using StringBuilder for code
>> generation, more caching of classes and such (TBD)
>> - - decoupling the code, properly separating status keeping, parsing,
>> dependency resolution and translation process from each other. It would
>> be nice if we could put these into separate packages even. (TBD)
>> - - create external API to enable efficient ant target (maybe integrate
>> one into NetRexxC.jar?), reusing caches etc (TBD)
>> - - start implementing some low hanging targets (detailed discussion about
>> this should start *after* being done with the above issues), like loop
>> over collections, annotations, transparent BigDecimal handling etc (TBD)
>>
>> If we can agree upon this, I could create the topic above into separate
>> issues, so detail discussions can go on in JIRA, then.
>>
>> Comments?
>>
>> - --
>> cu, Patric
>> -----BEGIN PGP SIGNATURE-----
>> Version: GnuPG v1.4.10 (GNU/Linux)
>> Comment: GnuPT 2.5.2
>>
>> iEYEARECAAYFAk3/mm0ACgkQfGgGu8y7ypDaBQCeKlbRV+r0O+V01UbBQTcB6cnQ
>> Zj0AoJtLk9j8a4NU5tqWg/LvmJLIpNvH
>> =nSlk
>> -----END PGP SIGNATURE-----
>> _______________________________________________
>> Ibm-netrexx mailing list
>> [hidden email]
>> Online Archive : http://ibm-netrexx.215625.n3.nabble.com/
>>
>
>
> _______________________________________________
> Ibm-netrexx mailing list
> [hidden email]
> Online Archive : http://ibm-netrexx.215625.n3.nabble.com/
>
>
> _______________________________________________
> Ibm-netrexx mailing list
> [hidden email]
> Online Archive : http://ibm-netrexx.215625.n3.nabble.com/
>

_______________________________________________
Ibm-netrexx mailing list
[hidden email]
Online Archive : http://ibm-netrexx.215625.n3.nabble.com/