Down the JAVA trail
Locked
1234
1234
 
|
One particular example of confusion about JAVA: a book I'm using shows:
import java.util.* but netrexxj/Edit won't deal with the *. Another question: The guisample.nrx deal; How does it work. Does the jEdit create the Java, then compile that and turn that over to JVM. I don't find many explanations and descriptions of the processes going on. BobH _______________________________________________ Ibm-netrexx mailing list [hidden email] |
|
|
Bob ;
Do you have Mike's book "The NetRexx Language" or the newer PDF version from here?: http://speleotrove.com/misc/NetRexx2.pdf If you look up the NetRexx syntax for the "import" statement, you will find that it is different from the Java import statement. The asterisk (*) is omitted in NetRexx - the wildcard is implied by the ending period. As for running the guisample.nrx program in jEdit, if you use the compiler plugin (NetRexxDE), you must first compile (upside down purple heart button), then pass the compiled class to the JVM (play button) for execution. If you use the interpreter plugin (NetRexxScript) , you can just select the play button and the program will be passed to Mike's NetRexx translator to produce an in memory interpretable version which is then passed directly to the JVM for execution. The plugins are open source, so you can look at the code for details on how they work. You might want to look at the plugin help files in the jEdit Help option for more details on using them. -- Kermit On 10/7/2010 12:57 PM, Robert Hamilton wrote: One particular example of confusion about JAVA: a book I'm using shows: _______________________________________________ Ibm-netrexx mailing list [hidden email] |
|
In reply to this post by Robert L Hamilton
On 7 October 2010 12:57, Robert Hamilton <[hidden email]> wrote: One particular example of confusion about JAVA: a book I'm using shows: NetRexx uses different syntax to import class libraries; you need to transpose the Java syntax to NetRexx. In this case (for NetRexx) it will be: import java.util.
jEdit is a programmer's editor: it doesn't compile anything. The NetRexxDE and NetRexxScript plug-ins are the magic that invoke the compilers and/or Java and they do it in their own special ways. However you compile the program though, you will be able to run the resulting guisample.class file via java (or javaw) completely outside of jEdit. (Always supposing your Java environment is correctly configured.) I don't find many explanations and descriptions of the processes going on. -- Can't tweet, won't tweet! _______________________________________________ Ibm-netrexx mailing list [hidden email]
Alan
-- Needs more cowbell. |
|
|
In reply to this post by Kermit Kiser
thanks; I have the 22 May 2009 version and one earlier. The reason for my question: I don't know if I'm dealing with JAVA or NetRexx in jEdit. And, If NetRexx can deal with JAVA why can it not deal with the JAVA 'import' statement?
Thanks for you time and enjoy the day. . . Bobh _______________________________________________ Ibm-netrexx mailing list [hidden email] |
|
|
+1 vote for Bob's request to enhance NetRexx to accept Java wildcard
syntax on import statements. I never did understand why I have to
remember a different and much less obvious syntax for that one.
-- Kermit On 10/7/2010 11:10 PM, Robert Hamilton wrote: thanks; I have the 22 May 2009 version and one earlier. The reason for my question: I don't know if I'm dealing with JAVA or NetRexx in jEdit. And, If NetRexx can deal with JAVA why can it not deal with the JAVA 'import' statement? _______________________________________________ Ibm-netrexx mailing list [hidden email] |
|
|
-1 For the following: For one it's not only a syntax difference it's also a semantics one. Java's "import pkgname.*" and NetRexx's "import pkgname." Are not equivalent at all. The former will import all classes in package pkgname while the latter will import these *AND* all classes in any sub-packages under pkgname. Being the case that NetRexx supports an additional import mode over java, there is no direct mapping to use here. I must say I favour the NetRexx way: let the compiler do as much work as possible for you :-) - Saludos / Kind regards, David Requena -----Original Message----- From: Kermit Kiser <[hidden email]> Sender: [hidden email] Date: Fri, 08 Oct 2010 09:44:41 To: IBM Netrexx<[hidden email]> Reply-To: IBM Netrexx <[hidden email]> Subject: Re: [Ibm-netrexx] Down the JAVA trail _______________________________________________ Ibm-netrexx mailing list [hidden email] _______________________________________________ Ibm-netrexx mailing list [hidden email] |
|
|
In reply to this post by Kermit Kiser
_______________________________________________ Ibm-netrexx mailing list [hidden email] |
|
|
In reply to this post by Kermit Kiser
It was more of a comment than a request for a change; I had been lulled into a 'comfort zone' where it didn't seem to matter,,, jEdit/NetRexx would deal with it.
On another deal, I'm running ooRexx (15 Aug 2009) on a server in England and I need to compute and display a date/time 6 hrs before to get on Texas time. I would bet that in jEdit/NetRexx/JAVA there is a built in function but I don't know of any in ooREXX v. 4.o.o of 15 Aug 2009. Thanx for the time and enjoy the Day or evening, whatever . . . BobH _______________________________________________ Ibm-netrexx mailing list [hidden email] |
|
|
In reply to this post by Mike Cowlishaw
I go for leaving NetRexx 2 as it was in 22 May 2009 description -- PERIOD. except any bug fixes, of course.
Bobh On Fri, Oct 8, 2010 at 1:51 PM, Mike Cowlishaw <[hidden email]> wrote:
_______________________________________________ Ibm-netrexx mailing list [hidden email] |
|
|
In reply to this post by Robert L Hamilton
On 7 October 2010 23:10, Robert Hamilton <[hidden email]> wrote: thanks; I have the 22 May 2009 version and one earlier. The reason for my question: I don't know if I'm dealing with JAVA or NetRexx in jEdit. And, If NetRexx can deal with JAVA why can it not deal with the JAVA 'import' statement? At it's simplest, when you're working with jEdit you're dealing with a text file. It only becomes Java or a NetRexx when you try to compile it. If you try to compile a text file with NetRexx statements in it with the JavaSidekick plugin you'll be just as unsuccessful as if you try processing a file full of Java statements via the NetRexxDE or NetRexxScript plug-ins. Alan. -- Can't tweet, won't tweet! _______________________________________________ Ibm-netrexx mailing list [hidden email]
Alan
-- Needs more cowbell. |
|
|
In reply to this post by Robert L Hamilton
On 10/8/2010 2:11 PM, Robert Hamilton wrote:
> I go for leaving NetRexx 2 as it was in 22 May 2009 description -- > PERIOD. except any bug fixes, of course. > Or emulate Donald Knuth and freeze the version number at something like Euler's constant or the Golden Ratio. Tom. _______________________________________________ Ibm-netrexx mailing list [hidden email] |
|
|
I am certainly happy with NetRexx the way it is -- I've used it for ten years -- but I worry about progressively losing Java interoperability at the class level, which I think is an important element in attracting others to NetRexx. David Requena says this has already happened.
On Fri, Oct 8, 2010 at 3:47 PM, Tom Maynard <[hidden email]> wrote:
_______________________________________________ Ibm-netrexx mailing list [hidden email] |
|
|
In reply to this post by Robert L Hamilton
say .DateTime~new~toTimeZone(-6)
will convert local time to the central timezone in ooRexx. Rick On Fri, Oct 8, 2010 at 3:06 PM, Robert Hamilton <[hidden email]> wrote: > It was more of a comment than a request for a change; I had been lulled into > a 'comfort zone' where it didn't seem to matter,,, jEdit/NetRexx would deal > with it. > > On another deal, I'm running ooRexx (15 Aug 2009) on a server in England and > I need to compute and display a date/time 6 hrs before to get on Texas > time. I would bet that in jEdit/NetRexx/JAVA there is a built in function > but I don't know of any in ooREXX v. 4.o.o of 15 Aug 2009. > > Thanx for the time and enjoy the Day or evening, whatever . . . > > BobH > > > > _______________________________________________ > Ibm-netrexx mailing list > [hidden email] > > > _______________________________________________ Ibm-netrexx mailing list [hidden email] |
|
|
In reply to this post by George Hovey-2
My mother was not by nature a talented cook. She depended on recipes
from friends, relatives, books, magazines, and the occasional leak from a prestigious restaurant. Unfortunately, she invariably modified the recipe somewhat from the original, regardless of its provenance. And the result was usually dismal. I can remember as a 12-year-old child, pleading with her to simply follow the recipe the first time. Then, if that was successful, she could get creative and try to improve on the original. My mother came to mind as I see a theme developing on this forum: 1. Someone proposes an "improvement" to NetRexx. 2. Much healthy discussion ensues. 3. Mike finally weighs in with the 12 perfectly good reasons why he did it the way that he did. 4. Most folks agree that they had not considered all of the factors, that Mike was right all along, and we should leave well enough alone. We do not yet have access to the NetRexx source code, and when we do get it we will need to do a fair amount of organizational work to make it "open" and available to everyone. There will be change management processes to establish, teams to populate, build machines to create, documentation to update, copyright notices to change, etc., etc., etc. Given that all this work will be done by volunteers (with real jobs, usually) I don't expect (or want) to see any significant change to NetRexx in its first year. As for bug fixes, I've been a member of this list since its inception; we went a very long time before the first legitimate bug was identified; I don't recall any after that. (I suspect that you could count them on one hand and still have enough fingers left over with which to properly salute Java-the-language in whatever cultural gesture you prefer. ;-) My primary point is (as it was to my mother) *don't change anything until you truly understand how it was intended to work*. Then you will have a base from which to make knowledgeable updates. Then, with full involvement of the "Open NetRexx Community", decisions can be made regarding where to direct efforts to enhance NetRexx. I agree that Java class interoperability is a far more significant concern than whether or not the 'import' or assignment statements should match Java's. But the decision should be the stakeholders' and at the moment we are still building the "Open NetRexx Community". Please do not take this to mean that I am trying to suppress a full and rousing discussion of NetRexx and ways in which it could be improved. All such ideas will are being archived and will be available when the time comes to consider enhancements. Bring 'em on. But right now, what we really need are bullet- and idiot-proof installation routines for NetRexx and its various editor/IDE/plugins on a slew of different platforms. We need a true "NetRexx Users Guide" as a companion volume to Mike's TNL reference. It will answer all the "why" and "when" questions that every beginner has. We need MANY more examples. In twenty years of teaching programming languages, I've found that you hook students with examples of useful code, which you then use to explicate the underlying syntax, grammar, and features. And yes, my first NetRexx course used a side-by-side comparison of a (pre-SWING) Java GUI app and its NetRexx equivalent. I had to back up quite a ways to cover everything that was involved to make that code work, but by then they were motivated to learn how. Most folks feel that if they had a couple of clear examples of code doing something akin to what they need, they could figure out how to make it do what they want. It's not the most efficient way of learning (many of my students are self-taught, then come to me to learn what they missed) but few are truly capable of learning it from a book. So who wants to start collecting code for the "NetRexx By Example" repository? -Chip- On 10/8/10 20:02 George Hovey said: > I am certainly happy with NetRexx the way it is -- I've used it for ten > years -- but I worry about progressively losing Java interoperability at > the class level, which I think is an important element in attracting > others to NetRexx. David Requena says this has already happened. > > On Fri, Oct 8, 2010 at 3:47 PM, Tom Maynard <[hidden email] > <mailto:[hidden email]>> wrote: > > On 10/8/2010 2:11 PM, Robert Hamilton wrote: > > I go for leaving NetRexx 2 as it was in 22 May 2009 description > -- PERIOD. except any bug fixes, of course. > > Or emulate Donald Knuth and freeze the version number at something > like Euler's constant or the Golden Ratio. Ibm-netrexx mailing list [hidden email] |
|
|
In reply to this post by Mike Cowlishaw
Your explanation, like David's, is very technical and no doubt
technically correct. But to me those explanations violate some basic
principles behind Rexx:
1) The computer language should be designed for the human user, not the computer. 2) The principle of "least astonishment". Both Bob and I expected "import" to allow the asterisk wildcard syntax which makes sense to a non-expert user. Nor does it hurt anything to allow it since NetRexx already provides the same use case anyway. -- Kermit On 10/8/2010 11:51 AM, Mike Cowlishaw wrote:
_______________________________________________ Ibm-netrexx mailing list [hidden email] |
|
|
In reply to this post by Aviatrexx
That sounds right to me. How about some kind of rough organization of the effort, eg a separate thread to discuss just the installation issue, that doesn't meander off in other directions? It might have a title that clearly implies this, like "Please, NetRexx Installation Issues ONLY". Didn't David Requena's group experiment produce a result of significance to this issue? He might start by discussing the implications of his result.
On Fri, Oct 8, 2010 at 6:49 PM, Chip Davis <[hidden email]> wrote: My mother was not by nature a talented cook. She depended on recipes from friends, relatives, books, magazines, and the occasional leak from a prestigious restaurant. Unfortunately, she invariably modified the recipe somewhat from the original, regardless of its provenance. _______________________________________________ Ibm-netrexx mailing list [hidden email] |
|
|
I'm thinking it is time for a Wiki, rather than a news-list-group
organization, for these topics. Unless there is some all-thoughtful person with lots of time who wishes to be "moderator." On 10/8/2010 8:46 PM, George Hovey wrote: > That sounds right to me. How about some kind of rough organization of > the effort, eg a separate thread to discuss just the installation > issue, that doesn't meander off in other directions? It might have a > title that clearly implies this, like "Please, NetRexx Installation > Issues ONLY". _______________________________________________ Ibm-netrexx mailing list [hidden email] |
|
|
How does it work?
On Fri, Oct 8, 2010 at 9:03 PM, Jeff Hennick <[hidden email]> wrote: I'm thinking it is time for a Wiki, rather than a news-list-group organization, for these topics. Unless there is some all-thoughtful person with lots of time who wishes to be "moderator." _______________________________________________ Ibm-netrexx mailing list [hidden email] |
|
|
In reply to this post by Kermit Kiser
Kermit,
'*' in Rexx and Netrexx means multiply, not
wild-card. Nowhere does it mean 'wildcard'. If one were
to add that meaning, why just in import? Any why just '*', why not '?' and
other wildcards, too? And perhaps more of reguler expression sytax, so one
could import all classes with numbers in their names, etc. (which could be
handy sometimes).
In short, by adding the special case you'd be adding an
anomaly -- and one that just adds a new meaning for an
operator character and hence verbosity and
complexity.
One _can_ put together a programming language by adding
snippets from other languages on a whim or any suggestion, but you'll end up
with Perl. That's OK if you like that kind of thing, but it is
incomprehensible to most people who are not dedicated
programmers.
Mike
_______________________________________________ Ibm-netrexx mailing list [hidden email] |
|
|
In reply to this post by Jeff Hennick
Wiki sounds good. More than one 'document' sounds better. Ie the documentation that we'll need, the learning pathway, could be a wiki document. This could either include installation issues, or an installation guide could be a separate wiki (to become a chapter in the eventual comprehensive guide).
We could have a wiki for NetRexx 3 ideas too, partly to keep them separate, and partly because when people read the discussion associated with each part, they may decide that their idea was not so hot after all, and reduce the 'noise'. It would also stop us from reinventing wheels. Connor. On 9 Oct 2010, at 02:03, Jeff Hennick wrote:
_______________________________________________ Ibm-netrexx mailing list [hidden email] |
«
Return to ibm-netrexx
|
1 view|%1 views
| Free forum by Nabble | Edit this page |