There is now a patch on the after3.01 test branch of the NetRexx
repository which modifies the default translator options slightly: If -exec or -arg (interpret) processing is requested, the following defaults are assumed: -nojava -nocompile -nocrossref This prevents unneeded files from being created during an interpretation run and also allows NetRexx programs to be executed without needing a Java compiler. As usual, if you don't want to checkout and build the compiler and runtime modules from the repository, you can download the latest test versions from the download area of the NetRexx Plus project on Kenai: http://kenai.com/projects/netrexx-plus/downloads Now here is a question for you: Are there others besides me who never use the crossref files created by the translator? Would it be a problem for anyone if the default translator option changed to "nocrossref"? -- Kermit _______________________________________________ Ibm-netrexx mailing list [hidden email] Online Archive : http://ibm-netrexx.215625.n3.nabble.com/ |
Kermit,
thanks for this, and I am sure that Rony will be happy. And to answer your question, I never use the crossrefs. best regards, René. On Mon, 03 Oct 2011 01:59:10 -0700, Kermit Kiser wrote: > There is now a patch on the after3.01 test branch of the NetRexx > repository which modifies the default translator options slightly: > > If -exec or -arg (interpret) processing is requested, the following > defaults are assumed: > > -nojava -nocompile -nocrossref > > This prevents unneeded files from being created during an > interpretation run and also allows NetRexx programs to be executed > without needing a Java compiler. > > As usual, if you don't want to checkout and build the compiler and > runtime modules from the repository, you can download the latest test > versions from the download area of the NetRexx Plus project on Kenai: > > http://kenai.com/projects/netrexx-plus/downloads > > Now here is a question for you: > > Are there others besides me who never use the crossref files created > by the translator? Would it be a problem for anyone if the default > translator option changed to "nocrossref"? > > -- Kermit > > _______________________________________________ > 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/ |
On 03.10.2011 11:21, rvjansen wrote: > Kermit, Thank you very much for your efforts! > thanks for this, and I am sure that Rony will be happy. +1 8-) Seriously, this will enhance the usability immediately (and remarkably) as newcomers/interested people who wish to experiment with NetRexx, but have no JDK installed are still able to use NetRexx without (misleading) errors. ---rony _______________________________________________ Ibm-netrexx mailing list [hidden email] Online Archive : http://ibm-netrexx.215625.n3.nabble.com/ |
In reply to this post by Kermit Kiser
Hello Kermit, and all,
I am NOT at all using -crossref (as it does NOT cross-reference used properties of other classes, only local var's). I would support the idea of having -nocrossref as the default. Needless to mention that I've implemented a GLOBAL cross-Reference (also Cross-Referencing all properties). I will release this shortly (this month) End of advertisement. Thomas. ======================================================== Am 03.10.2011 10:59, schrieb Kermit Kiser: > There is now a patch on the after3.01 test branch of the NetRexx > repository which modifies the default translator options slightly: > > If -exec or -arg (interpret) processing is requested, the following > defaults are assumed: > > -nojava -nocompile -nocrossref > > This prevents unneeded files from being created during an > interpretation run and also allows NetRexx programs to be executed > without needing a Java compiler. > > As usual, if you don't want to checkout and build the compiler and > runtime modules from the repository, you can download the latest test > versions from the download area of the NetRexx Plus project on Kenai: > > http://kenai.com/projects/netrexx-plus/downloads > > Now here is a question for you: > > Are there others besides me who never use the crossref files created > by the translator? Would it be a problem for anyone if the default > translator option changed to "nocrossref"? > > -- Kermit > > _______________________________________________ > Ibm-netrexx mailing list > [hidden email] > Online Archive : http://ibm-netrexx.215625.n3.nabble.com/ > > -- Thomas Schneider (Founder of www.thsitc.com) Member of the Rexx Languge Asscociation (www.rexxla.org) Member of the NetRexx Developer's Team (www.netrexx.org) _______________________________________________ Ibm-netrexx mailing list [hidden email] Online Archive : http://ibm-netrexx.215625.n3.nabble.com/
Thomas Schneider, Vienna, Austria (Europe) :-)
www.thsitc.com www.db-123.com |
In reply to this post by Kermit Kiser
> Now here is a question for you: > > Are there others besides me who never use the crossref files > created by the translator? Would it be a problem for anyone > if the default translator option changed to "nocrossref"? The crossref file is not needed by experienced NetRexx/Java folk, but I found it was much appreciated by 'new' people who were not always sure of the type of a given variable. So those are perhaps the people you should be asking? Mike _______________________________________________ Ibm-netrexx mailing list [hidden email] Online Archive : http://ibm-netrexx.215625.n3.nabble.com/ |
I'll offer a "friend of the court brief" on behalf of 'new' people -- I vaguely remember spending a lot of time looking at it when I was getting started. It will be just another confusing detail if they have to deliberately turn it on. And I guess I'm not experienced enough to never need it; I just turn it on/off in my batch file.
I hesitate to bring this up for fear of attracting the Wrath of Chip ;-), but I've wished along with Thomas that it gave more comprehensive typing. On Mon, Oct 3, 2011 at 9:28 AM, Mike Cowlishaw <[hidden email]> wrote:
_______________________________________________ Ibm-netrexx mailing list [hidden email] Online Archive : http://ibm-netrexx.215625.n3.nabble.com/ |
In reply to this post by rvjansen
René -
Rony seems to feel this one is important. Although I am currently focused on changes for the next release after 3.01, it would be easy to merge this change into the trunk for 3.01 if you want. We would just have to update the documentation to reflect the change. Let me know if you want me to do it. Do we have a preliminary target date for the next release after 3.01? That could be a factor in deciding whether or not to wait. As for defaulting to nocrossref, the feedback so far is 3-2 in favor, but as George pointed out it can be set as a personal preference by modifying the shell script. (I mostly use Ant for compiles other than testing - it is easy to turn off there.) Perhaps it would be better to make it an option in the "advanced" installer path. -- Kermit On 10/3/2011 2:21 AM, rvjansen wrote: > Kermit, > > thanks for this, and I am sure that Rony will be happy. > And to answer your question, I never use the crossrefs. > > best regards, > > René. > > On Mon, 03 Oct 2011 01:59:10 -0700, Kermit Kiser wrote: >> There is now a patch on the after3.01 test branch of the NetRexx >> repository which modifies the default translator options slightly: >> >> If -exec or -arg (interpret) processing is requested, the following >> defaults are assumed: >> >> -nojava -nocompile -nocrossref >> >> This prevents unneeded files from being created during an >> interpretation run and also allows NetRexx programs to be executed >> without needing a Java compiler. >> >> As usual, if you don't want to checkout and build the compiler and >> runtime modules from the repository, you can download the latest test >> versions from the download area of the NetRexx Plus project on Kenai: >> >> http://kenai.com/projects/netrexx-plus/downloads >> >> Now here is a question for you: >> >> Are there others besides me who never use the crossref files created >> by the translator? Would it be a problem for anyone if the default >> translator option changed to "nocrossref"? >> >> -- Kermit >> >> _______________________________________________ >> 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/ |
In reply to this post by George Hovey-2
I know I'm late responding to this thread, but I think you'll find
that I reserve the "Wrath of Chip" for suggestions that _require_ more typing, not things that _give_ more typing information. As for the 'crossref' option, I'm with you and Mike's, that it's better to give the novice user more information than less. If our newbie finds a file with xref info in it, he might find it useful. If it becomes an annoyance, we can hope he will look up the netrexxc command syntax and see that there is a flag to turn it off. (Personally, I'd like to see all such listings contain a line at the top noting the magic incantation to turn it off, not unlike the line at the bottom of a discussion group posting that reminds you how to unsubcribe.) If our neophyte doesn't even know that such an output listing exists, he may become frustrated/discouraged/disgusted with NetRexx before he ever finds out that such a helpful listing was available. Then there is the fact that it's a lot easier to simply delete an unneeded xref file than it is to re-run a large compile a second time when you forgot to reset the 'nocrossref' option. Just some more thoughts, no WoC intended... ;-) -Chip- On 10/3/11 15:57 George Hovey said: > I'll offer a "friend of the court brief" on behalf of 'new' people -- I > vaguely remember spending a lot of time looking at it when I was getting > started. It will be just another confusing detail if they have to > deliberately turn it on. And I guess I'm not experienced enough to > never need it; I just turn it on/off in my batch file. > > I hesitate to bring this up for fear of attracting the Wrath of Chip > ;-), but I've wished along with Thomas that it gave more comprehensive > typing. > > On Mon, Oct 3, 2011 at 9:28 AM, Mike Cowlishaw <[hidden email] > <mailto:[hidden email]>> wrote: > > > Now here is a question for you: > > > > Are there others besides me who never use the crossref files > > created by the translator? Would it be a problem for anyone > > if the default translator option changed to "nocrossref"? > > The crossref file is not needed by experienced NetRexx/Java folk, > but I found it > was much appreciated by 'new' people who were not always sure of the > type of a > given variable. So those are perhaps the people you should be asking? _______________________________________________ Ibm-netrexx mailing list [hidden email] Online Archive : http://ibm-netrexx.215625.n3.nabble.com/ |
You can call me a "control freak" if you want, but I still think it is
not a good idea to consume resources creating files that a programmer did not specifically request. The .java files which are of much more use to me than crossrefs are "on demand" only which seems to me a puzzling inconsistency. Anyway, you can rest assured that the default setting for crossrefs has not changed in NetRexxC except in the case where the programmer requests an interpretation rather than a compile. Instead, the NetRexx coded Ant task was changed so that crossrefs default to off when using Ant to compile. I would not be surprised to see Bill and David follow that approach in their IDE tools (Eclipse plugin, NetRexxDE) , so that advanced programmers using such tools only get crossrefs when they ask for them. I would still like to see an installer option to turn them off in the command line scripts. And I really hope you all are correct and that we actually have some of those "neophyte" NetRexx programmers so that we are not a dying breed! -- Kermit On 10/11/2011 11:31 AM, Chip Davis wrote: > I know I'm late responding to this thread, but I think you'll find > that I reserve the "Wrath of Chip" for suggestions that _require_ more > typing, not things that _give_ more typing information. > > As for the 'crossref' option, I'm with you and Mike's, that it's > better to give the novice user more information than less. > > If our newbie finds a file with xref info in it, he might find it > useful. If it becomes an annoyance, we can hope he will look up the > netrexxc command syntax and see that there is a flag to turn it off. > > (Personally, I'd like to see all such listings contain a line at the > top noting the magic incantation to turn it off, not unlike the line > at the bottom of a discussion group posting that reminds you how to > unsubcribe.) > > If our neophyte doesn't even know that such an output listing exists, > he may become frustrated/discouraged/disgusted with NetRexx before he > ever finds out that such a helpful listing was available. > > Then there is the fact that it's a lot easier to simply delete an > unneeded xref file than it is to re-run a large compile a second time > when you forgot to reset the 'nocrossref' option. > > Just some more thoughts, no WoC intended... ;-) > > -Chip- > > On 10/3/11 15:57 George Hovey said: >> I'll offer a "friend of the court brief" on behalf of 'new' people -- >> I vaguely remember spending a lot of time looking at it when I was >> getting started. It will be just another confusing detail if they >> have to deliberately turn it on. And I guess I'm not experienced >> enough to never need it; I just turn it on/off in my batch file. >> >> I hesitate to bring this up for fear of attracting the Wrath of Chip >> ;-), but I've wished along with Thomas that it gave more >> comprehensive typing. >> >> On Mon, Oct 3, 2011 at 9:28 AM, Mike Cowlishaw <[hidden email] >> <mailto:[hidden email]>> wrote: >> >> > Now here is a question for you: >> > >> > Are there others besides me who never use the crossref files >> > created by the translator? Would it be a problem for anyone >> > if the default translator option changed to "nocrossref"? >> >> The crossref file is not needed by experienced NetRexx/Java folk, >> but I found it >> was much appreciated by 'new' people who were not always sure of the >> type of a >> given variable. So those are perhaps the people you should be >> asking? > > _______________________________________________ > 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/ |
In reply to this post by Aviatrexx
Hello Chip, Mike,
*as far as xref (NetRexx Option -crossref) is concerned* *** it is ridiculous (in my old brain) *** That I'm getting only an xref of the local variables, BUT NOT of the imported and also used properties of the other (important) classes. **** FULL STOP *** Massa Thomas. ======================================================== Am 11.10.2011 20:31, schrieb Chip Davis: > I know I'm late responding to this thread, but I think you'll find > that I reserve the "Wrath of Chip" for suggestions that _require_ more > typing, not things that _give_ more typing information. > > As for the 'crossref' option, I'm with you and Mike's, that it's > better to give the novice user more information than less. > > If our newbie finds a file with xref info in it, he might find it > useful. If it becomes an annoyance, we can hope he will look up the > netrexxc command syntax and see that there is a flag to turn it off. > > (Personally, I'd like to see all such listings contain a line at the > top noting the magic incantation to turn it off, not unlike the line > at the bottom of a discussion group posting that reminds you how to > unsubcribe.) > > If our neophyte doesn't even know that such an output listing exists, > he may become frustrated/discouraged/disgusted with NetRexx before he > ever finds out that such a helpful listing was available. > > Then there is the fact that it's a lot easier to simply delete an > unneeded xref file than it is to re-run a large compile a second time > when you forgot to reset the 'nocrossref' option. > > Just some more thoughts, no WoC intended... ;-) > > -Chip- > > On 10/3/11 15:57 George Hovey said: >> I'll offer a "friend of the court brief" on behalf of 'new' people -- >> I vaguely remember spending a lot of time looking at it when I was >> getting started. It will be just another confusing detail if they >> have to deliberately turn it on. And I guess I'm not experienced >> enough to never need it; I just turn it on/off in my batch file. >> >> I hesitate to bring this up for fear of attracting the Wrath of Chip >> ;-), but I've wished along with Thomas that it gave more >> comprehensive typing. >> >> On Mon, Oct 3, 2011 at 9:28 AM, Mike Cowlishaw <[hidden email] >> <mailto:[hidden email]>> wrote: >> >> > Now here is a question for you: >> > >> > Are there others besides me who never use the crossref files >> > created by the translator? Would it be a problem for anyone >> > if the default translator option changed to "nocrossref"? >> >> The crossref file is not needed by experienced NetRexx/Java folk, >> but I found it >> was much appreciated by 'new' people who were not always sure of the >> type of a >> given variable. So those are perhaps the people you should be >> asking? > > _______________________________________________ > Ibm-netrexx mailing list > [hidden email] > Online Archive : http://ibm-netrexx.215625.n3.nabble.com/ > > -- Thomas Schneider (Founder of www.thsitc.com) Member of the Rexx Languge Asscociation (www.rexxla.org) Member of the NetRexx Developer's Team (www.netrexx.org) _______________________________________________ Ibm-netrexx mailing list [hidden email] Online Archive : http://ibm-netrexx.215625.n3.nabble.com/
Thomas Schneider, Vienna, Austria (Europe) :-)
www.thsitc.com www.db-123.com |
In reply to this post by Kermit Kiser
One might suspect that the reason a neophyte is learning NetRexx is
that he wants to avail himself of the Java platform, without incurring the trauma of learning the Java language. If so, he would find little in the .java file of value. A variable cross-reference listing however, is useful when compiling, but not so much when interpreting NetRexx. I have no problem with the option having a different default value depending on how a NetRexx program is executed. Resources consumed in the program development stage are not a critical concern. Having "crossref" the default for compilation is consistent with my exhortation to my students that the second line of any new Rexx program be "Trace Results". There's no sense in making your first debug run with it turned off. You want more information then, not less. One would assume that any package that used the NetRexx compiler under the covers would make the appropriate decision for the circumstances. I can't wait for a NetRexx IDE to become available to the neophytes so that people will stop asking for IDE features to be provided by the language itself... -Chip- On 10/11/11 23:26 Kermit Kiser said: > You can call me a "control freak" if you want, but I still think it is > not a good idea to consume resources creating files that a programmer > did not specifically request. The .java files which are of much more use > to me than crossrefs are "on demand" only which seems to me a puzzling > inconsistency. > > Anyway, you can rest assured that the default setting for crossrefs has > not changed in NetRexxC except in the case where the programmer requests > an interpretation rather than a compile. Instead, the NetRexx coded Ant > task was changed so that crossrefs default to off when using Ant to > compile. I would not be surprised to see Bill and David follow that > approach in their IDE tools (Eclipse plugin, NetRexxDE) , so that > advanced programmers using such tools only get crossrefs when they ask > for them. I would still like to see an installer option to turn them off > in the command line scripts. > > And I really hope you all are correct and that we actually have some of > those "neophyte" NetRexx programmers so that we are not a dying breed! > > -- Kermit > > > On 10/11/2011 11:31 AM, Chip Davis wrote: >> I know I'm late responding to this thread, but I think you'll find >> that I reserve the "Wrath of Chip" for suggestions that _require_ more >> typing, not things that _give_ more typing information. >> >> As for the 'crossref' option, I'm with you and Mike's, that it's >> better to give the novice user more information than less. >> >> If our newbie finds a file with xref info in it, he might find it >> useful. If it becomes an annoyance, we can hope he will look up the >> netrexxc command syntax and see that there is a flag to turn it off. >> >> (Personally, I'd like to see all such listings contain a line at the >> top noting the magic incantation to turn it off, not unlike the line >> at the bottom of a discussion group posting that reminds you how to >> unsubcribe.) >> >> If our neophyte doesn't even know that such an output listing exists, >> he may become frustrated/discouraged/disgusted with NetRexx before he >> ever finds out that such a helpful listing was available. >> >> Then there is the fact that it's a lot easier to simply delete an >> unneeded xref file than it is to re-run a large compile a second time >> when you forgot to reset the 'nocrossref' option. >> >> Just some more thoughts, no WoC intended... ;-) >> >> -Chip- >> >> On 10/3/11 15:57 George Hovey said: >>> I'll offer a "friend of the court brief" on behalf of 'new' people -- >>> I vaguely remember spending a lot of time looking at it when I was >>> getting started. It will be just another confusing detail if they >>> have to deliberately turn it on. And I guess I'm not experienced >>> enough to never need it; I just turn it on/off in my batch file. >>> >>> I hesitate to bring this up for fear of attracting the Wrath of Chip >>> ;-), but I've wished along with Thomas that it gave more >>> comprehensive typing. >>> >>> On Mon, Oct 3, 2011 at 9:28 AM, Mike Cowlishaw <[hidden email] >>> <mailto:[hidden email]>> wrote: >>> >>> > Now here is a question for you: >>> > >>> > Are there others besides me who never use the crossref files >>> > created by the translator? Would it be a problem for anyone >>> > if the default translator option changed to "nocrossref"? >>> >>> The crossref file is not needed by experienced NetRexx/Java folk, >>> but I found it >>> was much appreciated by 'new' people who were not always sure of the >>> type of a >>> given variable. So those are perhaps the people you should be >>> asking? >> >> _______________________________________________ >> 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/ |
Could not a NetRexx "neophyte" be a Java programmer looking for a better
tool? The problem with an IDE (this includes our current tools - the Eclipse plugin and the jEdit plugins NetRexxDE / NetRexxScript, and even Ant) is that there is a significant learning curve for the IDE environment itself which could create an insurmountable barrier for a neophyte who has not used that environment before. It may be best to direct them to the command line tools initially and just have a very good basic installer! I am working on a concept for a new type of IDE but it is nowhere near having any usable code to release. And upgrading the present tools probably should come first. -- Kermit On 10/12/2011 11:55 AM, Chip Davis wrote: > One might suspect that the reason a neophyte is learning NetRexx is > that he wants to avail himself of the Java platform, without incurring > the trauma of learning the Java language. If so, he would find little > in the .java file of value. > > A variable cross-reference listing however, is useful when compiling, > but not so much when interpreting NetRexx. I have no problem with the > option having a different default value depending on how a NetRexx > program is executed. Resources consumed in the program development > stage are not a critical concern. > > Having "crossref" the default for compilation is consistent with my > exhortation to my students that the second line of any new Rexx > program be "Trace Results". There's no sense in making your first > debug run with it turned off. You want more information then, not less. > > One would assume that any package that used the NetRexx compiler under > the covers would make the appropriate decision for the circumstances. > I can't wait for a NetRexx IDE to become available to the neophytes > so that people will stop asking for IDE features to be provided by the > language itself... > > -Chip- > > On 10/11/11 23:26 Kermit Kiser said: >> You can call me a "control freak" if you want, but I still think it >> is not a good idea to consume resources creating files that a >> programmer did not specifically request. The .java files which are of >> much more use to me than crossrefs are "on demand" only which seems >> to me a puzzling inconsistency. >> >> Anyway, you can rest assured that the default setting for crossrefs >> has not changed in NetRexxC except in the case where the programmer >> requests an interpretation rather than a compile. Instead, the >> NetRexx coded Ant task was changed so that crossrefs default to off >> when using Ant to compile. I would not be surprised to see Bill and >> David follow that approach in their IDE tools (Eclipse plugin, >> NetRexxDE) , so that advanced programmers using such tools only get >> crossrefs when they ask for them. I would still like to see an >> installer option to turn them off in the command line scripts. >> >> And I really hope you all are correct and that we actually have some >> of those "neophyte" NetRexx programmers so that we are not a dying >> breed! >> >> -- Kermit >> Ibm-netrexx mailing list [hidden email] Online Archive : http://ibm-netrexx.215625.n3.nabble.com/ |
Free forum by Nabble | Edit this page |