is NetRexx still a HUMAN oriented language?
Locked
 
|
I'm sure:
a) I will be either ignored! (ok.) b) *or I will go somebody on your nerves (again, as always) The Rexx Familiy of Languages does claim to be *Human orinented* That's good, ok. But *what* shall a human beeing do have to do with the so frequent: term part must start with a symbol message he/she is getting when compiling this human oriented language? For sure, *no human beeing*, except IT-professionals, do even *know* what a *term* is. *And*, frankly speaking, I did have to re-re-read Mike's deinition of a term in nrl1,2,3 .pdf *a lot of times* until I did get it... Not to annoy you, not to disrupt you: BUT: we will need a much easier explanation in the nrl3.pdf, I'm thinking. Also, the whole discussion about the handling of stubs seems (for me) rather complicated. But: I may be wrong ... I'm just feeling it's much more too technical (not simple human thought!) Massa Thomas, as my Egyptian friends, to whome I do say: Habibi, do say... PS: Just for thinking, NOT for OFFENDING anybody :-) -- 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 |
Re: is NetRexx still a HUMAN oriented language?
|
Is there concise listing of NetRexx error messages, like TRL's Section
17, "Error Numbers and Messages"? That section also has an explanatory paragraph or two for each. I can't find such a list in NetRexx 3 Documentation 22nd May 2009, nor via a Google Search. If there is not, or there is a list of "just" the messages without explanation, this seems like a good candidate for a wiki page. _______________________________________________ Ibm-netrexx mailing list [hidden email] Online Archive : http://ibm-netrexx.215625.n3.nabble.com/ |
|
|
You might be looking for the NetRexxC.properties file. It is contained
with the source for the NetRexx translator. The header and the first few lines are below. #01 --------------------------------------------------------------------- #02 NetRexx language processor -- messages file mfc #03 Copyright (c) IBM Corporation 1996, 2000. All Rights Reserved. #04 --------------------------------------------------------------------- #10 #11 This is the NetRexxC messages master file, which can be used #12 directly as a java properties file or from which some new properties #13 file (or other lookup database) can be generated or translated. #14 #15 Each line is a single message (which allows the file to be sorted); #16 the first word is the "key" (for example, 'cannot.extend.self') and #17 the remainder is the 'human-readable' message, with inserts #18 indicated as %1, %2, etc. (corresponding to the extra arguments #19 supplied to RxError, etc.). The inserts need not be be in ascending #20 numerical order, and may be repeated if required. #21 #22 Empty lines, and lines starting with '#' or blank, are ignored during #23 processing. The Java Properties class can use this file directly, #24 though NetRexxC used its own class (which was a little over ten #25 times faster, as measured 1996.11.02) until Java 1.1, since when it #26 is used as a ResourceBundle. #40 --------------------------------------------------------------------- #41 1996.10.29 Initial version #42 1996.11.02 First pass wording and consolidation completed #43 --------------------------------------------------------------------- abstract.cannot.have.optional Abstract methods cannot have optional arguments abstract.class.cannot.be.constructed An abstract type cannot be constructed abstract.constructor An abstract constructor cannot be invoked ... while.needs.boolean.type Expression type '%1' cannot be converted to '0' or '1' while.or.until.expected WHILE or UNTIL expected wrong.do.syntax.retry Wrong syntax for DO; retrying as a LOOP instruction There are 408 messages. On 11/10/2011 12:21 PM, Jeff Hennick wrote: > Is there concise listing of NetRexx error messages, like TRL's Section > 17, "Error Numbers and Messages"? That section also has an > explanatory paragraph or two for each. > > I can't find such a list in NetRexx 3 Documentation 22nd May 2009, nor > via a Google Search. > > If there is not, or there is a list of "just" the messages without > explanation, this seems like a good candidate for a wiki page. > > _______________________________________________ > Ibm-netrexx mailing list > [hidden email] > Online Archive : http://ibm-netrexx.215625.n3.nabble.com/ > > > > ----- > No virus found in this message. > Checked by AVG - www.avg.com > Version: 2012.0.1869 / Virus Database: 2092/4608 - Release Date: 11/10/11 > > _______________________________________________ Ibm-netrexx mailing list [hidden email] Online Archive : http://ibm-netrexx.215625.n3.nabble.com/ |
|
|
In reply to this post by Jeff Hennick
Hello Jeff,
wisely, Mike did code all errors/warnings/messages in an abbreviated form (like: is.no.variable). His abbreviated form, and the full ENGLISH messages, are there in the KENAI NetRexx Repository. When you cannot find them quickly *there* , give me a mail, and I will look up my (nearly endless) A4-BENE Folders there on my desk(s). As Alzheimer is obviously growing, I cannot recall the name of this file, for the minute :-( Thomas. ================================================================================== Am 10.11.2011 18:21, schrieb Jeff Hennick: > Is there concise listing of NetRexx error messages, like TRL's Section > 17, "Error Numbers and Messages"? That section also has an > explanatory paragraph or two for each. > > I can't find such a list in NetRexx 3 Documentation 22nd May 2009, nor > via a Google Search. > > If there is not, or there is a list of "just" the messages without > explanation, this seems like a good candidate for a wiki page. > > _______________________________________________ > 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 ThSITC
Thomas,
I sympathize with this but we need to design the solution. I would start with the following suggestions: TSO (the tmp processor) has an option in which you can enter a ? to get a more verbose explanation MQ Series has a separate commmand in which you can enter reason codes and get an explanation from the documentation. If we set up a database, for example, the wiki on kenai, with common errors which need more explanation, we could eventually turn this into more informative error messages. The hard part is to imagine the mental state of the novice programmer: I am sure most of us on this list have been sufficiently exposed to most of those error messages that they actually mean something to us now. The example you mention has a clear meaning to me, but I can imagine that it has not to someone else. I remember having struggled with the 'static method cannot' series of messages. This goes away with a clear mental model of oo implementation categories. Try to have a constructive idea coming out of these sentiments. best regards, René. On Thu, 10 Nov 2011 17:19:49 +0100, Thomas Schneider wrote: > I'm sure: > > a) I will be either ignored! (ok.) > b) *or I will go somebody on your nerves (again, as always) > > The Rexx Familiy of Languages does claim to be *Human orinented* > That's good, ok. > > But *what* shall a human beeing do have to do with the so frequent: > > term part must start with a symbol > > message he/she is getting when compiling this human oriented > language? > > For sure, *no human beeing*, except IT-professionals, do even *know* > what a *term* is. _______________________________________________ Ibm-netrexx mailing list [hidden email] Online Archive : http://ibm-netrexx.215625.n3.nabble.com/ |
Re: is NetRexx still a HUMAN oriented language?
|
|
I would most assuredly be qualified to generate the errors.
Bob Hamilton On Fri, Nov 11, 2011 at 9:11 AM, rvjansen <[hidden email]> wrote: Thomas, _______________________________________________ Ibm-netrexx mailing list [hidden email] Online Archive : http://ibm-netrexx.215625.n3.nabble.com/ |
|
|
In reply to this post by rvjansen
My experience on the ANSI committee indicates that any such effort to
clarify the error messages is A GOOD THING. That's how we got the much more specific and informative Rexx error message subcodes. While a wiki might be an elegant and extensible approach, it seems to me that an HTML file might be just as useful and a lot less weight. For example, the (necessarily) terse error message text would have a link that displays a more verbose exposition of the error, the common causes, suggested remedies, etc. In Thomas' case, it seems to be a simple matter of unfamiliarity with the terms (no pun intended) used in the message. Specifically, the words 'term' and 'symbol'. The error message text could include links to a glossary containing the definition of those terms (which could themselves have glossary links, recursively). I'm not suggesting we undertake a major project. We could start with a simple HTML file where each error message links to substantially the same text. The glossary could be built by taking the prose from NRL3. If folks would volunteer to take a message or glossary term and provide a more complete entry, it would not be a big job for any one person. Sort of the wiki collaboration model, but without the wiki overhead. As the Editor hasn't had much to contribute so far, I will volunteer to consolidate the contributions, as we still have the SVN tracking issue. Thoughts? -Chip- On 11/11/11 15:11 rvjansen said: > Thomas, > > I sympathize with this but we need to design the solution. I would start > with the following suggestions: > > TSO (the tmp processor) has an option in which you can enter a ? to get > a more verbose explanation > MQ Series has a separate commmand in which you can enter reason codes > and get an explanation from the documentation. > > If we set up a database, for example, the wiki on kenai, with common > errors which need more explanation, we could eventually turn this into > more informative error messages. > > The hard part is to imagine the mental state of the novice programmer: I > am sure most of us on this list have been sufficiently exposed to most > of those error messages that they actually mean something to us now. The > example you mention has a clear meaning to me, but I can imagine that it > has not to someone else. I remember having struggled with the 'static > method cannot' series of messages. This goes away with a clear mental > model of oo implementation categories. > > Try to have a constructive idea coming out of these sentiments. > > best regards, > > René. > > On Thu, 10 Nov 2011 17:19:49 +0100, Thomas Schneider wrote: >> I'm sure: >> >> a) I will be either ignored! (ok.) >> b) *or I will go somebody on your nerves (again, as always) >> >> The Rexx Familiy of Languages does claim to be *Human orinented* >> That's good, ok. >> >> But *what* shall a human beeing do have to do with the so frequent: >> >> term part must start with a symbol >> >> message he/she is getting when compiling this human oriented language? >> >> For sure, *no human beeing*, except IT-professionals, do even *know* >> what a *term* is. > > _______________________________________________ > 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/ |
Re: is NetRexx still a HUMAN oriented language?
|
|
Could I suggest the inclusion of 'Sample Code' ?
Bob Hamilton Richardson Texas USA On Fri, Nov 11, 2011 at 10:15 AM, Chip Davis <[hidden email]> wrote: My experience on the ANSI committee indicates that any such effort to clarify the error messages is A GOOD THING. That's how we got the much more specific and informative Rexx error message subcodes. _______________________________________________ Ibm-netrexx mailing list [hidden email] Online Archive : http://ibm-netrexx.215625.n3.nabble.com/ |
|
|
In reply to this post by Aviatrexx
I'm interested in this topic because one of the "to-do" items for the
Eclipse NetRexx plugin is the implementation of a NetRexx help facility integrated with Eclipse. The Eclipse IDE supports supplying error feedback and help information at several levels. The Java Development Environment is quite comprehensive, and my objective is to provide similar support for NetRexx. It includes the normal main menu "Help" item for documentation, dynamic (topic related) help triggered by F1, context (text hover) based popup windows containing descriptive information, and immediate syntax and semantic error indications via error messages and syntax coloring. Only the syntax error messages and coloring have been implemented so far. I don't know exactly what else will be generated for text or how it will be organized, but certainly everything I do will be freely available. Possibly it will be adaptable to a non-Eclipse environment. Bill On 11/11/2011 11:15 AM, Chip Davis wrote: > My experience on the ANSI committee indicates that any such effort to > clarify the error messages is A GOOD THING. That's how we got the > much more specific and informative Rexx error message subcodes. > > While a wiki might be an elegant and extensible approach, it seems to > me that an HTML file might be just as useful and a lot less weight. > > For example, the (necessarily) terse error message text would have a > link that displays a more verbose exposition of the error, the common > causes, suggested remedies, etc. > > In Thomas' case, it seems to be a simple matter of unfamiliarity with > the terms (no pun intended) used in the message. Specifically, the > words 'term' and 'symbol'. The error message text could include links > to a glossary containing the definition of those terms (which could > themselves have glossary links, recursively). > > I'm not suggesting we undertake a major project. We could start with > a simple HTML file where each error message links to substantially the > same text. The glossary could be built by taking the prose from NRL3. > > If folks would volunteer to take a message or glossary term and > provide a more complete entry, it would not be a big job for any one > person. Sort of the wiki collaboration model, but without the wiki > overhead. > > As the Editor hasn't had much to contribute so far, I will volunteer > to consolidate the contributions, as we still have the SVN tracking > issue. > > Thoughts? > > -Chip- > > On 11/11/11 15:11 rvjansen said: >> Thomas, >> >> I sympathize with this but we need to design the solution. I would >> start with the following suggestions: >> >> TSO (the tmp processor) has an option in which you can enter a ? to >> get a more verbose explanation >> MQ Series has a separate commmand in which you can enter reason codes >> and get an explanation from the documentation. >> >> If we set up a database, for example, the wiki on kenai, with common >> errors which need more explanation, we could eventually turn this >> into more informative error messages. >> >> The hard part is to imagine the mental state of the novice >> programmer: I am sure most of us on this list have been sufficiently >> exposed to most of those error messages that they actually mean >> something to us now. The example you mention has a clear meaning to >> me, but I can imagine that it has not to someone else. I remember >> having struggled with the 'static method cannot' series of messages. >> This goes away with a clear mental model of oo implementation >> categories. >> >> Try to have a constructive idea coming out of these sentiments. >> >> best regards, >> >> René. >> >> On Thu, 10 Nov 2011 17:19:49 +0100, Thomas Schneider wrote: >>> I'm sure: >>> >>> a) I will be either ignored! (ok.) >>> b) *or I will go somebody on your nerves (again, as always) >>> >>> The Rexx Familiy of Languages does claim to be *Human orinented* >>> That's good, ok. >>> >>> But *what* shall a human beeing do have to do with the so frequent: >>> >>> term part must start with a symbol >>> >>> message he/she is getting when compiling this human oriented language? >>> >>> For sure, *no human beeing*, except IT-professionals, do even *know* >>> what a *term* is. >> >> _______________________________________________ >> 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/ > > > > ----- > No virus found in this message. > Checked by AVG - www.avg.com > Version: 2012.0.1869 / Virus Database: 2092/4610 - Release Date: 11/11/11 > > _______________________________________________ Ibm-netrexx mailing list [hidden email] Online Archive : http://ibm-netrexx.215625.n3.nabble.com/ |
«
Return to ibm-netrexx
|
1 view|%1 views
| Free forum by Nabble | Edit this page |