NetRexx system environment variables
Locked
12345
12345
Re: A useful use of CURLY braces
 
|
Hi Fernando, the question (as I see it) is what parenthesis, brackets, and curly braces are for: Let's assume, that in his wisdom, Mr Mike .F. Cowlishaw would have defined: "{" means start, begin, do, etc "}" means end of the above (matching Curly Brace) Then, we might define: a = {1, 2, 3:5, 7, 12} prime_numbers={1,2,3,5,7,11,13, ...) *and*, by using the equivalence definition that ":" is a SYNONYM to ".." (in this context: b= {yes, no, maybe} Then, we could define some equivalences: TRUE == YES FALSE == NO (note that I'm using the '==' as the EQUIVALENCE operator!) What do you think ? ============================================================== Am 12.11.2010 09:53, schrieb Fernando Cassia:
--
Thomas Schneider Projects ReyC & LOGOS on www.KENAI.com _______________________________________________ Ibm-netrexx mailing list [hidden email]
Tom. (ths@db-123.com)
|
Re: NetRexx system environment variables
|
|
In reply to this post by George Hovey-2
Hello Geroge Hovey (*and* all).
Whe you did install NetRexx Syntax higlighting, provided by David Requena's NetRexxDE (successor of SatGuru Srivastava's NetRexxJe), the *DO* and *END* will become JEdit *keywords, and *are* *Syntax highlighted*. Be careful, however: do=13 if a=abc then do x=14; end might yield totally unexpected results ... Thomas Schneider ============================================ Am 12.11.2010 03:15, schrieb George Hovey: Kermit, --
Thomas Schneider Projects ReyC & LOGOS on www.KENAI.com _______________________________________________ Ibm-netrexx mailing list [hidden email]
Tom. (ths@db-123.com)
|
Re: NetRexx system environment variables
|
|
In reply to this post by Jeff Hennick
Hi there,
1.) I did use XEDIT on the CMS system for years.... 2.) then KEDIT, Then JEdit, then UltraEdit, (KEDIT: where I first detected PC Rexx, and thus Rexx) 3.) then, then then... My *advise* is to switch to *JEDIT*. Thomas Schneider. ========================================================= Am 12.11.2010 06:57, schrieb Jeff Hennick: > I have to put in a plug for KEDIT for Windows here. > > 1) It uses "Kexx" (essentially full Rexx) as its scripting language, and > 2) It color codes all of the matching levels of parens, brackets and > braces AND Do...End, or any other pairs, definable by file type or on > the fly. (Other syntax coloring as well.) > > Jeff > (Old XEDIT guy) > > On 11/11/2010 9:15 PM, George Hovey wrote: >> Kermit, >> The one (small) fly in the ointment is that jEdit has a superb >> facility for graphically indicating matching of parens, brackets and >> braces which, unfortunately, doesn't work with Do...End. In a >> previous incarnation as a C programmer I found this quite helpful. > _______________________________________________ > Ibm-netrexx mailing list > [hidden email] > > -- Thomas Schneider Projects ReyC & LOGOS on www.KENAI.com _______________________________________________ Ibm-netrexx mailing list [hidden email]
Tom. (ths@db-123.com)
|
|
|
In reply to this post by Thomas.Schneider.Wien
The Eclipse NetRexx editor colors the first "do" as a variable and
the second as a keyword.
If the statement is in error, such as "do = ;" the do is colored as an error (while you type). On 11/12/2010 9:53 AM, Thomas Schneider wrote: Hello Geroge Hovey (*and* all). _______________________________________________ Ibm-netrexx mailing list [hidden email] |
Re: NetRexx system environment variables
|
|
In reply to this post by Fernando Cassia-2
Hello Fernando: Curley braces are use since I've studied mathematics to define 'collections' (Mike, Rene, correct me, if that is *not* the correct term in english) ;-) Thus, what I am proposing: prime_numbers={1,2,3,5,7,11,13, ...} should define the prime numbers (did you yet find the algorithm for the ellipsis, by the way ??). You will be a multi-millionaire! The ellipsis (...) should be once forever defined aas 'etc.' (in english) ('usw.' in german) The '..' (synonym to ':' in value lists, or the hyphen in many notations, like regular expressions, should be defined as the range operator (and be equivalent to the hyphen, when enclosed in braces, and thus may not mean *minus*) Then, we could say: a = int {15,17,19} roman_letter={a-z} When the hyphen is part of the members of the collection, then a) either quotes are needed b) or the '..' must be used, thus giving roman_letter={a..z} *is equivalent to* {a-z} (in this simple case...) All of those rules have been quite successfully used in Mathematics in the past ! :-) So why not use them in Rey ?? Thomas. ================================================================ Am 11.11.2010 21:51, schrieb Fernando Cassia: > On Thu, Nov 11, 2010 at 4:45 PM, Tom Maynard<[hidden email]> wrote: >> Process p = "cmd /c path".execute() >> p.text.tokenize(";").each { entry -> println entry } >> >> But without the clarity, IMO. I admit, I may be expressing my own bias. >> [Note the curly brackets.] >> Tom. > What is with new languages and curly brackets? I hate them. I see > those in JavaFX and my head begins to spin in a 360-degree full > circle... not to mention having to type them myself... > > FC > > _______________________________________________ > Ibm-netrexx mailing list > [hidden email] > > -- Thomas Schneider Projects ReyC & LOGOS on www.KENAI.com _______________________________________________ Ibm-netrexx mailing list [hidden email]
Tom. (ths@db-123.com)
|
Re: NetRexx system environment variables
|
|
In reply to this post by Thomas.Schneider.Wien
Thomas, I have my own jEdit mode file for NetRexx which colors Do and End (and much else). The problem is to be sure they are matched correctly (am I the only one who gets this wrong?). Remember, Do, Loop and Select all use the same undifferentiated End. Again, a lot of problems go away if one takes advantage of the Label option. I don't know to expect from the example - I never use a keyword as a variable (although MFC has pointed out that I might be using a future keyword), or semicolons outside of strings. One of the joys of NetRexx is saying goodbye to semicolons. On Fri, Nov 12, 2010 at 9:53 AM, Thomas Schneider <[hidden email]> wrote:
_______________________________________________ Ibm-netrexx mailing list [hidden email] |
Re: NetRexx system environment variables
|
|
In reply to this post by Thomas.Schneider.Wien
On Fri, Nov 12, 2010 at 12:17 PM, Thomas Schneider <[hidden email]> wrote:
> > Hello Fernando: > > Curley braces are use since I've studied mathematics to define 'collections' > (Mike, Rene, correct me, if that is *not* the correct term in english) ;-) YES, but, one thing is USING them for specific instances, another is using the braces as an integral part of the language, used everywhere. (See the JavaFX examples I posted before). The difficulty is, at a quick glimpse of a code snipped, seeing what brace closes what, specially when there is a ton of those spread vertically... Curly brackets aka braces are a visual annoyance. FC _______________________________________________ Ibm-netrexx mailing list [hidden email] |
Re: NetRexx system environment variables
|
|
In reply to this post by Robert L Hamilton
Hello Robert again,
... you might want to look at my implementation of the *classic Rexx* *symbol* and *value* functions (at www.Rexx2Nrx.com) which, as far as I can see, do what you want to do.... Simply take the time an scan thru the www.Rexx2Nrx/RunTime section :-) Thomas Schneider. ============================================================ Am 11.11.2010 16:49, schrieb Robert Hamilton: This on-going exercise nudged me back to MFC's introduction to NetRexx 2 of 22 May 2009 ( page 13 in my offprint). In a section "Dealing with reality" is the definition of Astonishment Factor: "If a feature, accidentally misused, gives apparently unpredictable results, then it has a high Astonishment Factor and is therefore undesirable." --
Thomas Schneider Projects ReyC & LOGOS on www.KENAI.com _______________________________________________ Ibm-netrexx mailing list [hidden email]
Tom. (ths@db-123.com)
|
RE: NetRexx system environment variables
|
|
In reply to this post by Fernando Cassia-2
> Curly brackets aka braces are a visual annoyance. That sums it up rather well. And on today's relatively low resolution screens (compared to printed paper) they are too easily confused with parentheses -- which is one of the reasons style guidelines for C and similar languages often require or suggest that they are placed on a separate line, or they are too easily overlooked. Interestingly (perhaps), back in 1996, James Gosling and I were on a panel at the WWW5 conference in Paris; he was asked "what was the biggest mistake you made in Java?". His answer: "using curly braces" ... Mike _______________________________________________ Ibm-netrexx mailing list [hidden email] |
|
In reply to this post by Fernando Cassia-2
On 12 November 2010 07:37, Fernando Cassia <[hidden email]> wrote:
Peeps! Tom thoughtfully started another thread for the "curly brackets" conversation. Please keep this one on topic; it's supposed to be about environment variables: migrate the curly stuff to that thread so that in years to come we can do searches through the mailing list archive to see how these hot-button issues drove us to distraction. Personally though; I liked Curly best of all the Three Stooges. Way better than Shemp and Curly Joe: "nyuk-nyuk-nyuk!","woo-woo-woo!"... :-D A. -- Can't tweet, won't tweet! _______________________________________________ Ibm-netrexx mailing list [hidden email]
Alan
-- Needs more cowbell. |
Re: NetRexx system environment variables
|
|
In reply to this post by Mike Cowlishaw
Hello Mike, *and all*:
The well thing is, nowadays, in the year 2010, that: 1.) The meanning of CURLEY BRACES is well defined in mathematican and physician *notation*. 2.) Curley Braces do surround a *set* and/or *collection* of *members* Full stop. Thomas Schneider. ============================================================= Am 12.11.2010 17:39, schrieb Mike Cowlishaw: > >> Curly brackets aka braces are a visual annoyance. > That sums it up rather well. And on today's relatively low resolution screens > (compared to printed paper) they are too easily confused with parentheses -- > which is one of the reasons style guidelines for C and similar languages often > require or suggest that they are placed on a separate line, or they are too > easily overlooked. > > Interestingly (perhaps), back in 1996, James Gosling and I were on a panel at > the WWW5 conference in Paris; he was asked "what was the biggest mistake you > made in Java?". His answer: "using curly braces" ... > > Mike > > _______________________________________________ > Ibm-netrexx mailing list > [hidden email] > > -- Thomas Schneider Projects ReyC & LOGOS on www.KENAI.com _______________________________________________ Ibm-netrexx mailing list [hidden email]
Tom. (ths@db-123.com)
|
Re: NetRexx system environment variables
|
|
In reply to this post by alansam
On Fri, Nov 12, 2010 at 1:43 PM, Alan Sampson <[hidden email]> wrote:
> Personally though; I liked Curly best of all the Three Stooges. Way better > than Shemp and Curly Joe: "nyuk-nyuk-nyuk!","woo-woo-woo!"... :-D > > A. You stole my joke!. I was about to make a reference to Curly and curly brackets/braces... ;) Now back to the usual signal to noise ratio. :) FC _______________________________________________ Ibm-netrexx mailing list [hidden email] |
Re: NetRexx system environment variables
|
|
In reply to this post by Thomas.Schneider.Wien
Let me just add a comment.
Syntax highlighting is a core jEdit feature. It'll work with NetRexx even with no plugins installed. - Saludos / Kind regards, David Requena -----Original Message----- From: Thomas Schneider <[hidden email]> Sender: [hidden email] Date: Fri, 12 Nov 2010 15:53:31 To: IBM Netrexx<[hidden email]> Reply-To: [hidden email], IBM Netrexx <[hidden email]> Subject: Re: [Ibm-netrexx] NetRexx system environment variables _______________________________________________ Ibm-netrexx mailing list [hidden email] _______________________________________________ Ibm-netrexx mailing list [hidden email] |
Re: NetRexx system environment variables
|
|
In reply to this post by Thomas.Schneider.Wien
Thomas,
The fact that it's hard to follow your logic here would seem to indicate it's anti-netrexx in nature :-) - Saludos / Kind regards, David Requena -----Original Message----- From: Thomas Schneider <[hidden email]> Sender: [hidden email] Date: Fri, 12 Nov 2010 16:17:33 To: IBM Netrexx<[hidden email]> Reply-To: [hidden email], IBM Netrexx <[hidden email]> Subject: Re: [Ibm-netrexx] NetRexx system environment variables Hello Fernando: Curley braces are use since I've studied mathematics to define 'collections' (Mike, Rene, correct me, if that is *not* the correct term in english) ;-) Thus, what I am proposing: prime_numbers={1,2,3,5,7,11,13, ...} should define the prime numbers (did you yet find the algorithm for the ellipsis, by the way ??). You will be a multi-millionaire! The ellipsis (...) should be once forever defined aas 'etc.' (in english) ('usw.' in german) The '..' (synonym to ':' in value lists, or the hyphen in many notations, like regular expressions, should be defined as the range operator (and be equivalent to the hyphen, when enclosed in braces, and thus may not mean *minus*) Then, we could say: a = int {15,17,19} roman_letter={a-z} When the hyphen is part of the members of the collection, then a) either quotes are needed b) or the '..' must be used, thus giving roman_letter={a..z} *is equivalent to* {a-z} (in this simple case...) All of those rules have been quite successfully used in Mathematics in the past ! :-) So why not use them in Rey ?? Thomas. ================================================================ Am 11.11.2010 21:51, schrieb Fernando Cassia: > On Thu, Nov 11, 2010 at 4:45 PM, Tom Maynard<[hidden email]> wrote: >> Process p = "cmd /c path".execute() >> p.text.tokenize(";").each { entry -> println entry } >> >> But without the clarity, IMO. I admit, I may be expressing my own bias. >> [Note the curly brackets.] >> Tom. > What is with new languages and curly brackets? I hate them. I see > those in JavaFX and my head begins to spin in a 360-degree full > circle... not to mention having to type them myself... > > FC > > _______________________________________________ > Ibm-netrexx mailing list > [hidden email] > > -- Thomas Schneider Projects ReyC & LOGOS on www.KENAI.com _______________________________________________ Ibm-netrexx mailing list [hidden email] _______________________________________________ Ibm-netrexx mailing list [hidden email] |
Re: NetRexx system environment variables
|
|
In reply to this post by Fernando Cassia-2
I'm with George here, it's often difficult to know what end goes with what do, select, etc. Not at a first glance certainly. - Saludos / Kind regards, David Requena -----Original Message----- From: Fernando Cassia <[hidden email]> Sender: [hidden email] Date: Fri, 12 Nov 2010 12:37:50 To: <[hidden email]>; IBM Netrexx<[hidden email]> Reply-To: IBM Netrexx <[hidden email]> Subject: Re: [Ibm-netrexx] NetRexx system environment variables On Fri, Nov 12, 2010 at 12:17 PM, Thomas Schneider <[hidden email]> wrote: > > Hello Fernando: > > Curley braces are use since I've studied mathematics to define 'collections' > (Mike, Rene, correct me, if that is *not* the correct term in english) ;-) YES, but, one thing is USING them for specific instances, another is using the braces as an integral part of the language, used everywhere. (See the JavaFX examples I posted before). The difficulty is, at a quick glimpse of a code snipped, seeing what brace closes what, specially when there is a ton of those spread vertically... Curly brackets aka braces are a visual annoyance. FC _______________________________________________ Ibm-netrexx mailing list [hidden email] _______________________________________________ Ibm-netrexx mailing list [hidden email] |
Re: NetRexx system environment variables
|
|
In reply to this post by David Requena
It's more than worth it for this feature alone.
bobh On Fri, Nov 12, 2010 at 1:24 PM, David Requena <[hidden email]> wrote: Let me just add a comment. _______________________________________________ Ibm-netrexx mailing list [hidden email] |
|
|
In reply to this post by David Requena
On 11/12/2010 2:24 PM, David Requena wrote:
> Let me just add a comment. > > Syntax highlighting is a core jEdit feature. It'll work with NetRexx even with no plugins installed. > > - > Saludos / Kind regards, > David Requena There is a significant difference between basic syntax coloring and full semantic coloring. And there are numerous variations between the two. With basic syntax coloring, the color depends only on the content of the token, so A variable with the same name as a keyword is colored the same. With full semantic coloring, a language specific parse is done which allows the color (and font) to reflect the language specific usage of the token. For example, keywords, local variables, global variables and other types of tokens all with the same name can be colored differently. For NetRexx, SELECT and matching END keywords might be colored red, DO and matching END keywords could be blue, etc. (but that is not implemented quite yet :) For PL/I, "if if = then then then = else; else else = if;" becomes quite readable. _______________________________________________ Ibm-netrexx mailing list [hidden email] |
Re: NetRexx system environment variables
|
|
Bill,
My comment was simply addressed to clarify that what jEdit has in this regard is not related to my work, or Kermit's or Satguru's on our plugins. Thomas seemed to imply otherwise. Of course jEdit's 'modes' are regex based and no attempt is made to parse in any way the source being edited so, yes, it's pretty limited. - Saludos / Kind regards, David Requena -----Original Message----- From: Bill Fenlason <[hidden email]> Sender: [hidden email] Date: Fri, 12 Nov 2010 15:18:37 To: IBM Netrexx<[hidden email]> Reply-To: IBM Netrexx <[hidden email]> Subject: Re: [Ibm-netrexx] NetRexx system environment variables On 11/12/2010 2:24 PM, David Requena wrote: > Let me just add a comment. > > Syntax highlighting is a core jEdit feature. It'll work with NetRexx even with no plugins installed. > > - > Saludos / Kind regards, > David Requena There is a significant difference between basic syntax coloring and full semantic coloring. And there are numerous variations between the two. With basic syntax coloring, the color depends only on the content of the token, so A variable with the same name as a keyword is colored the same. With full semantic coloring, a language specific parse is done which allows the color (and font) to reflect the language specific usage of the token. For example, keywords, local variables, global variables and other types of tokens all with the same name can be colored differently. For NetRexx, SELECT and matching END keywords might be colored red, DO and matching END keywords could be blue, etc. (but that is not implemented quite yet :) For PL/I, "if if = then then then = else; else else = if;" becomes quite readable. _______________________________________________ Ibm-netrexx mailing list [hidden email] _______________________________________________ Ibm-netrexx mailing list [hidden email] |
|
|
David,
I guess I missed your point - sorry - I was a bit distracted by all the ropes and Mayans. I wasn't throwing stones at jEdit or your excellent work, just pointing out the wide range of coloring approaches. I've noticed that not everyone is familiar with the different possibilities. Actually, I think that there is an 80-20 rule at work, and that simply coloring keywords and recognizing comments (and quoted strings, etc.) goes a long way to aid productivity. Some of us like immediate error feedback IDE environments (I know I need all the help I can get :), but some excellent programmers find colors and fonts distracting and prefer to work "color blind", and yet others find error feedback annoying. To each his own. Of course most programmers would prefer to give up their religion (or spouse) rather than give up their favorite editor :). On 11/12/2010 6:42 PM, David Requena wrote: > Bill, > > My comment was simply addressed to clarify that what jEdit has in this regard is not related to my work, or Kermit's or Satguru's on our plugins. Thomas seemed to imply otherwise. > > Of course jEdit's 'modes' are regex based and no attempt is made to parse in any way the source being edited so, yes, it's pretty limited. > > - > Saludos / Kind regards, > David Requena > > -----Original Message----- > From: Bill Fenlason<[hidden email]> > Sender: [hidden email] > Date: Fri, 12 Nov 2010 15:18:37 > To: IBM Netrexx<[hidden email]> > Reply-To: IBM Netrexx<[hidden email]> > Subject: Re: [Ibm-netrexx] NetRexx system environment variables > > On 11/12/2010 2:24 PM, David Requena wrote: >> Let me just add a comment. >> >> Syntax highlighting is a core jEdit feature. It'll work with NetRexx even with no plugins installed. >> >> - >> Saludos / Kind regards, >> David Requena > There is a significant difference between basic syntax coloring and full > semantic coloring. And there are numerous variations between the two. > > With basic syntax coloring, the color depends only on the content of the > token, so A variable with the same name as a keyword is colored the same. > > With full semantic coloring, a language specific parse is done which > allows the color (and font) to reflect the language specific usage of > the token. > > For example, keywords, local variables, global variables and other types > of tokens all with the same name can be colored differently. > > For NetRexx, SELECT and matching END keywords might be colored red, DO > and matching END keywords could be blue, etc. (but that is not > implemented quite yet :) > > For PL/I, "if if = then then then = else; else else = if;" becomes quite > readable. > _______________________________________________ Ibm-netrexx mailing list [hidden email] |
Re: NetRexx system environment variables
|
|
Just a slight quibble about the behavior of the Eclipse NetRexx editor. You mention
The Eclipse NetRexx editor colors the first "do" as a variable and
the second as a keyword. ie, in "do=13".It is true that this is a legal assignment, but jEdit's response, though more primitive, seems to me more useful. As soon as "do" is typed it lights up as a keyword alerting one to the fact that this is a poor candidate for a variable name. If it is just a prefix of some symbol, then the light goes out when the third character is typed. That said, I appreciate that Eclipse's approach represents a higher level of intelligence. BTW would you care to outline how "full semantic coloring" is implemented? On Fri, Nov 12, 2010 at 8:25 PM, Bill Fenlason <[hidden email]> wrote: David, _______________________________________________ Ibm-netrexx mailing list [hidden email] |
«
Return to ibm-netrexx
|
1 view|%1 views
| Free forum by Nabble | Edit this page |