Menu
▾
▴
netrexx-pipelines
You can subscribe to this list here.
| 2019 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(4) |
Nov
(1) |
Dec
(2) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2020 |
Jan
|
Feb
|
Mar
(4) |
Apr
(3) |
May
(6) |
Jun
(7) |
Jul
(10) |
Aug
(9) |
Sep
(9) |
Oct
(1) |
Nov
(3) |
Dec
(5) |
| 2021 |
Jan
|
Feb
(1) |
Mar
(5) |
Apr
(1) |
May
|
Jun
|
Jul
|
Aug
(4) |
Sep
(3) |
Oct
(9) |
Nov
(1) |
Dec
(14) |
| 2022 |
Jan
(8) |
Feb
|
Mar
|
Apr
|
May
(1) |
Jun
(165) |
Jul
(8) |
Aug
|
Sep
|
Oct
(2) |
Nov
|
Dec
(5) |
| 2023 |
Jan
(19) |
Feb
(14) |
Mar
(2) |
Apr
(3) |
May
|
Jun
(2) |
Jul
(10) |
Aug
(6) |
Sep
|
Oct
|
Nov
|
Dec
|
| 2025 |
Jan
|
Feb
(10) |
Mar
(2) |
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Marc R. <re...@gm...> - 2023-01-16 12:15:23
|
Hi Leslie, method inStream of class Stage has 'Error(13,'inStream no stream selected')'. That seems relevant in your case. FWIW, it's not the compiler that errors, it is a StageError; Marc On 1/15/23 22:26, J Leslie Turriff wrote: > Hi, > This question is slightly related to "Question about delimited strings". > > When I run this pipeline > > | ~/bin/NetRexx > | $ pipe '(getHelp) < getOOrexxDocs.nrx | inside %/* -- Help File Text Start -% %*/ -- > Help File Text End -% | console' > | NOTE: Picked up JDK_JAVA_OPTIONS: -Dawt.useSystemAAFontSettings=on -Dswing.aatext=true > | : > | @14:27:11,leslie@pinto rc=0 > > from the directory where getOOrexx.Docs.nrx resides, it produces the expected output; but > when I run it from a different directory, e.g. > > | ~/bin/NetRexx/Pipelines > | $ pipe '(getHelp) < /home/leslie/bin/NetRexx/getOOrexxDocs.nrx | inside %/* -- Help File > Text Start -% %*/ -- Help File Text End -% | console' > | NOTE: Picked up JDK_JAVA_OPTIONS: -Dawt.useSystemAAFontSettings=on -Dswing.aatext=true > | RC=13 > | @15:24:32,leslie@pinto rc=0 > > instead of the expected output I get someone's RC=13. Note also that the last line shows > that the return code that bash sees is 0. > Where does this RC=13 come from? The only return codes mentioned (In the Progamming > Guide) are 0, 1 or 2 from the compiler. Might it be from Java? > > Leslie > -- > Platform: GNU/Linux > Hardware: x86_64 > Distribution: openSUSE Leap 15.4 > NetRexx: 4.03-GA build 260-20220503-1730 > > > _______________________________________________ > netrexx-pipelines mailing list > net...@li... > https://lists.sourceforge.net/lists/listinfo/netrexx-pipelines |
|
From: J L. T. <jlt...@ma...> - 2023-01-15 21:26:55
|
Hi, This question is slightly related to "Question about delimited strings". When I run this pipeline | ~/bin/NetRexx | $ pipe '(getHelp) < getOOrexxDocs.nrx | inside %/* -- Help File Text Start -% %*/ -- Help File Text End -% | console' | NOTE: Picked up JDK_JAVA_OPTIONS: -Dawt.useSystemAAFontSettings=on -Dswing.aatext=true | : | @14:27:11,leslie@pinto rc=0 from the directory where getOOrexx.Docs.nrx resides, it produces the expected output; but when I run it from a different directory, e.g. | ~/bin/NetRexx/Pipelines | $ pipe '(getHelp) < /home/leslie/bin/NetRexx/getOOrexxDocs.nrx | inside %/* -- Help File Text Start -% %*/ -- Help File Text End -% | console' | NOTE: Picked up JDK_JAVA_OPTIONS: -Dawt.useSystemAAFontSettings=on -Dswing.aatext=true | RC=13 | @15:24:32,leslie@pinto rc=0 instead of the expected output I get someone's RC=13. Note also that the last line shows that the return code that bash sees is 0. Where does this RC=13 come from? The only return codes mentioned (In the Progamming Guide) are 0, 1 or 2 from the compiler. Might it be from Java? Leslie -- Platform: GNU/Linux Hardware: x86_64 Distribution: openSUSE Leap 15.4 NetRexx: 4.03-GA build 260-20220503-1730 |
|
From: J L. T. <jlt...@ma...> - 2023-01-15 21:05:55
|
Jeff, Thanks for troubleshooting this for me. I find that testing this pipeline now works most of the time. (I will open a new thread concerning the times when it fails [for a different reason]). Leslie On 2023-01-15 07:25:46 Jeff Hennick wrote: > <html> > <head> > <meta http-equiv="Content-Type" content="text/html; charset=UTF-8"> > </head> > <body> > <p>Leslie,</p> > <p>I think I may have our solution. Look at this.</p> > <br> > <div > style="display:inline-block;white-space:pre;background-color:#012456;font-f >amily:'Cascadia Code',monospace;font-size:10pt;padding:4px;"><span > style="color:#D3D7CF;background-color:#012456;">PS > C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> </span><span > style="color:#FCE94F;background-color:#012456;">pipe </span><span > style="color:#06989A;background-color:#012456;">"(getHelp) < > getOOrexxDocs.nrx > > | console" > > </span><span style="color:#D3D7CF;background-color:#012456;">some start > lines /* -- Help File Text Start - > some > interior > lines > */ -- Help File Text End - > some end lines > PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> </span><span > style="color:#FCE94F;background-color:#012456;">pipe </span><span > style="color:#06989A;background-color:#012456;">"(getHelp) < > getOOrexxDocs.nrx > > | inside %/* -- Help File Text Start -% %*/ -- Help File Text End -% | > | console" > > </span><span style="color:#D3D7CF;background-color:#012456;">PS > C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes></span></div> > <p></p> > <div class="moz-cite-prefix">No output!</div> > <div class="moz-cite-prefix"><br> > </div> > <div class="moz-cite-prefix">Can you spot the difference?</div> > <div class="moz-cite-prefix"><br> > </div> > <div class="moz-cite-prefix">I have put a space in front of the > comments.</div> > <div class="moz-cite-prefix"><br> > </div> > <blockquote> > <div class="moz-cite-prefix"><i>inside</i> <b>Select > Records between Labels</b></div> > <blockquote> > <div class="moz-cite-prefix"><i>inside</i> selects groups of > records whose first record follows a record that begins with a > specified string. The end of each group can be specified by a > count of records to select, or > as a string that must be at the beginning of the first record > after the group. > </div> > </blockquote> > </blockquote> > <p>So while we have been chasing the delimiter character, all along > the stage and pipe were doing exactly what they should, with any > of the tested delimiters. In thinking about this, I was about to > suggest using the <i>between </i>stage<i>, </i>but it too works > with labels at the beginning of the records.</p> > <p>The suggested alternative in the CMS reference is:</p> > <p> > <blockquote type="cite">3. <i>pick</i> can do what <i>inside </i>does > and much more.</blockquote> > Unfortunately, the NetRexx <i>pick</i> implementation is > currently incomplete. That "much more" is a bear. We are lacking > the <i>from/to</i> options that we need here. Someday, I hope, > someone will -- since this is all volunteer driven open source > software -- improve our <i>pick</i>. Even if only incrementally.<br> > </p> > <p>At the moment, all I can suggest is in your <i>inside</i> > options to put in the exact number of SPACE characters to match > the file's records.</p> > <div > style="display:inline-block;white-space:pre;background-color:#012456;font-f >amily:'Cascadia Code',monospace;font-size:10pt;padding:4px;"><span > style="color:#D3D7CF;background-color:#012456;">PS > C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> </span><span > style="color:#FCE94F;background-color:#012456;">pipe </span><span > style="color:#06989A;background-color:#012456;">"(getHelp) < > getOOrexxDocs.nrx > > | inside % /* -- Help File Text Start -% % */ -- Help File Text End -% | > | console" > > </span><span style="color:#D3D7CF;background-color:#012456;">some > interior > lines > PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes></span></div> > <p></p> > <div class="moz-cite-prefix">Jeff</div> > <div class="moz-cite-prefix"><br> > </div> > <div class="moz-cite-prefix">On 1/13/2023 11:04 PM, J Leslie Turriff > wrote:<br> > </div> > <blockquote type="cite" > cite="mid:202...@ma..."> > <pre class="moz-quote-pre" wrap=""> I have several accounts on my > machine. One returns absolutely nothing; the other sets return code 13. > I wonder, is NetRexx sensitive to the version of Java in use? I recently > upgraded from Java 18 to Java 19; is that significant? > > Leslie > > On 2023-01-12 09:22:11 Jeff Hennick wrote: > </pre> > <blockquote type="cite"> > <pre class="moz-quote-pre" wrap="">Leslie, > I am not sure what is going on. If there is some setting in your system or > shell that is causing the problem. Could it not like the "/*"? (Or "like" > it too well?) I have not been able to duplicate the problem on my Windows > 10 system in PowerShell. I'm not working with the full file -- I'm not sure > where/how to get it -- so first here is my test file, then the results from > your pipe, which was copy and pasted from your posting. PS > C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> pipe "(getHelp) > < getOOrexxDocs.nrx | console" > some start lines > /* -- Help File Text Start - > some > interior > lines > */ -- Help File Text End - > some end lines > PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> pipe > "(getHelp) < getOOrexxDocs.nrx | inside %/* -- Help File Text Start -% > %*/ -- Help File Text End -% | console" some > interior > lines > PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> > > Are you getting some error message? No output with an abnormal return code? > </pre> > </blockquote> > <pre class="moz-quote-pre" wrap=""> > </pre> > </blockquote> > </body> > </html> -- |
|
From: Jeff H. <Je...@Je...> - 2023-01-15 13:25:56
|
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
<p>Leslie,</p>
<p>I think I may have our solution. Look at this.</p>
<br>
<div style="display:inline-block;white-space:pre;background-color:#012456;font-family:'Cascadia Code',monospace;font-size:10pt;padding:4px;"><span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> </span><span style="color:#FCE94F;background-color:#012456;">pipe </span><span style="color:#06989A;background-color:#012456;">"(getHelp) < getOOrexxDocs.nrx
| console"
</span><span style="color:#D3D7CF;background-color:#012456;">some start lines
/* -- Help File Text Start -
some
interior
lines
*/ -- Help File Text End -
some end lines
PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> </span><span style="color:#FCE94F;background-color:#012456;">pipe </span><span style="color:#06989A;background-color:#012456;">"(getHelp) < getOOrexxDocs.nrx
| inside %/* -- Help File Text Start -% %*/ -- Help File Text End -% | console"
</span><span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes></span></div>
<p></p>
<div class="moz-cite-prefix">No output!</div>
<div class="moz-cite-prefix"><br>
</div>
<div class="moz-cite-prefix">Can you spot the difference?</div>
<div class="moz-cite-prefix"><br>
</div>
<div class="moz-cite-prefix">I have put a space in front of the
comments.</div>
<div class="moz-cite-prefix"><br>
</div>
<blockquote>
<div class="moz-cite-prefix"><i>inside</i> <b>Select
Records between Labels</b></div>
<blockquote>
<div class="moz-cite-prefix"><i>inside</i> selects groups of
records whose first record follows a record that begins with a
specified string. The end of each group can be specified by a
count of records to select, or
as a string that must be at the beginning of the first record
after the group.
</div>
</blockquote>
</blockquote>
<p>So while we have been chasing the delimiter character, all along
the stage and pipe were doing exactly what they should, with any
of the tested delimiters. In thinking about this, I was about to
suggest using the <i>between </i>stage<i>, </i>but it too works
with labels at the beginning of the records.</p>
<p>The suggested alternative in the CMS reference is:</p>
<p>
<blockquote type="cite">3. <i>pick</i> can do what <i>inside </i>does
and much more.</blockquote>
Unfortunately, the NetRexx <i>pick</i> implementation is
currently incomplete. That "much more" is a bear. We are lacking
the <i>from/to</i> options that we need here. Someday, I hope,
someone will -- since this is all volunteer driven open source
software -- improve our <i>pick</i>. Even if only incrementally.<br>
</p>
<p>At the moment, all I can suggest is in your <i>inside</i>
options to put in the exact number of SPACE characters to match
the file's records.</p>
<div style="display:inline-block;white-space:pre;background-color:#012456;font-family:'Cascadia Code',monospace;font-size:10pt;padding:4px;"><span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> </span><span style="color:#FCE94F;background-color:#012456;">pipe </span><span style="color:#06989A;background-color:#012456;">"(getHelp) < getOOrexxDocs.nrx
| inside % /* -- Help File Text Start -% % */ -- Help File Text End -% | console"
</span><span style="color:#D3D7CF;background-color:#012456;">some
interior
lines
PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes></span></div>
<p></p>
<div class="moz-cite-prefix">Jeff</div>
<div class="moz-cite-prefix"><br>
</div>
<div class="moz-cite-prefix">On 1/13/2023 11:04 PM, J Leslie Turriff
wrote:<br>
</div>
<blockquote type="cite"
cite="mid:202...@ma...">
<pre class="moz-quote-pre" wrap=""> I have several accounts on my machine. One returns absolutely nothing; the other sets
return code 13.
I wonder, is NetRexx sensitive to the version of Java in use? I recently upgraded from
Java 18 to Java 19; is that significant?
Leslie
On 2023-01-12 09:22:11 Jeff Hennick wrote:
</pre>
<blockquote type="cite">
<pre class="moz-quote-pre" wrap="">Leslie,
I am not sure what is going on. If there is some setting in your system or
shell that is causing the problem. Could it not like the "/*"? (Or "like"
it too well?) I have not been able to duplicate the problem on my Windows
10 system in PowerShell. I'm not working with the full file -- I'm not sure
where/how to get it -- so first here is my test file, then the results from
your pipe, which was copy and pasted from your posting. PS
C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> pipe "(getHelp) <
getOOrexxDocs.nrx | console"
some start lines
/* -- Help File Text Start -
some
interior
lines
*/ -- Help File Text End -
some end lines
PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> pipe "(getHelp)
< getOOrexxDocs.nrx | inside %/* -- Help File Text Start -% %*/ -- Help
File Text End -% | console" some
interior
lines
PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes>
Are you getting some error message? No output with an abnormal return code?
</pre>
</blockquote>
<pre class="moz-quote-pre" wrap="">
</pre>
</blockquote>
</body>
</html>
|
|
From: J L. T. <jlt...@ma...> - 2023-01-14 04:04:43
|
I have several accounts on my machine. One returns absolutely nothing; the other sets return code 13. I wonder, is NetRexx sensitive to the version of Java in use? I recently upgraded from Java 18 to Java 19; is that significant? Leslie On 2023-01-12 09:22:11 Jeff Hennick wrote: > Leslie, > I am not sure what is going on. If there is some setting in your system or > shell that is causing the problem. Could it not like the "/*"? (Or "like" > it too well?) I have not been able to duplicate the problem on my Windows > 10 system in PowerShell. I'm not working with the full file -- I'm not sure > where/how to get it -- so first here is my test file, then the results from > your pipe, which was copy and pasted from your posting. PS > C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> pipe "(getHelp) < > getOOrexxDocs.nrx | console" > some start lines > /* -- Help File Text Start - > some > interior > lines > */ -- Help File Text End - > some end lines > PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> pipe "(getHelp) > < getOOrexxDocs.nrx | inside %/* -- Help File Text Start -% %*/ -- Help > File Text End -% | console" some > interior > lines > PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> > > Are you getting some error message? No output with an abnormal return code? -- |
|
From: Jeff H. <Je...@Je...> - 2023-01-12 15:22:25
|
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
<p>Leslie,</p>
<p>I am not sure what is going on. If there is some setting in your
system or shell that is causing the problem. Could it not like
the "/*"? (Or "like" it too well?) I have not been able to
duplicate the problem on my Windows 10 system in PowerShell.</p>
<p>I'm not working with the full file -- I'm not sure where/how to
get it -- so first here is my test file, then the results from
your pipe, which was copy and pasted from your posting.</p>
<div style="display:inline-block;white-space:pre;background-color:#012456;font-family:'Cascadia Code',monospace;font-size:10pt;padding:4px;"><span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> </span><span style="color:#FCE94F;background-color:#012456;">pipe </span><span style="color:#06989A;background-color:#012456;">"(getHelp) < getOOrexxDocs.nrx |
console"
</span><span style="color:#D3D7CF;background-color:#012456;">some start lines
/* -- Help File Text Start -
some
interior
lines
*/ -- Help File Text End -
some end lines
PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> </span><span style="color:#FCE94F;background-color:#012456;">pipe </span><span style="color:#06989A;background-color:#012456;">"(getHelp) < getOOrexxDocs.nrx |
inside %/* -- Help File Text Start -% %*/ -- Help File Text End -% | console"
</span><span style="color:#D3D7CF;background-color:#012456;">some
interior
lines
PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes></span></div>
<p></p>
<p></p>
<div class="moz-cite-prefix">Are you getting some error message? No
output with an abnormal return code?</div>
<div class="moz-cite-prefix"><br>
</div>
<div class="moz-cite-prefix">On 1/12/2023 1:35 AM, J Leslie Turriff
wrote:<br>
</div>
<blockquote type="cite"
cite="mid:202...@ma...">
<pre class="moz-quote-pre" wrap=""> This is what I tried:
| ~/bin/NetRexx
| $ pipe "(getHelp) < getOOrexxDocs.nrx | inside %/* -- Help File Text Start -% %*/ --
Help File Text End -% | console"
| NOTE: Picked up JDK_JAVA_OPTIONS: -Dawt.useSystemAAFontSettings=on -Dswing.aatext=true
| @00:28:55,leslie@pinto rc=0
I got the same result with !.
Leslie
On 2023-01-11 07:41:10 Jeff Hennick wrote:
</pre>
<blockquote type="cite">
<pre class="moz-quote-pre" wrap="">This is what we have in the NetRexx "Pipelines Stages.html" (
<a class="moz-txt-link-freetext" href="https://sourceforge.net/p/netrexx/code/ci/master/tree/documentation/njpipes">https://sourceforge.net/p/netrexx/code/ci/master/tree/documentation/njpipes</a>
/Pipeline%20Stages.html , unfortunately this is not available currently on
NetRexx.org): Delimited String
|--+--------+--+--delimitedString--+--|
+-STRing-+ +--DString----------+
Examples:
/abc/
,,
xf1f2f3
b11000001
str xabx
A delimited character string is written between two occurrences of a
delimiter character, as a hexadecimal literal, or as a binary literal. The
delimiter cannot be blank and it must not occur within the string. Two
adjacent delimiter characters represent the null string. It is suggested
that a special character be used as the delimiter, but this is not
enforced. A hexadecimal literal is specified by a leading H or X followed
by an even number of hexadecimal digits. A binary literal is specified by a
leading B followed by a string of 0 and 1; the number of binary digits must
an integral multiple of eight. The keyword STRING can be used to specify
that the delimited string contains a string that is terminated by delimiter
characters. And here is a modified version of one of the confirmation tests
using both % and ! as delimiters around /-containing strings: pipe (t)
-- inside dstring dstring ?
literal first one,second one,third one,fourth/one,fifth
one,sixth/one,seventh one,eighth one | split , |
inside %fourth/% !sixth/! |
cons pr /Raw: / |
c1: compare any ~2# 1 inside dstring dstring ~ equal ~*OK*~ notequal
~**FAIL** at rec \\c col \\b.~ less
/Less:\nActual:\n\\p\nExpected:\n\\s\n/ more
/More:\nActual:\n\\p\nExpected:\n\\s\n/ | cons ?
literal fifth one; | -- expected result |
split ; |
c1: ?
gives me
PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> java t
Raw: fifth one
2# 1 inside dstring dstring *OK*
So, I am not able to duplicate your problem at this time. Could you supply
a sample of what is failing?
Jeff
On 1/10/2023 11:30 PM, J Leslie Turriff wrote:
Hi,
In pipeline stage 'inside' I need to match strings that contain /.
I tried using ! and %, but they don't seem to work. What are safe string
delimiters to use in a netrexx pipeline? Can they be any character that
does not appear in the string, or are there restrictions?
Hopefully it works the same way as CMS/TSO Pipelines (from the
CMS/TSO Pipelines Author's
Edition 1.1.12):
| delimitedString:
|
| A delimited character string is written between two occurrences of a
| delimiter character, as a hexadecimal literal, or as a binary literal.
| The delimiter cannot be blank and it must not occur within the string.
| Two a adjacent delimiter characters represent the null string. It is
| suggested that a special character be used as the delimiter, but this is
| not enforced. However, it is advisable not to use alphanumeric
| characters, because a future release might add a keyword or a number as a
| valid option to a built-in program where only a delimited string is valid
| today.
|
| A hexadecimal literal is specified by a leading H or X followed by an even
| number of hexadecimal digits. A binary literal is specified by a leading B
| followed by a string of 0 and 1; the number of binary digits must an
| integral multiple of eight.
|
| The keyword STRING can be used to specify that the delimited string
| contains a string that is terminated by delimiter characters. This acts
| as a placeholder so that any non-blank character can be used as the
| delimiter character. Note that this use of the keyword is in addition to
| a keyword that is recognised by a built-in program. (Thus, split string
| string xabcx)
|
| Examples:
| /abc/
| ,,
| xf1f2f3
| b11000001
| str xabx
Leslie
--
_______________________________________________
netrexx-pipelines mailing list
<a class="moz-txt-link-abbreviated" href="mailto:net...@li...">net...@li...</a>
<a class="moz-txt-link-freetext" href="https://lists.sourceforge.net/lists/listinfo/netrexx-pipelines">https://lists.sourceforge.net/lists/listinfo/netrexx-pipelines</a>
</pre>
</blockquote>
<pre class="moz-quote-pre" wrap="">
</pre>
</blockquote>
</body>
</html>
|
|
From: J L. T. <jlt...@ma...> - 2023-01-12 06:35:38
|
This is what I tried: | ~/bin/NetRexx | $ pipe "(getHelp) < getOOrexxDocs.nrx | inside %/* -- Help File Text Start -% %*/ -- Help File Text End -% | console" | NOTE: Picked up JDK_JAVA_OPTIONS: -Dawt.useSystemAAFontSettings=on -Dswing.aatext=true | @00:28:55,leslie@pinto rc=0 I got the same result with !. Leslie On 2023-01-11 07:41:10 Jeff Hennick wrote: > This is what we have in the NetRexx "Pipelines Stages.html" ( > https://sourceforge.net/p/netrexx/code/ci/master/tree/documentation/njpipes >/Pipeline%20Stages.html , unfortunately this is not available currently on > NetRexx.org): Delimited String > |--+--------+--+--delimitedString--+--| > +-STRing-+ +--DString----------+ > Examples: > /abc/ > ,, > xf1f2f3 > b11000001 > str xabx > > A delimited character string is written between two occurrences of a > delimiter character, as a hexadecimal literal, or as a binary literal. The > delimiter cannot be blank and it must not occur within the string. Two > adjacent delimiter characters represent the null string. It is suggested > that a special character be used as the delimiter, but this is not > enforced. A hexadecimal literal is specified by a leading H or X followed > by an even number of hexadecimal digits. A binary literal is specified by a > leading B followed by a string of 0 and 1; the number of binary digits must > an integral multiple of eight. The keyword STRING can be used to specify > that the delimited string contains a string that is terminated by delimiter > characters. And here is a modified version of one of the confirmation tests > using both % and ! as delimiters around /-containing strings: pipe (t) > -- inside dstring dstring ? > literal first one,second one,third one,fourth/one,fifth > one,sixth/one,seventh one,eighth one | split , | > inside %fourth/% !sixth/! | > cons pr /Raw: / | > c1: compare any ~2# 1 inside dstring dstring ~ equal ~*OK*~ notequal > ~**FAIL** at rec \\c col \\b.~ less > /Less:\nActual:\n\\p\nExpected:\n\\s\n/ more > /More:\nActual:\n\\p\nExpected:\n\\s\n/ | cons ? > literal fifth one; | -- expected result | > split ; | > c1: ? > > gives me > PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> java t > Raw: fifth one > 2# 1 inside dstring dstring *OK* > > So, I am not able to duplicate your problem at this time. Could you supply > a sample of what is failing? > > > > Jeff > > > On 1/10/2023 11:30 PM, J Leslie Turriff wrote: > > Hi, > In pipeline stage 'inside' I need to match strings that contain /. > I tried using ! and %, but they don't seem to work. What are safe string > delimiters to use in a netrexx pipeline? Can they be any character that > does not appear in the string, or are there restrictions? > Hopefully it works the same way as CMS/TSO Pipelines (from the > CMS/TSO Pipelines Author's > > Edition 1.1.12): > | delimitedString: > | > | A delimited character string is written between two occurrences of a > | delimiter character, as a hexadecimal literal, or as a binary literal. > | The delimiter cannot be blank and it must not occur within the string. > | Two a adjacent delimiter characters represent the null string. It is > | suggested that a special character be used as the delimiter, but this is > | not enforced. However, it is advisable not to use alphanumeric > | characters, because a future release might add a keyword or a number as a > | valid option to a built-in program where only a delimited string is valid > | today. > | > | A hexadecimal literal is specified by a leading H or X followed by an even > | number of hexadecimal digits. A binary literal is specified by a leading B > | followed by a string of 0 and 1; the number of binary digits must an > | integral multiple of eight. > | > | The keyword STRING can be used to specify that the delimited string > | contains a string that is terminated by delimiter characters. This acts > | as a placeholder so that any non-blank character can be used as the > | delimiter character. Note that this use of the keyword is in addition to > | a keyword that is recognised by a built-in program. (Thus, split string > | string xabcx) > | > | Examples: > | /abc/ > | ,, > | xf1f2f3 > | b11000001 > | str xabx > > Leslie > -- > > > > _______________________________________________ > netrexx-pipelines mailing list > net...@li... > https://lists.sourceforge.net/lists/listinfo/netrexx-pipelines -- |
|
From: J L. T. <jlt...@ma...> - 2023-01-12 05:42:12
|
You're right, I sent this to the wrong list. I apologize. Leslie On 2023-01-11 07:42:49 Jeff Hennick wrote: > <html> > <head> > <meta http-equiv="Content-Type" content="text/html; charset=UTF-8"> > </head> > <body> > <p>This might get better responses out on a more general NetRexx > list. I'm not qualified to answer.<br> > </p> > <div class="moz-cite-prefix">On 1/10/2023 11:36 PM, J Leslie Turriff > wrote:<br> > </div> > <blockquote type="cite" > cite="mid:202...@ma..."> > <pre class="moz-quote-pre" wrap="">Hi, > > Say I want to use an indexed string with two levels of indices: > | list[level1,level2] > > Can I use > > | list.exists(index) > > to test just the first level? Or would it be > > | list.exists(index,) > > ? > > Leslie > -- > Platform: GNU/Linux > Hardware: x86_64 > Distribution: openSUSE Leap 15.4 > NetRexx: 4.03-GA build 260-20220503-1730 > > > _______________________________________________ > netrexx-pipelines mailing list > <a class="moz-txt-link-abbreviated" > href="mailto:net...@li...">netrexx-pipelines@lis >ts.sourceforge.net</a> <a class="moz-txt-link-freetext" > href="https://lists.sourceforge.net/lists/listinfo/netrexx-pipelines">https >://lists.sourceforge.net/lists/listinfo/netrexx-pipelines</a> </pre> > </blockquote> > </body> > </html> -- |
|
From: Marc R. <re...@gm...> - 2023-01-11 14:05:39
|
Hi Leslie,
by defining list[level1, level2] you do not define list[level1].
if you want to check existence of level1, you must assign it a value.
Check the following:
a=''
b=''
a['level1', 'level2']=1
b=a['level1']
if a.exists('level1') then say 'a[''level1''] exists'
if b.exists('level2') then say 'a[''level1'', ''level2''] exists'
a['level1'] = 1
if a.exists('level1') then say 'now a[''level1''] exists'
I think that is consistent with classic rexx's stems.
Marc
On 1/11/23 14:42, Jeff Hennick wrote:
> This might get better responses out on a more general NetRexx list. I'm not qualified to answer.
>
> On 1/10/2023 11:36 PM, J Leslie Turriff wrote:
>> Hi,
>> Say I want to use an indexed string with two levels of indices:
>> | list[level1,level2]
>> Can I use
>> | list.exists(index)
>> to test just the first level? Or would it be
>> | list.exists(index,)
>> ?
>>
>> Leslie
>> --
>> Platform: GNU/Linux
>> Hardware: x86_64
>> Distribution: openSUSE Leap 15.4
>> NetRexx: 4.03-GA build 260-20220503-1730
>>
>>
>> _______________________________________________
>> netrexx-pipelines mailing list
>> net...@li...
>> https://lists.sourceforge.net/lists/listinfo/netrexx-pipelines
>
>
> _______________________________________________
> netrexx-pipelines mailing list
> net...@li...
> https://lists.sourceforge.net/lists/listinfo/netrexx-pipelines
|
|
From: Jeff H. <Je...@Je...> - 2023-01-11 13:43:00
|
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
<p>This might get better responses out on a more general NetRexx
list. I'm not qualified to answer.<br>
</p>
<div class="moz-cite-prefix">On 1/10/2023 11:36 PM, J Leslie Turriff
wrote:<br>
</div>
<blockquote type="cite"
cite="mid:202...@ma...">
<pre class="moz-quote-pre" wrap="">Hi,
Say I want to use an indexed string with two levels of indices:
| list[level1,level2]
Can I use
| list.exists(index)
to test just the first level? Or would it be
| list.exists(index,)
?
Leslie
--
Platform: GNU/Linux
Hardware: x86_64
Distribution: openSUSE Leap 15.4
NetRexx: 4.03-GA build 260-20220503-1730
_______________________________________________
netrexx-pipelines mailing list
<a class="moz-txt-link-abbreviated" href="mailto:net...@li...">net...@li...</a>
<a class="moz-txt-link-freetext" href="https://lists.sourceforge.net/lists/listinfo/netrexx-pipelines">https://lists.sourceforge.net/lists/listinfo/netrexx-pipelines</a>
</pre>
</blockquote>
</body>
</html>
|
|
From: Jeff H. <Je...@Je...> - 2023-01-11 13:41:19
|
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
<p>This is what we have in the NetRexx "Pipelines Stages.html" (
<a class="moz-txt-link-freetext" href="https://sourceforge.net/p/netrexx/code/ci/master/tree/documentation/njpipes/Pipeline%20Stages.html">https://sourceforge.net/p/netrexx/code/ci/master/tree/documentation/njpipes/Pipeline%20Stages.html</a>
, unfortunately this is not available currently on NetRexx.org):</p>
<p>
<blockquote type="cite">
<h3>Delimited String</h3>
</blockquote>
<font face="monospace">|--+--------+--+--<i><span class="var">delimitedString</span></i>--+--|
</font></p>
<blockquote type="cite">
<div style="position: relative;">
<div class="desc">
<div class="p-ddiv"><font face="monospace"> +-STRing-+ +--<i><span
class="var">DString</span></i>----------+
</font></div>
<p><b>Examples:</b></p>
<pre class="examp"> /abc/
,,
xf1f2f3
b11000001
str xabx
</pre>
<p>A delimited character string is written between two
occurrences of a
delimiter character, as a hexadecimal literal, or as a
binary literal. The
delimiter cannot be blank and it must not occur within the
string. Two
adjacent delimiter characters represent the null string. It
is suggested
that a special character be used as the delimiter, but this
is not
enforced.</p>
<p>A hexadecimal literal is specified by a leading H or X
followed by an
even number of hexadecimal digits. A binary literal is
specified by a
leading B followed by a string of 0 and 1; the number of
binary digits
must an integral multiple of eight.</p>
<p>The keyword STRING can be used to specify that the
delimited string
contains a string that is terminated by delimiter
characters.</p>
</div>
</div>
</blockquote>
<p>And here is a modified version of one of the confirmation tests
using both % and ! as delimiters around /-containing strings:</p>
<p><font face="monospace">pipe (t)<br>
-- inside dstring dstring ?<br>
literal first one,second one,third one,fourth/one,fifth
one,sixth/one,seventh one,eighth one |<br>
split , |<br>
<b>inside %fourth/% !sixth/!</b> |<br>
cons pr /Raw: / |<br>
c1: compare any ~2# 1 inside dstring dstring ~ equal ~*OK*~
notequal ~**FAIL** at rec \\c col \\b.~ less
/Less:\nActual:\n\\p\nExpected:\n\\s\n/ more
/More:\nActual:\n\\p\nExpected:\n\\s\n/ |<br>
cons ?<br>
literal fifth one; | -- expected result |<br>
split ; |<br>
c1: ? </font><br>
</p>
<p>gives me</p>
<div style="display:inline-block;white-space:pre;background-color:#012456;font-family:'Cascadia Code',monospace;font-size:10pt;padding:4px;"><span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> </span><span style="color:#FCE94F;background-color:#012456;">java </span><span style="color:#D3D7CF;background-color:#012456;">t
Raw: fifth one
2# 1 inside dstring dstring *OK*</span></div>
<p></p>
<div class="moz-cite-prefix">So, I am not able to duplicate your
problem at this time. Could you supply a sample of what is
failing?<br>
</div>
<div class="moz-cite-prefix"><br>
</div>
<div class="moz-cite-prefix">Jeff</div>
<div class="moz-cite-prefix"><br>
</div>
<div class="moz-cite-prefix">On 1/10/2023 11:30 PM, J Leslie Turriff
wrote:<br>
</div>
<blockquote type="cite"
cite="mid:202...@ma...">
<pre class="moz-quote-pre" wrap="">Hi,
In pipeline stage 'inside' I need to match strings that contain /. I tried using !
and %, but they don't seem to work. What are safe string delimiters to use in a netrexx
pipeline? Can they be any character that does not appear in the string, or are there
restrictions?
Hopefully it works the same way as CMS/TSO Pipelines (from the CMS/TSO Pipelines Author's
Edition 1.1.12):
| delimitedString:
|
| A delimited character string is written between two occurrences of a
| delimiter character, as a hexadecimal literal, or as a binary literal. The
| delimiter cannot be blank and it must not occur within the string. Two a
| adjacent delimiter characters represent the null string. It is suggested
| that a special character be used as the delimiter, but this is not
| enforced. However, it is advisable not to use alphanumeric characters,
| because a future release might add a keyword or a number as a valid option
| to a built-in program where only a delimited string is valid today.
|
| A hexadecimal literal is specified by a leading H or X followed by an even
| number of hexadecimal digits. A binary literal is specified by a leading B
| followed by a string of 0 and 1; the number of binary digits must an
| integral multiple of eight.
|
| The keyword STRING can be used to specify that the delimited string
| contains a string that is terminated by delimiter characters. This acts as
| a placeholder so that any non-blank character can be used as the delimiter
| character. Note that this use of the keyword is in addition to a keyword
| that is recognised by a built-in program. (Thus, split string string xabcx)
|
| Examples:
| /abc/
| ,,
| xf1f2f3
| b11000001
| str xabx
Leslie
--
_______________________________________________
netrexx-pipelines mailing list
<a class="moz-txt-link-abbreviated" href="mailto:net...@li...">net...@li...</a>
<a class="moz-txt-link-freetext" href="https://lists.sourceforge.net/lists/listinfo/netrexx-pipelines">https://lists.sourceforge.net/lists/listinfo/netrexx-pipelines</a>
</pre>
</blockquote>
</body>
</html>
|
|
From: J L. T. <jlt...@ma...> - 2023-01-11 04:36:56
|
Hi, Say I want to use an indexed string with two levels of indices: | list[level1,level2] Can I use | list.exists(index) to test just the first level? Or would it be | list.exists(index,) ? Leslie -- Platform: GNU/Linux Hardware: x86_64 Distribution: openSUSE Leap 15.4 NetRexx: 4.03-GA build 260-20220503-1730 |
|
From: J L. T. <jlt...@ma...> - 2023-01-11 04:30:50
|
Hi, In pipeline stage 'inside' I need to match strings that contain /. I tried using ! and %, but they don't seem to work. What are safe string delimiters to use in a netrexx pipeline? Can they be any character that does not appear in the string, or are there restrictions? Hopefully it works the same way as CMS/TSO Pipelines (from the CMS/TSO Pipelines Author's Edition 1.1.12): | delimitedString: | | A delimited character string is written between two occurrences of a | delimiter character, as a hexadecimal literal, or as a binary literal. The | delimiter cannot be blank and it must not occur within the string. Two a | adjacent delimiter characters represent the null string. It is suggested | that a special character be used as the delimiter, but this is not | enforced. However, it is advisable not to use alphanumeric characters, | because a future release might add a keyword or a number as a valid option | to a built-in program where only a delimited string is valid today. | | A hexadecimal literal is specified by a leading H or X followed by an even | number of hexadecimal digits. A binary literal is specified by a leading B | followed by a string of 0 and 1; the number of binary digits must an | integral multiple of eight. | | The keyword STRING can be used to specify that the delimited string | contains a string that is terminated by delimiter characters. This acts as | a placeholder so that any non-blank character can be used as the delimiter | character. Note that this use of the keyword is in addition to a keyword | that is recognised by a built-in program. (Thus, split string string xabcx) | | Examples: | /abc/ | ,, | xf1f2f3 | b11000001 | str xabx Leslie -- |
|
From: Jeff H. <Je...@Je...> - 2023-01-11 02:15:55
|
<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
</head>
<body>
<p>I have finally added the <b>ALL</b> stage to NetRexx Pipelines.</p>
<p>This almost matches the CMS Pipelines version, except for some
symbol notation. NetRexx & Pipelines "eat" the "|" and "("
& ")" before the stage gets to see them. So, for OR use "!"
and for grouping use "[" and "]". IBM uses the 3270's "¬" for
NOT, here KEX's "^" or NetRexx's own "\" -- but it needs to be
doubled up as "\\" -- also work.</p>
<p>The internal logic of CMS is to construct a new sub-pipe with
LOCATE and NLOCATE stages; NetRexx simply uses the logic handling
built into NetRexx. This is noticeable in the output of %debug
and %dump options.<br>
</p>
<p>There are also confirmation tests, and the HTML Stages file has
been updated with this. That file now also has the ability to
show only the changes by version number (4.05, 3.11, and 3.09). <br>
</p>
<p>This update has been pushed for the next NetRexx release.</p>
<p>[I think this completes all the major stages from CMS that need
to be in NetRexx.]</p>
<p>Jeff</p>
<table>
<tbody>
<tr class="c405" style="">
<td class="name"><span id="all">all
<br>
<span class="n405">
</span></span></td>
<td class="desc">Select Lines Containing Strings (or Not)
<div>
<div class="p-ddiv">
<div class="p-diag box">
<font face="monospace">▶▶--ALL---+----------+--|
expression |---▶◄ <br>
</font></div>
<div class="p-diag box"><font face="monospace">
+--%debug--+ (6) <br>
</font></div>
<div class="p-diag box"><font face="monospace">
+--%dump---+ (7)
</font></div>
</div>
<div class="p-note">
Notes:
<ul>
<li>(1) "expression" consists of one or more
delimitedstrings separated by logical ANDs, ORs, and
NOTs, and grouped, if needed, by parentheses.
</li>
<li>(2) "&" is used for AND.
</li>
<li>(3) Since "|" is the default stage separator, "!"
may be used for OR.
</li>
<li>(4) Since NetRexx uses "(" and ")" for options --
which are not used in the ALL stage -- "[" and "]"
must be used for parentheses.
</li>
<li>(5) CMS Pipelines, having originated on 3270
terminals, uses "¬" for NOT. This symbol is not
readily typed on terminals running NetRexx
Pipelines, so as alternatives, "\\", used by
NetRexx, (it needs to be doubled to "escape" it) or
"^", used by KEX, NOT symbols may be used as
alternatives.
</li>
<li>(6) %debug (must be lowercase) NetRexx Pipelines
writes the logic line to the file ALL.DEBUG in the
current directory. Windows may make it all.debug .
CMS Pipelines writes the constructed pipeline (of
LOCATE and NLOCATE stages) to ALL DEBUG A.
</li>
<li>(7) %dump (must be lowercase) - writes to the
primary output stream as the first record. NetRexx
Pipelines writes the logic line. CMS Pipelines
writes constructed pipeline.
</li>
<li>CMS Pipelines uses is own logic order. NetRexx
Pipelines uses regular NetRexx
</li>
</ul>
Examples:
<ul>
<li class="examp"> literal NetRexx is Good,NetRexx is
Great,NetRexx is Fantastic |
<br>
split , |
<br>
<span class="red">all /a/</span> |
<br>
cons
<br>
<br>
▶NetRexx is Great
<br>
▶NetRexx is Fantastic
<br>
</li>
<li class="examp">
literal NetRexx is Good,NetRexx is Great,NetRexx is
Fantastic |
<br>
split , |
<br>
<span class="red">all / G/ & [/oo/ ! /F/]</span>
|
<br>
cons
<br>
<br>
▶NetRexx is Good
<br>
</li>
<li class="examp">
literal NetRexx is Good,NetRexx is Great,NetRexx is
Fantastic |
<br>
split , |
<br>
<span class="red">all /R/ & [/oo/ ! /F/]</span>
|
<br>
cons
<br>
<br>
▶NetRexx is Good
<br>
▶NetRexx is Fantastic
</li>
</ul>
</div>
</div>
</td>
</tr>
</tbody>
</table>
</body>
</html>
|
|
From: J L. T. <jlt...@ma...> - 2023-01-05 11:37:19
|
Wow! Now I can hardly wait for the next release! Thank you so much. Leslie On 2023-01-04 13:10:24 Jeff Hennick wrote: > I have made a major update to the SPEC stage (and expect it is the last > for it for a while). It will be in the next release of NetRexx. It is > currently available from SourceForge. Since it uses the "stageExit" > method, it can only be compiled into the full NetRexxC.jar. > > CMS Pipelines has built into the SPEC stage its own programming language. > It is Rexx-like-but-not-quite-Rexx. For NetRexx Pipelines, I did not -- > and recommend we do not -- attempt to duplicate it. What we have built in, > and I have used instead, is the worlds best scripting language: NetRexx. > This gives all the power, but with a somewhat different syntax, of the CMS > version. I have incorporated this into the "Conversion" phase, with the > key word NETREXX (or NR) and a delimitedString containing the NetRexx > source. In running, this code is encapsulated in a method. The data > selected in the Input phase is available as the variable data. And > whatever is returned is passed to the Output phase of the stage. (As a > convenience, if the last statement is not RETURN, the statement "return > data" is automatically added.) So all of these pipes work the same (giving > cba): pipe "literal abc | spec 1-* NR /return data.reverse/ 1 | cons" > pipe "literal abc | spec 1-* NR @rev = data.reverse; return rev@ 1 | cons" > pipe "literal abc | spec 1-* NR %data = data.reverse% 1 | cons" > CMS also has COUNTERs. Here in NetRexx, COUNTER is a class property and > can be used as a full fledged NetRexx indexed variable, for example > counter[3] or counter['gold']. Counters are persistent for the life of the > stage and shared across all NR invocations within a SPEC stage. (CMS has a > limited number of counters available, and suggests specifying the number > needed in the SPEC options. NetRexx has no such limitations and the > COUNTERS option is ignored.) This NetRexx code can also use Identified > Fields, below, similarly to counters. The syntax is field["a"], for > example. Like CMS, fields are identified by a single letter, case > sensitive, so up to 52 are available. DANGER: Since the full power of > NetRexx is available, you can get yourself in trouble! Also, this code is > three levels deep, and error messages may point to the wrong thing, at the > "wrong time." Debugging can be a bear. You have been warned. > > These new Field Identifiers are are specified by a letter and colon before > an input range. (There may be no blank between the letter and colon, there > may be a space between the identifier and the range specifier.) The letter > is any of the 52 a-zA-Z, and is case sensitive. > > Another new feature may make the sometimes dense options string a little > easier to read. Groups of Id-Input-Convert-Output may now have a > non-functional comma separating them. (At least at this time, the comma > must have a BLANK after it, and optionally before it and must be in the > right place.) > > This update now fully supports multiple output streams as well as inputs. > The output range of "." has been added, it is of use with Identified Fields > and NR. It causes no output from that input range. Also,in NetRexx > Pipelines only, an input range of "." gives a data field of "". This may > be of use with the new NETREXX converter phase. > > There are confirmation tests for (I think) all of the relevant options in > the examples/pipes directory. > > The documentation is now: > > > . > . > ========= > Jeff -- |
|
From: Jeff H. <Je...@Je...> - 2023-01-04 19:10:41
|
<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
</head>
<body>
I have made a major update to the <b>SPEC</b> stage (and expect it
is the last for it for a while). It will be in the next release of
NetRexx. It is currently available from SourceForge. Since it uses
the "stageExit" method, it can only be compiled into the full
NetRexxC.jar.<br>
<p>CMS Pipelines has built into the SPEC stage its own programming
language. It is Rexx-like-but-not-quite-Rexx. For NetRexx
Pipelines, I did not -- and recommend we do not -- attempt to
duplicate it. What we have built in, and I have used instead, is
the worlds best scripting language: <b>NetRexx</b>. This gives
all the power, but with a somewhat different syntax, of the CMS
version. I have incorporated this into the "Conversion" phase,
with the key word <b>NETREXX</b> (or <b>NR</b>) and a
delimitedString containing the NetRexx source. In running, this
code is encapsulated in a method. The data selected in the Input
phase is available as the variable <font face="monospace"><b>data</b></font>.
And whatever is returned is passed to the Output phase of the
stage. (As a convenience, if the last statement is not RETURN,
the statement "return data" is automatically added.) So all of
these pipes work the same (giving cba):</p>
<ul>
<li><font face="monospace">pipe "literal abc | <b>spec 1-* NR
/return data.reverse/</b> 1 | cons"</font></li>
<li><font face="monospace">pipe "literal abc | <b>spec 1-* NR
@rev = data.reverse; return rev@</b> 1 | cons"</font></li>
<li><font face="monospace">pipe "literal abc | <b>spec 1-* NR
%data = data.reverse%</b> 1 | cons"</font></li>
</ul>
<p>CMS also has <b>COUNTER</b>s. Here in NetRexx, <b>COUNTER</b>
is a class property and can be used as a full fledged NetRexx
indexed variable, for example <font face="monospace">counter[3] </font>or
<font face="monospace">counter['gold']</font>. Counters are
persistent for the life of the stage and shared across all <b>NR</b>
invocations within a SPEC stage. (CMS has a limited number of
counters available, and suggests specifying the number needed in
the SPEC options. NetRexx has no such limitations and the
COUNTERS option is ignored.)</p>
<p>This NetRexx code can also use Identified Fields, below,
similarly to counters. The syntax is <font face="monospace"><b>field["a"]</b></font>,
for example. Like CMS, fields are identified by a single letter,
case sensitive, so up to 52 are available.</p>
<blockquote>
<p><b>DANGER</b>: Since the full power of NetRexx is available,
you can get yourself in trouble! Also, this code is three
levels deep, and error messages may point to the wrong thing, at
the "wrong time." Debugging can be a bear. You have been
warned.<br>
</p>
</blockquote>
<p>These new Field Identifiers are are specified by a letter and
colon before an input range. (There may be no blank between the
letter and colon, there may be a space between the identifier and
the range specifier.) The letter is any of the 52 a-zA-Z, and is
case sensitive.<br>
</p>
<p>Another new feature may make the sometimes dense options string a
little easier to read. Groups of Id-Input-Convert-Output may now
have a non-functional comma separating them. (At least at this
time, the comma must have a BLANK after it, and optionally before
it and must be in the right place.) <br>
</p>
<p>This update now fully supports multiple output streams as well as
inputs. The output range of "." has been added, it is of use with
Identified Fields and NR. It causes no output from that input
range. Also,in NetRexx Pipelines only, an input range of "."
gives a data field of "". This may be of use with the new <b>NETREXX</b>
converter phase.<br>
</p>
<p>There are confirmation tests for (I think) all of the relevant
options in the <font face="monospace">examples/pipes</font>
directory.<br>
</p>
<p>The documentation is now:<br>
</p>
<p> <img src="cid:par...@Je..." alt=""></p>
<p><img src="cid:par...@Je..." alt=""></p>
<p>.<img src="cid:par...@Je..." alt=""></p>
<p><img src="cid:par...@Je..." alt=""></p>
<p>.<img src="cid:par...@Je..." alt=""></p>
<p>=========</p>
<p>Jeff<br>
</p>
</body>
</html> |
|
From: René J. <rvj...@xs...> - 2023-01-04 15:46:20
|
Interesting … yes, I think this is one step further in the evolution of the pipes processor from file/pipe-generate/compile only to ‘just in time’. It is amazing that this takes almost no time now (on my fastest computer-that is). This is a very good idea - thank you Marc. René. > On 4 Jan 2023, at 11:24, Marc Remes <re...@gm...> wrote: > > I always found it strange that NetRexx Pipelines did not accept its source from stdin, counter-intuitive on *nix. > I've updated runner.nrx to do just that. > > Previously, when executing org.netrexx.njpipes.pipes.runner without an argument, a message 'specify a pipeline' was given. > Now, it will read the pipeline from stdin. > Also, a -h option was added, and the parsing of the pipeline options - between () - have been sanitised. > > bin/pipe does not have to readline anymore, as runner does it now. > This allows addressing pipelines directly : > > << > $ cat pipetest.nrx > > address system > > 'pipe "(sep ! end %) literal one ! cons%"' > > 'echo "literal two | cons" | pipe' > > 'echo "literal three | cons" >pipein.txt' > > 'pipe -f pipein.txt' > > > address pipe '(pipetest sep ! end ") literal four ! cons"' > > address pipe > > '(pipetest sep ! end ") literal five ! cons"' > > '(pipetest sep | end ?)'- -- split lines > 'literal six |'- -- write comments after - > 'cons ?' > > '(pipetest sep | end ?)'- > ' literal seven '- > '| cons' > > $ nrc -exec pipetest > NetRexx portable processor 4.05-alpha build 2,078-20230104-1557 > Copyright (c) RexxLA, 2011,2022. All rights reserved. > Parts Copyright (c) IBM Corporation, 1995,2008. > Program pipetest.nrx > ===== Exec: pipetest ===== > one > two > three > four > five > six > seven > Processing of 'pipetest.nrx' complete > >>> > > Mileage on Windows may vary.. > > Marc > > > > _______________________________________________ > netrexx-pipelines mailing list > net...@li... > https://lists.sourceforge.net/lists/listinfo/netrexx-pipelines |
|
From: Marc R. <re...@gm...> - 2023-01-04 15:24:30
|
I always found it strange that NetRexx Pipelines did not accept its source from stdin, counter-intuitive on *nix. I've updated runner.nrx to do just that. Previously, when executing org.netrexx.njpipes.pipes.runner without an argument, a message 'specify a pipeline' was given. Now, it will read the pipeline from stdin. Also, a -h option was added, and the parsing of the pipeline options - between () - have been sanitised. bin/pipe does not have to readline anymore, as runner does it now. This allows addressing pipelines directly : << $ cat pipetest.nrx address system 'pipe "(sep ! end %) literal one ! cons%"' 'echo "literal two | cons" | pipe' 'echo "literal three | cons" >pipein.txt' 'pipe -f pipein.txt' address pipe '(pipetest sep ! end ") literal four ! cons"' address pipe '(pipetest sep ! end ") literal five ! cons"' '(pipetest sep | end ?)'- -- split lines 'literal six |'- -- write comments after - 'cons ?' '(pipetest sep | end ?)'- ' literal seven '- '| cons' $ nrc -exec pipetest NetRexx portable processor 4.05-alpha build 2,078-20230104-1557 Copyright (c) RexxLA, 2011,2022. All rights reserved. Parts Copyright (c) IBM Corporation, 1995,2008. Program pipetest.nrx ===== Exec: pipetest ===== one two three four five six seven Processing of 'pipetest.nrx' complete >> Mileage on Windows may vary.. Marc |
|
From: Jeff H. <Je...@Je...> - 2022-12-25 17:40:44
|
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
<p>Marc,</p>
<p>Excellent! That works. Thank you.<br>
</p>
<p>Incidentally, I am continuing work on the SPEC stage, including
what I think is a very useful extension.</p>
<p>Jeff<br>
</p>
<div class="moz-cite-prefix">On 12/25/2022 10:34 AM, Marc Remes
wrote:<br>
</div>
<blockquote type="cite"
cite="mid:90c...@gm...">Hi Jeff,
<br>
<br>
the example is wrong but the text is correct. Options -gen and
-keep are used on the pipc command.
<br>
<br>
The example should read 'pipc -gen -keep testpipe', where
testpipe.njp defines the pipeline(s).
<br>
<br>
Marc
<br>
<br>
<br>
On 12/24/22 18:23, Jeff Hennick wrote:
<br>
<blockquote type="cite">I am getting an error I can not trace.
I'd like to try to compile a pipe and keep the JAVA file, but am
failing.
<br>
<br>
The Pipelines Guide and Reference, page 33, says
<br>
<br>
The following options can be used on the pipc command, in
addition to the ones specified in the previous chapter for the
Pipes Runner:
<br>
<br>
-gen Generate the NetREXX source file. The pipeline needs a
name.
<br>
<br>
-keep Keep the Java source which is generated from the
NetREXX source.
<br>
<br>
Example: pipe (testpipe -gen -keep)
<br>
<br>
This will generate the NetREXX source as well as keep the
java source.
<br>
<br>
First off, I am assuming, in the Example, that should be pipc
and be followed by the name of the njp file. But Windows
PowerShell does not like any of the combinations I try.
<br>
<br>
Any pointers in getting the JAVA file would be appreciated.
<br>
<br>
-------
<br>
<br>
The error I am getting trying to compile a test pipe -- with a
test version of the SPEC stage, is
<br>
<br>
java.lang.NullPointerException at
org.netrexx.njpipes.stages.specz.stageExit(specz.java:1198) at
org.netrexx.njpipes.pipes.compiler.compile(compiler.java:574) at
org.netrexx.njpipes.pipes.compiler.compile(compiler.java:166) at
org.netrexx.njpipes.pipes.pipe2nrx.run(pipe2nrx.java:47) at
org.netrexx.njpipes.pipes.ThreadPool.run(ThreadPool.java:76) at
java.lang.Thread.run(Unknown Source) Exiting due to uncaught
Exception null
<br>
<br>
It is that "specz.java:1198" that I'd like to locate and tie
back to my NR source.
<br>
<br>
<br>
<br>
_______________________________________________
<br>
netrexx-pipelines mailing list
<br>
<a class="moz-txt-link-abbreviated" href="mailto:net...@li...">net...@li...</a>
<br>
<a class="moz-txt-link-freetext" href="https://lists.sourceforge.net/lists/listinfo/netrexx-pipelines">https://lists.sourceforge.net/lists/listinfo/netrexx-pipelines</a>
<br>
</blockquote>
<br>
<fieldset class="moz-mime-attachment-header"></fieldset>
<br>
<fieldset class="moz-mime-attachment-header"></fieldset>
<pre class="moz-quote-pre" wrap="">_______________________________________________
netrexx-pipelines mailing list
<a class="moz-txt-link-abbreviated" href="mailto:net...@li...">net...@li...</a>
<a class="moz-txt-link-freetext" href="https://lists.sourceforge.net/lists/listinfo/netrexx-pipelines">https://lists.sourceforge.net/lists/listinfo/netrexx-pipelines</a>
</pre>
</blockquote>
</body>
</html>
|
|
From: Marc R. <re...@gm...> - 2022-12-25 15:35:04
|
Hi Jeff, the example is wrong but the text is correct. Options -gen and -keep are used on the pipc command. The example should read 'pipc -gen -keep testpipe', where testpipe.njp defines the pipeline(s). Marc On 12/24/22 18:23, Jeff Hennick wrote: > I am getting an error I can not trace. I'd like to try to compile a pipe and keep the JAVA file, but am failing. > > The Pipelines Guide and Reference, page 33, says > > The following options can be used on the pipc command, in addition to the ones specified in the previous chapter for the Pipes Runner: > > -gen Generate the NetREXX source file. The pipeline needs a name. > > -keep Keep the Java source which is generated from the NetREXX source. > > Example: pipe (testpipe -gen -keep) > > This will generate the NetREXX source as well as keep the java source. > > First off, I am assuming, in the Example, that should be pipc and be followed by the name of the njp file. But Windows PowerShell does not like any of the combinations I try. > > Any pointers in getting the JAVA file would be appreciated. > > ------- > > The error I am getting trying to compile a test pipe -- with a test version of the SPEC stage, is > > java.lang.NullPointerException at org.netrexx.njpipes.stages.specz.stageExit(specz.java:1198) at org.netrexx.njpipes.pipes.compiler.compile(compiler.java:574) at org.netrexx.njpipes.pipes.compiler.compile(compiler.java:166) at org.netrexx.njpipes.pipes.pipe2nrx.run(pipe2nrx.java:47) at org.netrexx.njpipes.pipes.ThreadPool.run(ThreadPool.java:76) at java.lang.Thread.run(Unknown Source) Exiting due to uncaught Exception null > > It is that "specz.java:1198" that I'd like to locate and tie back to my NR source. > > > > _______________________________________________ > netrexx-pipelines mailing list > net...@li... > https://lists.sourceforge.net/lists/listinfo/netrexx-pipelines |
|
From: Jeff H. <Je...@Je...> - 2022-12-24 17:23:47
|
<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
</head>
<body>
<p>I am getting an error I can not trace. I'd like to try to
compile a pipe and keep the JAVA file, but am failing.</p>
<p>The Pipelines Guide and Reference, page 33, says <br>
</p>
<blockquote>
<p>The following options can be used on the pipc command, in
addition to the ones
specified in the previous chapter for the Pipes Runner: <br>
</p>
<p><font face="monospace">-gen</font> Generate the NetREXX source
file. The pipeline needs a name.</p>
<p><font face="monospace">-keep</font> Keep the Java source which
is generated from the NetREXX source. <br>
</p>
<p>Example:
<font face="monospace">pipe (testpipe -gen -keep)</font> <br>
</p>
<p>This will generate the NetREXX source as well as keep the java
source.</p>
</blockquote>
<p>First off, I am assuming, in the Example, that should be <font
face="monospace">pipc</font> and be followed by the name of the
njp file. But Windows PowerShell does not like any of the
combinations I try.</p>
<p>Any pointers in getting the JAVA file would be appreciated.</p>
<p>-------</p>
<p>The error I am getting trying to compile a test pipe -- with a
test version of the SPEC stage, is</p>
<div style="display:inline-block;white-space:pre;background-color:#012456;font-family:'Cascadia Code',monospace;font-size:10pt;padding:4px;"><span style="color:#D3D7CF;background-color:#012456;">java.lang.NullPointerException
at org.netrexx.njpipes.stages.specz.stageExit(specz.java:1198)
at org.netrexx.njpipes.pipes.compiler.compile(compiler.java:574)
at org.netrexx.njpipes.pipes.compiler.compile(compiler.java:166)
at org.netrexx.njpipes.pipes.pipe2nrx.run(pipe2nrx.java:47)
at org.netrexx.njpipes.pipes.ThreadPool.run(ThreadPool.java:76)
at java.lang.Thread.run(Unknown Source)
Exiting due to uncaught Exception null</span></div>
<p></p>
<p>It is that "specz.java:1198" that I'd like to locate and tie back
to my NR source.<br>
</p>
</body>
</html>
|
|
From: J L. T. <jlt...@ma...> - 2022-12-22 00:47:31
|
Wonderful! Thank you so very much! Leslie On 2022-12-21 12:38:58 Jeff Hennick wrote: > I have just uploaded an updated SPEC stage. It will be included in the > next release of NetRexx. (The source is now available from SourceForge, > but since it uses advanced features it can only be compiled by rebuilding > the full NetRexx jars. Doable.) > > This fixes a known bug with X2C/C2X. (I had used the NetRexx builtin, but > it works with a singe character, not a string like classic Rexx.) Also > fixed, I hope, was the proper way to handle multiple input streams. The big > addition is FIELDS, now in addition to characters and words. And > WORDSEPARATOR and FIELDSEPARATOR, and their abbreviations. As an extension > to the CMS separators, which are limited to a single character, NetRexx > Pipelines now also supports multi-character separators. These are > specified by a "QWord." This is a simple single word, which is optionally > surrounded by quote marks, single or double. Quotes are necessary if the > "word" includes a SPACE, or begins with a quote mark. These contrived > examples are from specs_tests_9.njp: > > -- specs wordseparator multichar multiseps? > literal @#one@#@#@#two@#@#three@#four | > specs wordseparator @# w2 1 | > cons > > The output is two. > And > -- specs multi fieldseparator multichar multiseps? > literal $@#$one$@#@#@#two@#@#$three$@#four | > specs fieldseparator @# f5 1 fieldseparator $ f5 nw | > cons > > With output of two three. > > > > > And, NetRexx Pipelines has the two newish stages for rearranging content: > PARSE and GREP / REGEX. -- |
|
From: Jeff H. <Je...@Je...> - 2022-12-21 18:55:29
|
<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
</head>
<body>
<p>I have just uploaded an updated <b>SPEC</b> stage. It will be
included in the next release of NetRexx. (The source is now
available from SourceForge, but since it uses advanced features it
can only be compiled by rebuilding the full NetRexx jars.
Doable.)<br>
</p>
<p>This fixes a known bug with X2C/C2X. (I had used the NetRexx
builtin, but it works with a singe character, not a string like
classic Rexx.) Also fixed, I hope, was the proper way to handle
multiple input streams.</p>
<p>The big addition is <b>FIELDS</b>, now in addition to characters
and words. And <b>WORDSEPARATOR</b> and <b>FIELDSEPARATOR</b>,
and their abbreviations. As an extension to the CMS separators,
which are limited to a single character, NetRexx Pipelines now
also supports multi-character separators. These are specified by
a "QWord." This is a simple single word, which is optionally
surrounded by quote marks, single or double. Quotes are necessary
if the "word" includes a SPACE, or begins with a quote mark.</p>
<p>These contrived examples are from <b>specs_tests_9.njp</b>: <br>
</p>
<font face="monospace">-- specs wordseparator multichar multiseps?<br>
literal @#one@#@#@#two@#@#three@#four |<br>
specs wordseparator @# w2 1 |<br>
cons</font><br>
<p>The output is <b>two</b>.</p>
<p>And</p>
<font face="monospace">-- specs multi fieldseparator multichar
multiseps?</font><br>
<font face="monospace"> literal $@#$one$@#@#@#two@#@#$three$@#four
|</font><br>
<font face="monospace"> specs fieldseparator @# f5 1
fieldseparator $ f5 nw |</font><br>
<font face="monospace"> cons <br>
</font>
<p>With output of <b>two three</b>.</p>
<p><br>
</p>
<p><img src="cid:par...@Je..." alt=""></p>
<p><img src="cid:par...@Je..." alt=""></p>
<p><img src="cid:par...@Je..." alt=""></p>
<p><img src="cid:par...@Je..." alt=""></p>
<p><img src="cid:par...@Je..." alt=""></p>
<p><br>
</p>
<p>And, NetRexx Pipelines has the two newish stages for rearranging
content: <b>PARSE</b> and <b>GREP / REGEX</b>.<br>
<font face="monospace"></font></p>
</body>
</html> |
|
From: Marc R. <re...@gm...> - 2022-10-22 13:50:18
|
Hi Jeff,
a NetRexx program returns the value specified by an 'exit' clause when run.
The translator returns any of the following :
EXIT_OK =int 0 -- Exit values
EXIT_WARN =int 1 -- Exit values
EXIT_ERROR =int 2 -- Exit values
I've reproduced your issue, this 'RC=16' comes from the run method in specj.
Above this method, there is the comment 'The run method of specs should never get called'
Not being familiar with pipelines, I don't know why this method is not be called, nor why it is called when having a local class generated without package statement.
To fix your issue and test your new specj stage, leave the package statement as is and rebuild the NetRexxC/F.jar ('ant' from netrexx-code directory) to have the specj class properly 'packaged' ;
$ pipe 'literal one two three | specj word -2 1 | console'
two
Marc
On 10/22/22 04:27, Jeff Hennick wrote:
> [cross post with ne...@gr... previously]
>
> I am working on the NetRexx Pipelines *spec* stage, fixing a bug and adding functionality (FIELDS, etc.).
>
> But I am running into a NetRexx roadblock.
>
> I have checked out the source for the working stage, *spec.nrx*, and for testing modified it in only two ways: changed both the file name and the class name to *specj*.
>
> This I can compile using the *pipc* utility. But when I use this stage in a pipe, I get an error message that the class was found, but with the wrong full name. So I commented out the *package* statement (as I have done for testing many stages) and recompiled.
>
> Now running the pipe, it immediately returns with *RC=16*. I have added *trace results*, and nothing is executed before the RC=16.
>
> I can find no list or reference of NetRexx Standard Return Codes. Is there one? Knowing the history of NetRexx, I looked for Rexx Standard Return Codes. That says 16 is *Label not found*. Is that its meaning in NetRexx? Is there anyway of learning what label might have not been found?
>
> For the record, specj,and spec, uses the less straightforward stageExit method, where the stage processes the options then on the fly builds a new, simpler, sub-stage to read the records. I have successfully divided it into the two halves: the new options are processed and the sub-stage is created and instead of being immediately run it is put out. I can compile that and successfully use it in a pipe. I just can't get it to all work together.
>
> In adding FIELDS, I have enhanced the SEPARATOR for fields and words from CMS's single character to a full string. This will be useful for the character pairs CRLF, or COMMA+SPACE for examples.
>
> Thank you for any help you can give on this.
>
> Jeff Hennick
>
>
> _._,_._,_
>
>
> _______________________________________________
> netrexx-pipelines mailing list
> net...@li...
> https://lists.sourceforge.net/lists/listinfo/netrexx-pipelines
|
|
From: Jeff H. <Je...@Je...> - 2022-10-22 02:45:32
|
<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
</head>
<body>
<p>[cross post with <a class="moz-txt-link-abbreviated" href="mailto:ne...@gr...">ne...@gr...</a> previously]</p>
<p>I am working on the NetRexx Pipelines <b>spec</b> stage, fixing
a bug and adding functionality (FIELDS, etc.).</p>
<p>But I am running into a NetRexx roadblock.</p>
<p>I have checked out the source for the working stage, <b>spec.nrx</b>,
and for testing modified it in only two ways: changed both the
file name and the class name to <b>specj</b>.</p>
<p>This I can compile using the <b>pipc</b> utility. But when I
use this stage in a pipe, I get an error message that the class
was found, but with the wrong full name. So I commented out the <b>package</b>
statement (as I have done for testing many stages) and recompiled.</p>
<p>Now running the pipe, it immediately returns with <b>RC=16</b>.
I have added <b>trace results</b>, and nothing is executed before
the RC=16.</p>
<p>I can find no list or reference of NetRexx Standard Return
Codes. Is there one? Knowing the history of NetRexx, I looked
for Rexx Standard Return Codes. That says 16 is <b>Label not
found</b>. Is that its meaning in NetRexx? Is there anyway of
learning what label might have not been found?</p>
<p>For the record, specj,and spec, uses the less straightforward
stageExit method, where the stage processes the options then on
the fly builds a new, simpler, sub-stage to read the records. I
have successfully divided it into the two halves: the new options
are processed and the sub-stage is created and instead of being
immediately run it is put out. I can compile that and
successfully use it in a pipe. I just can't get it to all work
together.</p>
<p>In adding FIELDS, I have enhanced the SEPARATOR for fields and
words from CMS's single character to a full string. This will be
useful for the character pairs CRLF, or COMMA+SPACE for examples.</p>
<p>Thank you for any help you can give on this.</p>
<p>Jeff Hennick</p>
<br>
<div width="1" style="color:white;clear:both">_._,_._,_</div>
</body>
</html>
|
