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: Jeff H. <Je...@Je...> - 2025-03-02 21:59:34
|
<!DOCTYPE html>
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
<p>Lelie,</p>
<p>Good catch. Actually, we can take arrayr and arrayw out of the
documentation as they are "helper classes" for array. (Array
looks to see if it first in the pipe or not and then uses the
right helper.) When I first saw the list of files, I included
them in the documentation due to ignorance; I recently learned
more. I will take them out, probably in time for this next
edition. I am still too ignorant about arraya for action right
now.</p>
<p>Jeff<br>
</p>
<div class="moz-cite-prefix">On 3/2/2025 11:18 AM, J Leslie Turriff
via netrexx-pipelines wrote:<br>
</div>
<blockquote type="cite"
cite="mid:202...@ma...">
<pre wrap="" class="moz-quote-pre"> In the Pipelines Guide and Reference list of built-in stages, we see
array | Read or Write an Array
| Pipes for NetRexx
arraya | Read or Write an Array
| Pipes for NetRexx
arrayr | Read or Write an Array
| Pipes for NetRexx
arrayw | Read or Write an Array
| Pipes for NetRexx
Should these not say
array | Read From an Array
| Pipes for NetRexx
arraya | Append To or Create an Array
| Pipes for NetRexx
arrayr | Read From an Array
| Pipes for NetRexx
arrayw | Replace or Create an Array
| Pipes for NetRexx
as file, filea, filer, filew do?
Since arrays are fixed-length, does the pipeline define the array if
necessary?
Leslie
--
Platform: Linux
Distribution: openSUSE Leap 15.6 - x86_64
NetRexx 4.06-GA build 152-20240304-0612
openjdk version "21.0.3" 2024-04-16
OpenJDK Runtime Environment (build 21.0.3+9-suse-lp155.51.3-x8664)
OpenJDK 64-Bit Server VM (build 21.0.3+9-suse-lp155.51.3-x8664, mixed mode,
sharing)
_______________________________________________
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...> - 2025-03-02 16:19:14
|
In the Pipelines Guide and Reference list of built-in stages, we see
array | Read or Write an Array
| Pipes for NetRexx
arraya | Read or Write an Array
| Pipes for NetRexx
arrayr | Read or Write an Array
| Pipes for NetRexx
arrayw | Read or Write an Array
| Pipes for NetRexx
Should these not say
array | Read From an Array
| Pipes for NetRexx
arraya | Append To or Create an Array
| Pipes for NetRexx
arrayr | Read From an Array
| Pipes for NetRexx
arrayw | Replace or Create an Array
| Pipes for NetRexx
as file, filea, filer, filew do?
Since arrays are fixed-length, does the pipeline define the array if
necessary?
Leslie
--
Platform: Linux
Distribution: openSUSE Leap 15.6 - x86_64
NetRexx 4.06-GA build 152-20240304-0612
openjdk version "21.0.3" 2024-04-16
OpenJDK Runtime Environment (build 21.0.3+9-suse-lp155.51.3-x8664)
OpenJDK 64-Bit Server VM (build 21.0.3+9-suse-lp155.51.3-x8664, mixed mode,
sharing)
|
|
From: Jeff H. <Je...@Je...> - 2025-02-23 18:29:54
|
<!DOCTYPE html>
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
<p>Fixed.</p>
<p>I have pushed a fix that converts the stage name to lowercase
prior to processing.</p>
<p>Now upper, lower, or mixed case stage names compile nicely.</p>
<p>Jeff<br>
</p>
<div class="moz-cite-prefix">On 2/22/2025 1:59 PM, Jeff Hennick
wrote:<br>
</div>
<blockquote type="cite"
cite="mid:6e0...@Je...">
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<p>I am having a problem with Pipelines, discovered while
preparing for the Symposium. It is using the PARSE stage: it
works great if the stage name in the pipe is all lowercase,
"parse," but not if it is uppercase, "PARSE." Other stage names
work fine upper or lower. I don't know if there are other
problem names or not.<br>
</p>
<p>Here, m3 is lowercase and m4 upper. </p>
<p>Here I compile and run them. Below is the listings of the
files. <br>
</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\nr5\examples\pipes> </span><span
style="color:#FCE94F;background-color:#012456;">pipc </span><span
style="color:#D3D7CF;background-color:#012456;">m3</span>
<span style="color:#D3D7CF;background-color:#012456;">pipe (m3 ) literal blue brown red | parse /_1 . / /Part z1, has _1 units/ | cons</span>
<span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr5\examples\pipes> </span><span
style="color:#FCE94F;background-color:#012456;">java </span><span
style="color:#D3D7CF;background-color:#012456;">m3</span>
<span style="color:#D3D7CF;background-color:#012456;">Part z1, has blue units</span>
<span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr5\examples\pipes> </span><span
style="color:#FCE94F;background-color:#012456;">pipc </span><span
style="color:#D3D7CF;background-color:#012456;">m4</span>
<span style="color:#D3D7CF;background-color:#012456;">pipe (m4 ) literal blue brown red | PARSE /_1 . / /Part z1, has _1 units/ | cons</span>
<span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr5\examples\pipes> </span><span
style="color:#FCE94F;background-color:#012456;">java </span><span
style="color:#D3D7CF;background-color:#012456;">m4</span>
<span style="color:#D3D7CF;background-color:#012456;">RC=16</span>
<span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr5\examples\pipes> </span><span
style="color:#FCE94F;background-color:#012456;">cat </span><span
style="color:#D3D7CF;background-color:#012456;">m3.njp</span>
<span style="color:#D3D7CF;background-color:#012456;">pipe (m3)</span>
<span style="color:#D3D7CF;background-color:#012456;"> literal blue brown red |</span>
<span style="color:#D3D7CF;background-color:#012456;"> parse /_1 . / /Part z1, has _1 units/ |</span>
<span style="color:#D3D7CF;background-color:#012456;"> cons</span>
<span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr5\examples\pipes> </span><span
style="color:#FCE94F;background-color:#012456;">cat </span><span
style="color:#D3D7CF;background-color:#012456;">m4.njp</span>
<span style="color:#D3D7CF;background-color:#012456;">pipe (m4)</span>
<span style="color:#D3D7CF;background-color:#012456;"> literal blue brown red |</span>
<span style="color:#D3D7CF;background-color:#012456;"> PARSE /_1 . / /Part z1, has _1 units/ |</span>
<span style="color:#D3D7CF;background-color:#012456;"> cons</span>
<span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr5\examples\pipes></span></div>
<br>
<fieldset class="moz-mime-attachment-header"></fieldset>
<br>
<fieldset class="moz-mime-attachment-header"></fieldset>
<pre wrap="" class="moz-quote-pre">_______________________________________________
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...> - 2025-02-23 14:27:41
|
<!DOCTYPE html>
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
<p>Yes. It works fine. Except for the "echo" after compiling.</p>
<p>Jeff<br>
</p>
<div class="moz-cite-prefix">On 2/23/2025 1:04 AM, J Leslie Turriff
via netrexx-pipelines wrote:<br>
</div>
<blockquote type="cite"
cite="mid:202...@ma...">
<pre wrap="" class="moz-quote-pre"> Um... you're mixing parentheses and braces in your prefix literal ... stage.
Leslie
On 2025-02-22 15:00:34 Jeff Hennick wrote:
</pre>
<blockquote type="cite">
<pre wrap="" class="moz-quote-pre">Compiling a pipe that uses the {var} structure to include a NetRexx
variable works, but the "echo" is not correct. See the case below, where it
should be arg(string {'avar'}) it is missing two punctuation marks:
arg(string 'avar'}
PS C:\Users\Jeff\Documents\nr5\examples\pipes> pipc testpipe.nrx
pipe (testpipe_apipe stall 1000) stem F | sort | prefix literal arg(string
'avar'} | console | stem T PS C:\Users\Jeff\Documents\nr5\examples\pipes>
</pre>
</blockquote>
<pre wrap="" class="moz-quote-pre">--
Platform: Linux
Distribution: openSUSE Leap 15.6 - x86_64
NetRexx 4.06-GA build 152-20240304-0612
openjdk version "21.0.3" 2024-04-16
OpenJDK Runtime Environment (build 21.0.3+9-suse-lp155.51.3-x8664)
OpenJDK 64-Bit Server VM (build 21.0.3+9-suse-lp155.51.3-x8664, mixed mode,
sharing)
_______________________________________________
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...> - 2025-02-23 06:04:34
|
Um... you're mixing parentheses and braces in your prefix literal ... stage.
Leslie
On 2025-02-22 15:00:34 Jeff Hennick wrote:
> Compiling a pipe that uses the {var} structure to include a NetRexx
> variable works, but the "echo" is not correct. See the case below, where it
> should be arg(string {'avar'}) it is missing two punctuation marks:
> arg(string 'avar'}
>
> PS C:\Users\Jeff\Documents\nr5\examples\pipes> pipc testpipe.nrx
> pipe (testpipe_apipe stall 1000) stem F | sort | prefix literal arg(string
> 'avar'} | console | stem T PS C:\Users\Jeff\Documents\nr5\examples\pipes>
--
Platform: Linux
Distribution: openSUSE Leap 15.6 - x86_64
NetRexx 4.06-GA build 152-20240304-0612
openjdk version "21.0.3" 2024-04-16
OpenJDK Runtime Environment (build 21.0.3+9-suse-lp155.51.3-x8664)
OpenJDK 64-Bit Server VM (build 21.0.3+9-suse-lp155.51.3-x8664, mixed mode,
sharing)
|
|
From: René J. <rvj...@xs...> - 2025-02-22 23:58:29
|
this is just a hunch: both SELECT and PARSE are NetRexx keywords. René. > On 22 Feb 2025, at 21:45, Jeff Hennick <Je...@je...> wrote: > > I have found another with a similar problem: SELECT. > > The only difference in these two is the capitalization of SELECT: > > PS C:\Users\Jeff\Documents\nr5\examples\pipes> pipc m4 > pipe (m4 ) literal blue white green| literal green blue white | literal brown red white | literal blue brown red | SELECT multiple /parse rec . C .; r = 'blue red white'.wordpos(c); return r/ | cons > PS C:\Users\Jeff\Documents\nr5\examples\pipes> java m4 > Error: the RUN method should never be called. > RC=16 > PS C:\Users\Jeff\Documents\nr5\examples\pipes> pipc m4 > pipe (m4 ) literal blue white green| literal green blue white | literal brown red white | literal blue brown red | select multiple /parse rec . C .; r = 'blue red white'.wordpos(c); return r/ | cons > PS C:\Users\Jeff\Documents\nr5\examples\pipes> java m4 > blue brown red > PS C:\Users\Jeff\Documents\nr5\examples\pipes> > > On 2/22/2025 1:59 PM, Jeff Hennick wrote: >> I am having a problem with Pipelines, discovered while preparing for the Symposium. It is using the PARSE stage: it works great if the stage name in the pipe is all lowercase, "parse," but not if it is uppercase, "PARSE." Other stage names work fine upper or lower. I don't know if there are other problem names or not. >> >> Here, m3 is lowercase and m4 upper. >> >> Here I compile and run them. Below is the listings of the files. >> >> PS C:\Users\Jeff\Documents\nr5\examples\pipes> pipc m3 >> pipe (m3 ) literal blue brown red | parse /_1 . / /Part z1, has _1 units/ | cons >> PS C:\Users\Jeff\Documents\nr5\examples\pipes> java m3 >> Part z1, has blue units >> PS C:\Users\Jeff\Documents\nr5\examples\pipes> pipc m4 >> pipe (m4 ) literal blue brown red | PARSE /_1 . / /Part z1, has _1 units/ | cons >> PS C:\Users\Jeff\Documents\nr5\examples\pipes> java m4 >> RC=16 >> PS C:\Users\Jeff\Documents\nr5\examples\pipes> cat m3.njp >> pipe (m3) >> >> literal blue brown red | >> parse /_1 . / /Part z1, has _1 units/ | >> cons >> >> PS C:\Users\Jeff\Documents\nr5\examples\pipes> cat m4.njp >> pipe (m4) >> >> literal blue brown red | >> PARSE /_1 . / /Part z1, has _1 units/ | >> cons >> >> PS C:\Users\Jeff\Documents\nr5\examples\pipes> >> >> >> _______________________________________________ >> netrexx-pipelines mailing list >> net...@li... <mailto: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...> - 2025-02-22 21:00:51
|
<!DOCTYPE html>
<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
</head>
<body>
<p>Compiling a pipe that uses the <b>{var}</b> structure to include
a NetRexx variable works, but the "echo" is not correct.</p>
<p>See the case below, where it should be <b>arg(string {'avar'})</b>
it is missing two punctuation marks: <b>arg(string 'avar'}</b><br>
</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\nr5\examples\pipes> </span><span
style="color:#FCE94F;background-color:#012456;">pipc </span><span
style="color:#D3D7CF;background-color:#012456;">testpipe.nrx</span>
<span style="color:#D3D7CF;background-color:#012456;">pipe (testpipe_apipe stall 1000) stem F | sort | prefix literal arg(string 'av</span><span
style="color:#D3D7CF;background-color:#012456;">ar'} | console | stem T</span>
<span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr5\examples\pipes></span></div>
<p></p>
<p><br>
</p>
</body>
</html>
|
|
From: Jeff H. <Je...@Je...> - 2025-02-22 20:45:18
|
<!DOCTYPE html>
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
<p>I have found another with a similar problem: SELECT.</p>
<p>The only difference in these two is the capitalization of SELECT:</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\nr5\examples\pipes> </span><span
style="color:#FCE94F;background-color:#012456;">pipc </span><span
style="color:#D3D7CF;background-color:#012456;">m4</span>
<span style="color:#D3D7CF;background-color:#012456;">pipe (m4 ) literal blue white green| literal green blue white | literal brown re</span><span
style="color:#D3D7CF;background-color:#012456;">d white | literal blue brown red | SELECT multiple /parse rec . C .; r = 'blue</span><span
style="color:#D3D7CF;background-color:#012456;"> red white'.wordpos(c); return r/ | cons</span>
<span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr5\examples\pipes> </span><span
style="color:#FCE94F;background-color:#012456;">java </span><span
style="color:#D3D7CF;background-color:#012456;">m4</span>
<span style="color:#D3D7CF;background-color:#012456;">Error: the RUN method should never be called.</span>
<span style="color:#D3D7CF;background-color:#012456;">RC=16</span>
<span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr5\examples\pipes> </span><span
style="color:#FCE94F;background-color:#012456;">pipc </span><span
style="color:#D3D7CF;background-color:#012456;">m4</span>
<span style="color:#D3D7CF;background-color:#012456;">pipe (m4 ) literal blue white green| literal green blue white | literal brown re</span><span
style="color:#D3D7CF;background-color:#012456;">d white | literal blue brown red | select multiple /parse rec . C .; r = 'blue</span><span
style="color:#D3D7CF;background-color:#012456;"> red white'.wordpos(c); return r/ | cons</span>
<span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr5\examples\pipes> </span><span
style="color:#FCE94F;background-color:#012456;">java </span><span
style="color:#D3D7CF;background-color:#012456;">m4</span>
<span style="color:#D3D7CF;background-color:#012456;">blue brown red</span>
<span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr5\examples\pipes></span></div>
<p></p>
<div class="moz-cite-prefix">On 2/22/2025 1:59 PM, Jeff Hennick
wrote:<br>
</div>
<blockquote type="cite"
cite="mid:6e0...@Je...">
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<p>I am having a problem with Pipelines, discovered while
preparing for the Symposium. It is using the PARSE stage: it
works great if the stage name in the pipe is all lowercase,
"parse," but not if it is uppercase, "PARSE." Other stage names
work fine upper or lower. I don't know if there are other
problem names or not.<br>
</p>
<p>Here, m3 is lowercase and m4 upper. </p>
<p>Here I compile and run them. Below is the listings of the
files. <br>
</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\nr5\examples\pipes> </span><span
style="color:#FCE94F;background-color:#012456;">pipc </span><span
style="color:#D3D7CF;background-color:#012456;">m3</span>
<span style="color:#D3D7CF;background-color:#012456;">pipe (m3 ) literal blue brown red | parse /_1 . / /Part z1, has _1 units/ | cons</span>
<span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr5\examples\pipes> </span><span
style="color:#FCE94F;background-color:#012456;">java </span><span
style="color:#D3D7CF;background-color:#012456;">m3</span>
<span style="color:#D3D7CF;background-color:#012456;">Part z1, has blue units</span>
<span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr5\examples\pipes> </span><span
style="color:#FCE94F;background-color:#012456;">pipc </span><span
style="color:#D3D7CF;background-color:#012456;">m4</span>
<span style="color:#D3D7CF;background-color:#012456;">pipe (m4 ) literal blue brown red | PARSE /_1 . / /Part z1, has _1 units/ | cons</span>
<span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr5\examples\pipes> </span><span
style="color:#FCE94F;background-color:#012456;">java </span><span
style="color:#D3D7CF;background-color:#012456;">m4</span>
<span style="color:#D3D7CF;background-color:#012456;">RC=16</span>
<span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr5\examples\pipes> </span><span
style="color:#FCE94F;background-color:#012456;">cat </span><span
style="color:#D3D7CF;background-color:#012456;">m3.njp</span>
<span style="color:#D3D7CF;background-color:#012456;">pipe (m3)</span>
<span style="color:#D3D7CF;background-color:#012456;"> literal blue brown red |</span>
<span style="color:#D3D7CF;background-color:#012456;"> parse /_1 . / /Part z1, has _1 units/ |</span>
<span style="color:#D3D7CF;background-color:#012456;"> cons</span>
<span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr5\examples\pipes> </span><span
style="color:#FCE94F;background-color:#012456;">cat </span><span
style="color:#D3D7CF;background-color:#012456;">m4.njp</span>
<span style="color:#D3D7CF;background-color:#012456;">pipe (m4)</span>
<span style="color:#D3D7CF;background-color:#012456;"> literal blue brown red |</span>
<span style="color:#D3D7CF;background-color:#012456;"> PARSE /_1 . / /Part z1, has _1 units/ |</span>
<span style="color:#D3D7CF;background-color:#012456;"> cons</span>
<span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr5\examples\pipes></span></div>
<br>
<fieldset class="moz-mime-attachment-header"></fieldset>
<br>
<fieldset class="moz-mime-attachment-header"></fieldset>
<pre wrap="" class="moz-quote-pre">_______________________________________________
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...> - 2025-02-22 19:15:23
|
<!DOCTYPE html>
<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
</head>
<body>
<p>I am having a problem with Pipelines, discovered while preparing
for the Symposium. It is using the PARSE stage: it works great if
the stage name in the pipe is all lowercase, "parse," but not if
it is uppercase, "PARSE." Other stage names work fine upper or
lower. I don't know if there are other problem names or not.<br>
</p>
<p>Here, m3 is lowercase and m4 upper. </p>
<p>Here I compile and run them. Below is the listings of the files.
<br>
</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\nr5\examples\pipes> </span><span
style="color:#FCE94F;background-color:#012456;">pipc </span><span
style="color:#D3D7CF;background-color:#012456;">m3</span>
<span style="color:#D3D7CF;background-color:#012456;">pipe (m3 ) literal blue brown red | parse /_1 . / /Part z1, has _1 units/ | cons</span>
<span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr5\examples\pipes> </span><span
style="color:#FCE94F;background-color:#012456;">java </span><span
style="color:#D3D7CF;background-color:#012456;">m3</span>
<span style="color:#D3D7CF;background-color:#012456;">Part z1, has blue units</span>
<span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr5\examples\pipes> </span><span
style="color:#FCE94F;background-color:#012456;">pipc </span><span
style="color:#D3D7CF;background-color:#012456;">m4</span>
<span style="color:#D3D7CF;background-color:#012456;">pipe (m4 ) literal blue brown red | PARSE /_1 . / /Part z1, has _1 units/ | cons</span>
<span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr5\examples\pipes> </span><span
style="color:#FCE94F;background-color:#012456;">java </span><span
style="color:#D3D7CF;background-color:#012456;">m4</span>
<span style="color:#D3D7CF;background-color:#012456;">RC=16</span>
<span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr5\examples\pipes> </span><span
style="color:#FCE94F;background-color:#012456;">cat </span><span
style="color:#D3D7CF;background-color:#012456;">m3.njp</span>
<span style="color:#D3D7CF;background-color:#012456;">pipe (m3)</span>
<span style="color:#D3D7CF;background-color:#012456;"> literal blue brown red |</span>
<span style="color:#D3D7CF;background-color:#012456;"> parse /_1 . / /Part z1, has _1 units/ |</span>
<span style="color:#D3D7CF;background-color:#012456;"> cons</span>
<span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr5\examples\pipes> </span><span
style="color:#FCE94F;background-color:#012456;">cat </span><span
style="color:#D3D7CF;background-color:#012456;">m4.njp</span>
<span style="color:#D3D7CF;background-color:#012456;">pipe (m4)</span>
<span style="color:#D3D7CF;background-color:#012456;"> literal blue brown red |</span>
<span style="color:#D3D7CF;background-color:#012456;"> PARSE /_1 . / /Part z1, has _1 units/ |</span>
<span style="color:#D3D7CF;background-color:#012456;"> cons</span>
<span style="color:#D3D7CF;background-color:#012456;">PS C:\Users\Jeff\Documents\nr5\examples\pipes></span></div>
<p></p>
</body>
</html>
|
|
From: J L. T. <jlt...@ma...> - 2025-02-14 23:47:20
|
Hm. I had to clone the source code directory to figure out what's going on to cause this RC=13 condition. As one can see below, from working directory ~/data/Documents/eBooks/SF, the absolute path /run/media/leslie/HONORVERSE/index.htm is APPENDED to the working directory, so isn't found. % @02:18:44 leslie@pinto % wd=~/data/Documents/eBooks/SF % $ pipe '(debug 64) < /run/media/leslie/HONORVERSE/index.htm|take 5|console' % diskr_1 Error - File "/home/leslie/data/Documents/eBooks/SF//run/media/leslie/HONORVERSE/index.htm" not found rc=13 % RC=13 % rc=0 I was able to find it by changing the debug option from 63 to 64, which displays the name of the stage and outputs the error message instead of the list of stage statuses. So I have two issues to report: 1) Stages < (and probably diskr, filer) do not properly handle absolute paths 2) The debug options are not documented, so one has to download the source code to determine how to use them. Leslie -- Platform: Linux Distribution: openSUSE Leap 15.6 - x86_64 NetRexx 4.06-GA build 152-20240304-0612 openjdk version "21.0.3" 2024-04-16 OpenJDK Runtime Environment (build 21.0.3+9-suse-lp155.51.3-x8664) OpenJDK 64-Bit Server VM (build 21.0.3+9-suse-lp155.51.3-x8664, mixed mode, sharing) |
|
From: J L. T. <jlt...@ma...> - 2025-02-14 23:31:38
|
Inside a pipeline, the ? is the default separator for multi-stage streams. It can be overridden with the end option. Leslie On 2023-02-26 11:24:21 Jeff Hennick wrote: > In building a new HELP stage, and wanting to handle all the stage names and > their synonyms, I have run into a problem with an argument of just the > question mark: "?", which is a synonym for HELP in CMS. Here are the first > few lines traced. First with the character > which is processed all > right. But with the character ? , it never gets through. > > PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> pipe "ahelp >" > --- ahelp.nrx [Thread-1,njPipes] > 25 *=* method run() > 26 *=* data = Rexx "" > > >v> data "" > > 28 *=* parse arg() stage_to_help > > >v> stage_to_help ">" > > PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> pipe "ahelp ?" > --- ahelp.nrx [Thread-1,njPipes] > 25 *=* method run() > 26 *=* data = Rexx "" > > >v> data "" > > 28 *=* parse arg() stage_to_help > > >v> stage_to_help "" > > Some exploration with raw NetRexx shows the Shell does not like passing >, > but all is right with?: PS > C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> type toast.nrx /* > This wishes you the best of health. */ > class toast > method main(argwords=String[]) static > arg=Rexx(argwords) > say 'Cheers!' arg > > PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> java toast > Cheers! > PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> java toast xyz > Cheers! xyz > PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> java toast > > At line:1 char:13 > + java toast > > + ~ > Missing file specification after redirection operator. > + CategoryInfo : ParserError: (:) [], > ParentContainsErrorRecordEx ception > + FullyQualifiedErrorId : MissingFileSpecification > > PS C:\Users\Jeff\Documents\nr\netrexx-code\examples\pipes> java toast ? > Cheers! ? > > So, NetRexx alone is not "eating" the "?" parameter, it must be Pipelines > that is. Is this a feature of Pipelines that I have missed, or a bug? > Anyone's help would be appreciated. > Jeff Hennick |
|
From: J L. T. <jlt...@ma...> - 2025-02-14 22:58:05
|
(Oops! Sent this to the wrong list!) I have a CD containing various open-source books in HTML format that I wish to reformat to make them more readable. I am able to read these books using various Linux tools, and even the jEdit (Java-based) text editor, but a NetRexx pipe fails with rc=13 (though it returns rc=0 to the shell). Here is the output from several commands that read from the CD and from a regular disk drive. I have set the debugging option as suggested in the Language Reference, but the output doesn't seem very helpful, at least to me. % @16:55:25 leslie@pinto % wd=~/data/Documents/eBooks/SF % $ head -n5 /run/media/leslie/HONORVERSE/index.htm % <html> % % <head> % <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1"> % <title>Honor Harrington Index Page</title> % rc=0 % @16:55:48 leslie@pinto % wd=~/data/Documents/eBooks/SF % $ pipe '(debug 63) < /run/media/leslie/HONORVERSE/index.htm|take 5|console' % Starting pipe p10a417d p10a417d@2e4538fc % Starting diskr_1 p10a417d@2e4538fc % Starting take_2 p10a417d@2e4538fc % Autocommitting take_2 p10a417d@2e4538fc to -1 % Starting console_3 p10a417d@2e4538fc % Preparing to stop diskr_1 p10a417d@2e4538fc 13 3 % Ending Pipe p10a417d p10a417d@2e4538fc 96 13 % Autocommitting diskr_1 p10a417d@2e4538fc to -1 % Autocommitting console_3 p10a417d@2e4538fc to -1 % Committed console_3 p10a417d@2e4538fc to -1 % RC=13 % rc=0 % @16:56:01 leslie@pinto % wd=~/data/Documents/eBooks/SF % $ pipe '< Authors/DavidWeber/index.html|take 5|console' % <!doctype html> % <html> % <head> % <meta charset=utf-8> % <link rel=stylesheet href=/home/leslie/Documents/Books/simple.css> % rc=0 The debug option is described only as a bitmap, but the meanings of the bits are not explained: % debug % The debug option specifies a bitmask for debugging the execu- % tion of a pipe; (debug 63), for example, generates a rather com- % plete debugging trail. Anyway, can you explain why this pipeline is failing? Leslie -- Platform: Linux Distribution: openSUSE Leap 15.6 - x86_64 NetRexx 4.06-GA build 152-20240304-0612 openjdk version "21.0.3" 2024-04-16 OpenJDK Runtime Environment (build 21.0.3+9-suse-lp155.51.3-x8664) OpenJDK 64-Bit Server VM (build 21.0.3+9-suse-lp155.51.3-x8664, mixed mode, sharing) |
|
From: Marc R. <re...@gm...> - 2023-08-05 10:59:45
|
Hi Leslie,
once again, when you redirect a process (as the command stage does), that process is not connected to a tty anymore, as pipe 'command tty | console' says.
That is why the pipe 'command stty --all | console' has a correct empty output.
When using a specific device with -F, you are interrogating the characteristics of that device, not necessarily the terminal or console you are running in.
Linux has two types of TTYs, physical and pseudo terminals. TTY comes from teletype, from the old days where people interfaced with computers through physical wires.
Don't confuse 'terminal' with the terminal application.
The stty command does not only allow to collect the characteristics of a TTY, but also allows to change them.
Consider this:
$ ps axo user,tty,pid,comm | grep bash
xxxxxx pts/0 7644 bash
xxxxxx tty3 30797 bash
xxxxxx pts/1 34414 bash
I have three bash programs running, one connected to a 'real' TTY /dev/tty3, started from the console prompt Ctrl-Alt-F3.
Two other bash programs are started from the gnome terminal program, and are connected to /dev/pts/0 and /dev/pts/1.
$ pipe 'command ps | cons'
PID TTY TIME CMD
7644 pts/0 00:00:00 bash
42258 pts/0 00:00:05 java
42289 pts/0 00:00:00 ps
The pipeline stages are running as children of process 7644, connected to /dev/pts/0.
$ pipe 'command stty --all -F /dev/pts/1 | cons'
speed 38400 baud; rows 30; columns 96; line = 0;
intr = ^C; quit = ^\; erase = ^?; kill = ^U; eof = ^D; eol = <undef>;
eol2 = <undef>; swtch = <undef>; start = ^Q; stop = ^S; susp = ^Z; rprnt = ^R;
werase = ^W; lnext = <undef>; discard = <undef>; min = 1; time = 0;
-parenb -parodd -cmspar cs8 -hupcl -cstopb cread -clocal -crtscts
-ignbrk -brkint -ignpar -parmrk -inpck -istrip -inlcr -igncr -icrnl ixon -ixoff
-iuclc -ixany -imaxbel iutf8
opost -olcuc -ocrnl onlcr -onocr -onlret -ofill -ofdel nl0 cr0 tab0 bs0 vt0 ff0
isig -icanon iexten -echo echoe echok -echonl -noflsh -xcase -tostop -echoprt
echoctl echoke -flusho -extproc
Collected tty specs of sibling bash process 34414, connected to /dev/pts/1
$ pipe 'command stty cols 20 -F /dev/pts/1 | cons'
Changed number of columns of /dev/pts/1 (bash process 34414)
$ pipe 'command stty --all -F /dev/pts/1 | cons'
speed 38400 baud; rows 30; columns 20; line = 0;
intr = ^C; quit = ^\; erase = ^?; kill = ^U; eof = ^D; eol = <undef>;
eol2 = <undef>; swtch = <undef>; start = ^Q; stop = ^S; susp = ^Z; rprnt = ^R;
werase = ^W; lnext = <undef>; discard = <undef>; min = 1; time = 0;
-parenb -parodd -cmspar cs8 -hupcl -cstopb cread -clocal -crtscts
-ignbrk -brkint -ignpar -parmrk -inpck -istrip -inlcr -igncr -icrnl ixon -ixoff
-iuclc -ixany -imaxbel iutf8
opost -olcuc -ocrnl onlcr -onocr -onlret -ofill -ofdel nl0 cr0 tab0 bs0 vt0 ff0
isig -icanon iexten -echo echoe echok -echonl -noflsh -xcase -tostop -echoprt
echoctl echoke -flusho -extproc
Indeed, number of columns changed.
$ pipe 'command stty --all -F /dev/pts/0 | cons'
speed 38400 baud; rows 30; columns 120; line = 0;
intr = ^C; quit = ^\; erase = ^?; kill = ^U; eof = ^D; eol = <undef>;
eol2 = <undef>; swtch = <undef>; start = ^Q; stop = ^S; susp = ^Z; rprnt = ^R;
werase = ^W; lnext = ^V; discard = ^O; min = 1; time = 0;
-parenb -parodd -cmspar cs8 -hupcl -cstopb cread -clocal -crtscts
-ignbrk -brkint -ignpar -parmrk -inpck -istrip -inlcr -igncr icrnl ixon -ixoff
-iuclc -ixany -imaxbel iutf8
opost -olcuc -ocrnl onlcr -onocr -onlret -ofill -ofdel nl0 cr0 tab0 bs0 vt0 ff0
isig icanon iexten echo echoe echok -echonl -noflsh -xcase -tostop -echoprt
echoctl echoke -flusho -extproc
Number of columns of /dev/pts/0 (i.e where parent bash process is connected to) is unchanged.
If you need to know the number of rows of the terminal where the pipelines script is running, you need to know which TTY the parent process is connected to.
And it could be that that parent process does not have a TTY, for example when it's started as a daemon, in which case it will produce an empty stty --all output..
More reading about the old and newer days :
https://itsfoss.com/what-is-tty-in-linux/
https://www.linusakesson.net/programming/tty/
Marc
|
|
From: J L. T. <jlt...@ma...> - 2023-08-05 03:23:13
|
On 2023-08-04 06:14:27 Rony G. Flatscher wrote: > On 04.08.2023 06:30, J Leslie Turriff wrote: > > On 2023-08-02 06:45:16 Rony G. Flatscher wrote: > >> On 31.07.2023 16:27, Marc Remes wrote: > >>> pipe 'command stty --all|console also runs fine.. Its output is empty > >>> because it is not connected to a real/pseudo terminal, as echo dog |tty > >>> reports. > >>> > >>> Indeed, we're not doing bash redirection, it is now java that is doing > >>> the redirection. The command stage starts the stty process by r.exec > >>> and connects to its stdin. For the stty process there's no difference > >>> whether the redirection is done by bash or java, in both cases its > >>> stdin is not a terminal. And thus stty has no effect. > >> > >> According to the JavaDocs for Process > >> <https://docs.oracle.com/javase/8/docs/api/java/lang/Process.html> "... > >> By default, the created subprocess does not have its own terminal or > >> console. ..." so this may be an explanation that stty would not find it > >> and what Marc tries to point out. Not sure about ProcessBuilder > >> <https://docs.oracle.com/javase/8/docs/api/java/lang/ProcessBuilder.html > >>> can be used instead and making a difference? > >> > >> Maybe doing something like "stty -F `tty` --all" would work and if not > > > > supplying explicitly the device path, if known? > > You're a genius! It took several tries, and one has to know of a valid > > tty to plug into the command, but yes, that's the key. > > > > I tried these in the jEdit console: > > | ~> stty -F 'tty' --all > > > > as you suggested, but stty seems to want to set tty attributes when it > > sees -F; so I tried > > > > | ~> stty --all -F 'tty' > > > > but it said "stty: tty: No such file or directory" so I ran the w command > > to find a "real" tty, and it seems like /dev/pts/## (where ## is any of > > the ttys defined to the system) works, and the output from --all is not > > written to that device, but to the console where stty was invoked! so we > > see something like the attached output. The tty must belong to the > > account invoking stty or we get "Permission denied". > > > > Thank you! Thank you! > > You are very welcome, glad that I could help a little bit. > > One more remark: have you tried the back-ticks ` around "tty"? > > E.g. in your example (that did not work) you write: > > stty --all -F 'tty' > > If you write it this way, would it work for you? > > echo `tty` > stty --all -F `tty` > > ---rony I had not tried that before, but I just did and it fails when issued from the jEdit console. See the attached snapshot. I'm fairly happy with what I have that is working; now I just have to get my nrx pipeline program that uses it to work. It seems that mixing the address and the pipe commands in a simple pipeline program (stty-rows.njp, also attached) makes the pipe compiler choke; so using nrx pipelines to simplify what my bash version does might require an even longer NetRexx program. :-( Leslie |
|
From: Rony G. F. <Ron...@wu...> - 2023-08-04 11:14:44
|
On 04.08.2023 06:30, J Leslie Turriff wrote: > > On 2023-08-02 06:45:16 Rony G. Flatscher wrote: >> On 31.07.2023 16:27, Marc Remes wrote: >>> pipe 'command stty --all|console also runs fine.. Its output is empty >>> because it is not connected to a real/pseudo terminal, as echo dog |tty >>> reports. >>> >>> Indeed, we're not doing bash redirection, it is now java that is doing >>> the redirection. The command stage starts the stty process by r.exec and >>> connects to its stdin. For the stty process there's no difference whether >>> the redirection is done by bash or java, in both cases its stdin is not a >>> terminal. And thus stty has no effect. >> According to the JavaDocs for Process >> <https://docs.oracle.com/javase/8/docs/api/java/lang/Process.html> "... By >> default, the created subprocess does not have its own terminal or console. >> ..." so this may be an explanation that stty would not find it and what >> Marc tries to point out. Not sure about ProcessBuilder >> <https://docs.oracle.com/javase/8/docs/api/java/lang/ProcessBuilder.html> >> can be used instead and making a difference? >> >> Maybe doing something like "stty -F `tty` --all" would work and if not > supplying explicitly the device path, if known? > You're a genius! It took several tries, and one has to know of a valid tty > to plug into the command, but yes, that's the key. > > I tried these in the jEdit console: > | ~> stty -F 'tty' --all > as you suggested, but stty seems to want to set tty attributes when it > sees -F; so I tried > | ~> stty --all -F 'tty' > but it said "stty: tty: No such file or directory" so I ran the w command to > find a "real" tty, and it seems like /dev/pts/## (where ## is any of the ttys > defined to the system) works, and the output from --all is not written to > that device, but to the console where stty was invoked! so we see something > like the attached output. The tty must belong to the account invoking stty > or we get "Permission denied". > > Thank you! Thank you! You are very welcome, glad that I could help a little bit. One more remark: have you tried the back-ticks ` around "tty"? E.g. in your example (that did not work) you write: stty --all -F 'tty' If you write it this way, would it work for you? echo `tty` stty --all -F `tty` ---rony |
|
From: J L. T. <jlt...@ma...> - 2023-08-04 04:47:08
|
Well. A suggestion by Rony Flatscher found a solution to this issue. It seems that stty, unless told otherwise, does not use the most usual method of writing to the terminal, but it can be told to do so, using the --file=<device> flag, where <device> can be any "real" tty owned by the issuer of the stty command; so one can use the w command to see what ttys are available: | $ w | grep leslie | Cyllarus pts/15 :0 Mon02 3days 0.00s 0.00s su - leslie | leslie pts/33 :1 Mon12 4.00s 5:34 0.00s grep --color=auto leslie | leslie pts/34 :1 Mon12 28:53 6:09 0.02s su - | leslie pts/36 :1 Mon12 3days 0.03s 0.03s /bin/bash | leslie pts/35 :1 Mon12 2days 0.09s 0.09s /bin/bash | leslie pts/37 :1 Mon12 3days 0.03s 0.03s /bin/bash | leslie pts/38 :1 Mon12 3days 0.03s 0.03s /bin/bash | leslie pts/39 :1 Mon12 3days 0.03s 0.03s /bin/bash | leslie pts/40 :1 Mon12 2days 0.07s 0.02s su - | leslie pts/41 :1 Mon12 2days 0.07s 0.02s su - | leslie pts/42 :1 Mon12 2days 2.35s 0.01s /usr/bin/links file:///usr/share/doc/fish/index.html | leslie pts/43 :1 Mon12 3days 0.03s 0.03s /bin/bash | leslie pts/44 :1 Mon12 3days 0.00s 2:33 kded [tdeinit] --new-startup | leslie pts/45 :1 Tue16 2days 2.13s 2.13s /usr/bin/fish | developm pts/52 :2 Tue22 2days 0.00s 0.00s su - leslie | rc=0 | | $ stty --file=/dev/pts/45 --all | speed 38400 baud; rows 54; columns 135; line = 0; | intr = ^C; quit = <undef>; erase = ^?; kill = ^U; eof = ^D; eol = <undef>; eol2 = <undef>; swtch = <undef>; start = ^Q; | stop = ^S; susp = <undef>; rprnt = ^R; werase = ^W; lnext = ^V; discard = ^O; min = 1; time = 0; | -parenb -parodd -cmspar cs8 -hupcl -cstopb cread -clocal -crtscts | -ignbrk -brkint -ignpar -parmrk -inpck -istrip -inlcr -igncr -icrnl -ixon -ixoff -iuclc -ixany -imaxbel iutf8 | opost -olcuc -ocrnl onlcr -onocr -onlret -ofill -ofdel nl0 cr0 tab0 bs0 vt0 ff0 | isig -icanon -iexten -echo echoe echok -echonl -noflsh -xcase -tostop -echoprt echoctl echoke -flusho -extproc | | rc=0 Interestingly, any tty device owned by the user can be specified; stty writes its output to the tty where the command is invoked, not the one specified by the --file= option. I used the console feature of the jEdit text editor, which is implemented in Java, to verify that this works, and that the following pipeline, more compact than the bash script I used to use, works as well: | $ pipe 'command stty --all --file=/dev/pts/33|take first 1|specs word 5 next|strip trailing /;/|console' | 54 | rc=0 Leslie On 2023-07-31 09:27:09 Marc Remes wrote: > Leslie, > > pipe 'command stty --all|console also runs fine.. Its output is empty > because it is not connected to a real/pseudo terminal, as echo dog |tty > reports. > > Indeed, we're not doing bash redirection, it is now java that is doing the > redirection. The command stage starts the stty process by r.exec and > connects to its stdin. For the stty process there's no difference whether > the redirection is done by bash or java, in both cases its stdin is not a > terminal. And thus stty has no effect. > > Marc > > On 7/31/23 16:00, J Leslie Turriff wrote: > > On 2023-07-31 06:20:50 Marc Remes wrote: > >> On 7/31/23 13:03, Rony G. Flatscher wrote: > >>> According to "https://linux.die.net/man/1/stty" "stty --all" would > >>> print all settings (to stdout I assume). > >> > >> True : > >> > >> $ stty --all > >> speed 38400 baud; rows 30; columns 99; line = 0; > >> intr = ^C; quit = ^\; erase = ^?; kill = ^U; eof = ^D; eol = <undef>; > >> eol2 = <undef>; swtch = <undef>; start = ^Q; stop = ^S; susp = ^Z; rprnt > >> = ^R; werase = ^W; lnext = ^V; discard = ^O; min = 1; time = 0; > >> -parenb -parodd -cmspar cs8 -hupcl -cstopb cread -clocal -crtscts > >> -ignbrk -brkint -ignpar -parmrk -inpck -istrip -inlcr -igncr icrnl ixon > >> -ixoff -iuclc -ixany -imaxbel iutf8 > >> opost -olcuc -ocrnl onlcr -onocr -onlret -ofill -ofdel nl0 cr0 tab0 bs0 > >> vt0 ff0 isig icanon iexten echo echoe echok -echonl -noflsh -xcase > >> -tostop -echoprt echoctl echoke -flusho -extproc > >> $ echo dog | stty > >> stty: 'standard input': Inappropriate ioctl for device > >> > >> The first command is attached to a terminal (either a 'real' terminal, > >> but more likely a pseudo-terminal), and shows its settings. The second > >> command's stdin is the stdout of the echo command, and says so.. > >> > >> Also check the tty command: > >> $ tty > >> /dev/pts/6 > >> $ echo dog | tty > >> not a tty > >> > >> Marc > > > > But: > > | @08:49:08 ─▶leslie@pinto◀─ > > | wd=~/bin/NetRexx/Pipelines > > | $ pipe 'command ls -l|console' > > | total 224 > > | -rw-r--r-- 1 leslie users 3741 Jun 29 01:44 condense.class > > | -rw-r--r-- 1 leslie users 1298 Jun 29 00:07 GenSymLinks.class > > | -rw-r--r-- 1 leslie users 4688 Jun 28 23:53 > > | GenSymLinks_gensymlinks.class > > | > > | -rw-r--r-- 1 leslie users 2403 Jun 19 15:50 treeOutput.txt > > | rc=0 > > > > works just fine, whereas > > > > | @08:49:20 ─▶leslie@pinto◀─ > > | wd=~/bin/NetRexx/Pipelines > > | $ pipe 'command stty --all|console' > > | rc=0 > > > > produces no output. What is the difference between the stage "command ls > > -l" (which works) and "command stty --all" (which does not)? We're not > > doing bash redirection here; this is inside the NetRexx pipeline. > > > > Leslie > > _______________________________________________ > netrexx-pipelines mailing list > net...@li... > https://lists.sourceforge.net/lists/listinfo/netrexx-pipelines |
|
From: J L. T. <jlt...@ma...> - 2023-08-04 04:30:51
|
On 2023-08-02 06:45:16 Rony G. Flatscher wrote: > On 31.07.2023 16:27, Marc Remes wrote: > > pipe 'command stty --all|console also runs fine.. Its output is empty > > because it is not connected to a real/pseudo terminal, as echo dog |tty > > reports. > > > > Indeed, we're not doing bash redirection, it is now java that is doing > > the redirection. The command stage starts the stty process by r.exec and > > connects to its stdin. For the stty process there's no difference whether > > the redirection is done by bash or java, in both cases its stdin is not a > > terminal. And thus stty has no effect. > > According to the JavaDocs for Process > <https://docs.oracle.com/javase/8/docs/api/java/lang/Process.html> "... By > default, the created subprocess does not have its own terminal or console. > ..." so this may be an explanation that stty would not find it and what > Marc tries to point out. Not sure about ProcessBuilder > <https://docs.oracle.com/javase/8/docs/api/java/lang/ProcessBuilder.html> > can be used instead and making a difference? > > Maybe doing something like "stty -F `tty` --all" would work and if not supplying explicitly the device path, if known? You're a genius! It took several tries, and one has to know of a valid tty to plug into the command, but yes, that's the key. I tried these in the jEdit console: | ~> stty -F 'tty' --all as you suggested, but stty seems to want to set tty attributes when it sees -F; so I tried | ~> stty --all -F 'tty' but it said "stty: tty: No such file or directory" so I ran the w command to find a "real" tty, and it seems like /dev/pts/## (where ## is any of the ttys defined to the system) works, and the output from --all is not written to that device, but to the console where stty was invoked! so we see something like the attached output. The tty must belong to the account invoking stty or we get "Permission denied". Thank you! Thank you! Leslie > > --- > > In ooRexx it works as expected, i.e. > > cmd="stty --all" > cmd > > -- or redirecting > arr=.array~new > address system cmd with output using (arr) > say 'arr~items:' arr~items > say arr > > ---rony > > > On 7/31/23 16:00, J Leslie Turriff wrote: > >> On 2023-07-31 06:20:50 Marc Remes wrote: > >>> On 7/31/23 13:03, Rony G. Flatscher wrote: > >>>> According to "https://linux.die.net/man/1/stty" "stty --all" would > >>>> print all settings (to stdout I assume). > >>> > >>> True : > >>> > >>> $ stty --all > >>> speed 38400 baud; rows 30; columns 99; line = 0; > >>> intr = ^C; quit = ^\; erase = ^?; kill = ^U; eof = ^D; eol = <undef>; > >>> eol2 = <undef>; swtch = <undef>; start = ^Q; stop = ^S; susp = ^Z; > >>> rprnt = ^R; werase = ^W; lnext = ^V; discard = ^O; min = 1; time = 0; > >>> -parenb -parodd -cmspar cs8 -hupcl -cstopb cread -clocal -crtscts > >>> -ignbrk -brkint -ignpar -parmrk -inpck -istrip -inlcr -igncr icrnl ixon > >>> -ixoff -iuclc -ixany -imaxbel iutf8 > >>> opost -olcuc -ocrnl onlcr -onocr -onlret -ofill -ofdel nl0 cr0 tab0 bs0 > >>> vt0 ff0 isig icanon iexten echo echoe echok -echonl -noflsh -xcase > >>> -tostop -echoprt echoctl echoke -flusho -extproc > >>> $ echo dog | stty > >>> stty: 'standard input': Inappropriate ioctl for device > >>> > >>> The first command is attached to a terminal (either a 'real' terminal, > >>> but more likely a pseudo-terminal), and shows its settings. The second > >>> command's stdin is the stdout of the echo command, and says so.. > >>> > >>> Also check the tty command: > >>> $ tty > >>> /dev/pts/6 > >>> $ echo dog | tty > >>> not a tty > >>> > >>> Marc > >> > >> But: > >> | @08:49:08 leslie@pinto > >> | wd=~/bin/NetRexx/Pipelines > >> | $ pipe 'command ls -l|console' > >> | total 224 > >> | -rw-r--r-- 1 leslie users 3741 Jun 29 01:44 condense.class > >> | -rw-r--r-- 1 leslie users 1298 Jun 29 00:07 GenSymLinks.class > >> | -rw-r--r-- 1 leslie users 4688 Jun 28 23:53 > >> | GenSymLinks_gensymlinks.class : > >> | -rw-r--r-- 1 leslie users 2403 Jun 19 15:50 treeOutput.txt > >> | rc=0 > >> > >> works just fine, whereas > >> > >> | @08:49:20 ─▶leslie@pinto◀─ > >> | wd=~/bin/NetRexx/Pipelines > >> | $ pipe 'command stty --all|console' > >> | rc=0 > >> > >> produces no output. What is the difference between the stage "command > >> ls -l" (which works) and "command stty --all" (which does not)? We're > >> not doing bash redirection here; this is inside the NetRexx pipeline. > >> > >> Leslie |
|
From: Rony G. F. <Ron...@wu...> - 2023-08-02 11:45:32
|
On 31.07.2023 16:27, Marc Remes wrote: > pipe 'command stty --all|console also runs fine.. Its output is empty because it is not connected > to a real/pseudo terminal, as echo dog |tty reports. > > Indeed, we're not doing bash redirection, it is now java that is doing the redirection. > The command stage starts the stty process by r.exec and connects to its stdin. > For the stty process there's no difference whether the redirection is done by bash or java, in > both cases its stdin is not a terminal. And thus stty has no effect. According to the JavaDocs for Process <https://docs.oracle.com/javase/8/docs/api/java/lang/Process.html> "... By default, the created subprocess does not have its own terminal or console. ..." so this may be an explanation that stty would not find it and what Marc tries to point out. Not sure about ProcessBuilder <https://docs.oracle.com/javase/8/docs/api/java/lang/ProcessBuilder.html> can be used instead and making a difference? Maybe doing something like "stty -F `tty` --all" would work and if not supplying explicitly the device path, if known? --- In ooRexx it works as expected, i.e. cmd="stty --all" cmd -- or redirecting arr=.array~new address system cmd with output using (arr) say 'arr~items:' arr~items say arr ---rony > > On 7/31/23 16:00, J Leslie Turriff wrote: >> On 2023-07-31 06:20:50 Marc Remes wrote: >>> On 7/31/23 13:03, Rony G. Flatscher wrote: >>>> According to "https://linux.die.net/man/1/stty" "stty --all" would print >>>> all settings (to stdout I assume). >>> >>> True : >>> >>> $ stty --all >>> speed 38400 baud; rows 30; columns 99; line = 0; >>> intr = ^C; quit = ^\; erase = ^?; kill = ^U; eof = ^D; eol = <undef>; eol2 >>> = <undef>; swtch = <undef>; start = ^Q; stop = ^S; susp = ^Z; rprnt = ^R; >>> werase = ^W; lnext = ^V; discard = ^O; min = 1; time = 0; >>> -parenb -parodd -cmspar cs8 -hupcl -cstopb cread -clocal -crtscts >>> -ignbrk -brkint -ignpar -parmrk -inpck -istrip -inlcr -igncr icrnl ixon >>> -ixoff -iuclc -ixany -imaxbel iutf8 >>> opost -olcuc -ocrnl onlcr -onocr -onlret -ofill -ofdel nl0 cr0 tab0 bs0 vt0 >>> ff0 isig icanon iexten echo echoe echok -echonl -noflsh -xcase -tostop >>> -echoprt echoctl echoke -flusho -extproc >>> $ echo dog | stty >>> stty: 'standard input': Inappropriate ioctl for device >>> >>> The first command is attached to a terminal (either a 'real' terminal, but >>> more likely a pseudo-terminal), and shows its settings. The second >>> command's stdin is the stdout of the echo command, and says so.. >>> >>> Also check the tty command: >>> $ tty >>> /dev/pts/6 >>> $ echo dog | tty >>> not a tty >>> >>> Marc >>> >> But: >> >> | @08:49:08 ─▶leslie@pinto◀─ >> | wd=~/bin/NetRexx/Pipelines >> | $ pipe 'command ls -l|console' >> | total 224 >> | -rw-r--r-- 1 leslie users 3741 Jun 29 01:44 condense.class >> | -rw-r--r-- 1 leslie users 1298 Jun 29 00:07 GenSymLinks.class >> | -rw-r--r-- 1 leslie users 4688 Jun 28 23:53 GenSymLinks_gensymlinks.class >> | : >> | -rw-r--r-- 1 leslie users 2403 Jun 19 15:50 treeOutput.txt >> | rc=0 >> >> works just fine, whereas >> >> | @08:49:20 ─▶leslie@pinto◀─ >> | wd=~/bin/NetRexx/Pipelines >> | $ pipe 'command stty --all|console' >> | rc=0 >> >> produces no output. What is the difference between the stage "command ls -l" >> (which works) and "command stty --all" (which does not)? We're not doing >> bash redirection here; this is inside the NetRexx pipeline. >> >> Leslie |
|
From: Marc R. <re...@gm...> - 2023-07-31 14:27:20
|
Leslie, pipe 'command stty --all|console also runs fine.. Its output is empty because it is not connected to a real/pseudo terminal, as echo dog |tty reports. Indeed, we're not doing bash redirection, it is now java that is doing the redirection. The command stage starts the stty process by r.exec and connects to its stdin. For the stty process there's no difference whether the redirection is done by bash or java, in both cases its stdin is not a terminal. And thus stty has no effect. Marc On 7/31/23 16:00, J Leslie Turriff wrote: > On 2023-07-31 06:20:50 Marc Remes wrote: >> On 7/31/23 13:03, Rony G. Flatscher wrote: >>> According to "https://linux.die.net/man/1/stty" "stty --all" would print >>> all settings (to stdout I assume). >> >> True : >> >> $ stty --all >> speed 38400 baud; rows 30; columns 99; line = 0; >> intr = ^C; quit = ^\; erase = ^?; kill = ^U; eof = ^D; eol = <undef>; eol2 >> = <undef>; swtch = <undef>; start = ^Q; stop = ^S; susp = ^Z; rprnt = ^R; >> werase = ^W; lnext = ^V; discard = ^O; min = 1; time = 0; >> -parenb -parodd -cmspar cs8 -hupcl -cstopb cread -clocal -crtscts >> -ignbrk -brkint -ignpar -parmrk -inpck -istrip -inlcr -igncr icrnl ixon >> -ixoff -iuclc -ixany -imaxbel iutf8 >> opost -olcuc -ocrnl onlcr -onocr -onlret -ofill -ofdel nl0 cr0 tab0 bs0 vt0 >> ff0 isig icanon iexten echo echoe echok -echonl -noflsh -xcase -tostop >> -echoprt echoctl echoke -flusho -extproc >> $ echo dog | stty >> stty: 'standard input': Inappropriate ioctl for device >> >> The first command is attached to a terminal (either a 'real' terminal, but >> more likely a pseudo-terminal), and shows its settings. The second >> command's stdin is the stdout of the echo command, and says so.. >> >> Also check the tty command: >> $ tty >> /dev/pts/6 >> $ echo dog | tty >> not a tty >> >> Marc >> > But: > > | @08:49:08 ─▶leslie@pinto◀─ > | wd=~/bin/NetRexx/Pipelines > | $ pipe 'command ls -l|console' > | total 224 > | -rw-r--r-- 1 leslie users 3741 Jun 29 01:44 condense.class > | -rw-r--r-- 1 leslie users 1298 Jun 29 00:07 GenSymLinks.class > | -rw-r--r-- 1 leslie users 4688 Jun 28 23:53 GenSymLinks_gensymlinks.class > | : > | -rw-r--r-- 1 leslie users 2403 Jun 19 15:50 treeOutput.txt > | rc=0 > > works just fine, whereas > > | @08:49:20 ─▶leslie@pinto◀─ > | wd=~/bin/NetRexx/Pipelines > | $ pipe 'command stty --all|console' > | rc=0 > > produces no output. What is the difference between the stage "command ls -l" > (which works) and "command stty --all" (which does not)? We're not doing > bash redirection here; this is inside the NetRexx pipeline. > > Leslie |
|
From: J L. T. <jlt...@ma...> - 2023-07-31 14:00:16
|
On 2023-07-31 06:20:50 Marc Remes wrote: > On 7/31/23 13:03, Rony G. Flatscher wrote: > > According to "https://linux.die.net/man/1/stty" "stty --all" would print > > all settings (to stdout I assume). > > True : > > $ stty --all > speed 38400 baud; rows 30; columns 99; line = 0; > intr = ^C; quit = ^\; erase = ^?; kill = ^U; eof = ^D; eol = <undef>; eol2 > = <undef>; swtch = <undef>; start = ^Q; stop = ^S; susp = ^Z; rprnt = ^R; > werase = ^W; lnext = ^V; discard = ^O; min = 1; time = 0; > -parenb -parodd -cmspar cs8 -hupcl -cstopb cread -clocal -crtscts > -ignbrk -brkint -ignpar -parmrk -inpck -istrip -inlcr -igncr icrnl ixon > -ixoff -iuclc -ixany -imaxbel iutf8 > opost -olcuc -ocrnl onlcr -onocr -onlret -ofill -ofdel nl0 cr0 tab0 bs0 vt0 > ff0 isig icanon iexten echo echoe echok -echonl -noflsh -xcase -tostop > -echoprt echoctl echoke -flusho -extproc > $ echo dog | stty > stty: 'standard input': Inappropriate ioctl for device > > The first command is attached to a terminal (either a 'real' terminal, but > more likely a pseudo-terminal), and shows its settings. The second > command's stdin is the stdout of the echo command, and says so.. > > Also check the tty command: > $ tty > /dev/pts/6 > $ echo dog | tty > not a tty > > Marc > But: | @08:49:08 ─▶leslie@pinto◀─ | wd=~/bin/NetRexx/Pipelines | $ pipe 'command ls -l|console' | total 224 | -rw-r--r-- 1 leslie users 3741 Jun 29 01:44 condense.class | -rw-r--r-- 1 leslie users 1298 Jun 29 00:07 GenSymLinks.class | -rw-r--r-- 1 leslie users 4688 Jun 28 23:53 GenSymLinks_gensymlinks.class | : | -rw-r--r-- 1 leslie users 2403 Jun 19 15:50 treeOutput.txt | rc=0 works just fine, whereas | @08:49:20 ─▶leslie@pinto◀─ | wd=~/bin/NetRexx/Pipelines | $ pipe 'command stty --all|console' | rc=0 produces no output. What is the difference between the stage "command ls -l" (which works) and "command stty --all" (which does not)? We're not doing bash redirection here; this is inside the NetRexx pipeline. Leslie -- Platform: Linux Distribution: openSUSE Leap 15.4 (x86_64) java version "19.0.1" 2022-10-18 Java(TM) SE Runtime Environment (build 19.0.1+10-21) Java HotSpot(TM) 64-Bit Server VM (build 19.0.1+10-21, mixed mode, sharing) NetRexx portable processor 4.05-beta build 62-20230422-0022 |
|
From: Marc R. <re...@gm...> - 2023-07-31 11:21:01
|
On 7/31/23 13:03, Rony G. Flatscher wrote: > According to "https://linux.die.net/man/1/stty" "stty --all" would print all settings (to stdout I assume). True : $ stty --all speed 38400 baud; rows 30; columns 99; line = 0; intr = ^C; quit = ^\; erase = ^?; kill = ^U; eof = ^D; eol = <undef>; eol2 = <undef>; swtch = <undef>; start = ^Q; stop = ^S; susp = ^Z; rprnt = ^R; werase = ^W; lnext = ^V; discard = ^O; min = 1; time = 0; -parenb -parodd -cmspar cs8 -hupcl -cstopb cread -clocal -crtscts -ignbrk -brkint -ignpar -parmrk -inpck -istrip -inlcr -igncr icrnl ixon -ixoff -iuclc -ixany -imaxbel iutf8 opost -olcuc -ocrnl onlcr -onocr -onlret -ofill -ofdel nl0 cr0 tab0 bs0 vt0 ff0 isig icanon iexten echo echoe echok -echonl -noflsh -xcase -tostop -echoprt echoctl echoke -flusho -extproc $ echo dog | stty stty: 'standard input': Inappropriate ioctl for device The first command is attached to a terminal (either a 'real' terminal, but more likely a pseudo-terminal), and shows its settings. The second command's stdin is the stdout of the echo command, and says so.. Also check the tty command: $ tty /dev/pts/6 $ echo dog | tty not a tty Marc |
|
From: Rony G. F. <Ron...@wu...> - 2023-07-31 11:03:16
|
On 31.07.2023 12:57, Marc Remes wrote: > a 'redirected pipe' is a program that has its stdin (or stdout/stderr) redirected. > > For example 'echo dog | cat', here cat is reading its stdin from the stdout of the echo command, > The cat process is not connected to a tty, hence stty has no effect. > > The pipelines 'command' stage does exactly that: a java process that runs a program with its > stdin/stdout redirected. According to "https://linux.die.net/man/1/stty" "stty --all" would print all settings (to stdout I assume). ---rony > > On 7/31/23 02:05, J Leslie Turriff wrote >> Sorry, I don't understand. When I run stty --all in bash (no redirection) I >> get seven lines of output on stdout; when I run it as an operand of the >> command stage I get no lines. What is the redirected pipe you're referring >> to? >> >> Leslie >> >> On 2023-07-30 12:56:29 Marc Remes wrote: >>> Hi Leslie, >>> >>> because a redirected pipe is not a tty, stty has no effect (and no output). >>> >>> Marc >>> >>> On 7/29/23 09:39, J Leslie Turriff wrote: >>>> Marc, >>>> stty is not the same as tty; tty is a device, but stty is a tool to >>>> query or change a tty's settings. >>>> >>>> On 2023-07-28 06:35:18 Marc Remes wrote: >>>>> Hi Leslie, >>>>> >>>>> A redirected pipe is not a tty. That's why the output is empty. >>>>> try pipe 'command tty | console' >>>>> There are methods to connect a pseudo tty, but I don't think we need to >>>>> adapt njpipes to support that. >>>>> >>>>> ref: https://gamlor.info/posts-output/2021-05-29-tty-in-java/en/ >>>>> >>>>> Marc |
|
From: Marc R. <re...@gm...> - 2023-07-31 10:57:33
|
Hi Leslie, a 'redirected pipe' is a program that has its stdin (or stdout/stderr) redirected. For example 'echo dog | cat', here cat is reading its stdin from the stdout of the echo command, The cat process is not connected to a tty, hence stty has no effect. The pipelines 'command' stage does exactly that: a java process that runs a program with its stdin/stdout redirected. Marc On 7/31/23 02:05, J Leslie Turriff wrote > Sorry, I don't understand. When I run stty --all in bash (no redirection) I > get seven lines of output on stdout; when I run it as an operand of the > command stage I get no lines. What is the redirected pipe you're referring > to? > > Leslie > > On 2023-07-30 12:56:29 Marc Remes wrote: >> Hi Leslie, >> >> because a redirected pipe is not a tty, stty has no effect (and no output). >> >> Marc >> >> On 7/29/23 09:39, J Leslie Turriff wrote: >>> Marc, >>> stty is not the same as tty; tty is a device, but stty is a tool to >>> query or change a tty's settings. >>> >>> On 2023-07-28 06:35:18 Marc Remes wrote: >>>> Hi Leslie, >>>> >>>> A redirected pipe is not a tty. That's why the output is empty. >>>> try pipe 'command tty | console' >>>> There are methods to connect a pseudo tty, but I don't think we need to >>>> adapt njpipes to support that. >>>> >>>> ref: https://gamlor.info/posts-output/2021-05-29-tty-in-java/en/ >>>> >>>> Marc >>>> >>>> >>>> _______________________________________________ >>>> 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 >> >> _______________________________________________ >> 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: J L. T. <jlt...@ma...> - 2023-07-31 00:05:27
|
Sorry, I don't understand. When I run stty --all in bash (no redirection) I get seven lines of output on stdout; when I run it as an operand of the command stage I get no lines. What is the redirected pipe you're referring to? Leslie On 2023-07-30 12:56:29 Marc Remes wrote: > Hi Leslie, > > because a redirected pipe is not a tty, stty has no effect (and no output). > > Marc > > On 7/29/23 09:39, J Leslie Turriff wrote: > > Marc, > > stty is not the same as tty; tty is a device, but stty is a tool to > > query or change a tty's settings. > > > > On 2023-07-28 06:35:18 Marc Remes wrote: > >> Hi Leslie, > >> > >> A redirected pipe is not a tty. That's why the output is empty. > >> try pipe 'command tty | console' > >> There are methods to connect a pseudo tty, but I don't think we need to > >> adapt njpipes to support that. > >> > >> ref: https://gamlor.info/posts-output/2021-05-29-tty-in-java/en/ > >> > >> Marc > >> > >> > >> _______________________________________________ > >> 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 > > _______________________________________________ > netrexx-pipelines mailing list > net...@li... > https://lists.sourceforge.net/lists/listinfo/netrexx-pipelines |
|
From: Marc R. <re...@gm...> - 2023-07-30 17:56:45
|
Hi Leslie, because a redirected pipe is not a tty, stty has no effect (and no output). Marc On 7/29/23 09:39, J Leslie Turriff wrote: > Marc, > stty is not the same as tty; tty is a device, but stty is a tool to query or > change a tty's settings. > > On 2023-07-28 06:35:18 Marc Remes wrote: >> Hi Leslie, >> >> A redirected pipe is not a tty. That's why the output is empty. >> try pipe 'command tty | console' >> There are methods to connect a pseudo tty, but I don't think we need to >> adapt njpipes to support that. >> >> ref: https://gamlor.info/posts-output/2021-05-29-tty-in-java/en/ >> >> Marc >> >> >> _______________________________________________ >> 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 |
