bug#29669: repl-print - requet for iprovement

bug#29669: repl-print - requet for iprovement

[David Pirotte]

[-- Attachment #1.1: Type: text/plain, Size: 287 bytes --]

Hello,

Attached a patch so repl-print becomes 'friendly' wrt large (huge) lists,
arrays, srfi-4 bytevectors ...

	I looked at the documentation but I could not spot anything/places where
	things would need to be changed wrt the above patch?

	Dito wrt tests?

Thanks,
David

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1.2: 0001-Updating-repl-print-to-use-truncated-print.patch --]
[-- Type: text/x-patch, Size: 1780 bytes --]

From d920d22efe3e77d23004122e21cec420c402f531 Mon Sep 17 00:00:00 2001
From: David Pirotte <david@altosw.be>
Date: Mon, 11 Dec 2017 21:28:24 -0200
Subject: [PATCH] Updating repl-print to use truncated-print

* module/system/repl/common.scm (repl-print): Use (truncated-print val),
  not (write val).  With this update, repl-print becomes 'friendly' wrt
  large (huge) lists, arrays, srfi-4 butevoectors ...
---
 module/system/repl/common.scm | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/module/system/repl/common.scm b/module/system/repl/common.scm
index 42d5c24ae..47799957d 100644
--- a/module/system/repl/common.scm
+++ b/module/system/repl/common.scm
@@ -1,6 +1,6 @@
 ;;; Repl common routines
 
-;; Copyright (C) 2001, 2008-2016  Free Software Foundation, Inc.
+;; Copyright (C) 2001, 2008-2017  Free Software Foundation, Inc.
 
 ;;; This library is free software; you can redistribute it and/or
 ;;; modify it under the terms of the GNU Lesser General Public
@@ -28,6 +28,7 @@
   #:autoload (language tree-il optimize) (optimize)
   #:use-module (ice-9 control)
   #:use-module (ice-9 history)
+  #:use-module (ice-9 pretty-print)
   #:export (<repl> make-repl repl-language repl-options
             repl-tm-stats repl-gc-stats repl-debug
             repl-welcome repl-prompt
@@ -223,7 +224,7 @@ See <http://www.gnu.org/licenses/lgpl.html>, for more details.")
           ;; should be printed with the generic printer, `write'. The
           ;; language-printer is something else: it prints expressions of
           ;; a given language, not the result of evaluation.
-          (write val)
+          (truncated-print val)
           (newline))))))
 
 (define (repl-option-ref repl key)
-- 
2.15.1


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

bug#29669: repl-print - requet for iprovement

[Ludovic Courtès]

Hello,

David Pirotte <david@altosw.be> skribis:

> From d920d22efe3e77d23004122e21cec420c402f531 Mon Sep 17 00:00:00 2001
> From: David Pirotte <david@altosw.be>
> Date: Mon, 11 Dec 2017 21:28:24 -0200
> Subject: [PATCH] Updating repl-print to use truncated-print
>
> * module/system/repl/common.scm (repl-print): Use (truncated-print val),
>   not (write val).  With this update, repl-print becomes 'friendly' wrt
>   large (huge) lists, arrays, srfi-4 butevoectors ...

Note that it’s already possible to do this:

--8<---------------cut here---------------start------------->8---
scheme@(guile-user)> ,use (ice-9 pretty-print)
scheme@(guile-user)> ,o print (lambda (repl obj) (truncated-print obj) (newline))
scheme@(guile-user)> (iota 500)
$20 = (0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 # …)
--8<---------------cut here---------------end--------------->8---

So the question becomes: should we change the default?

I have a slight preference for keeping the default as it is to avoid
surprises, but no strong opinion.

Andy, Mark, others, WDYT?

Ludo’.

bug#29669: repl-print - requet for iprovement

[Andy Wingo]

On Sun 01 Jul 2018 18:26, ludo@gnu.org (Ludovic Courtès) writes:

> Hello,
>
> David Pirotte <david@altosw.be> skribis:
>
>> From d920d22efe3e77d23004122e21cec420c402f531 Mon Sep 17 00:00:00 2001
>> From: David Pirotte <david@altosw.be>
>> Date: Mon, 11 Dec 2017 21:28:24 -0200
>> Subject: [PATCH] Updating repl-print to use truncated-print
>>
>> * module/system/repl/common.scm (repl-print): Use (truncated-print val),
>>   not (write val).  With this update, repl-print becomes 'friendly' wrt
>>   large (huge) lists, arrays, srfi-4 butevoectors ...
>
> Note that it’s already possible to do this:
>
> scheme@(guile-user)> ,use (ice-9 pretty-print)
> scheme@(guile-user)> ,o print (lambda (repl obj) (truncated-print obj) (newline))
> scheme@(guile-user)> (iota 500)
> $20 = (0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 # …)
>
> So the question becomes: should we change the default?
>
> I have a slight preference for keeping the default as it is to avoid
> surprises, but no strong opinion.
>
> Andy, Mark, others, WDYT?

Hoo, I don't know.  If we were to do this it should be controllable by
REPL options, I think; we'd need the ability to go back and forth.  But
if we have the option I think it could make sense for it to be on by
default, like what GDB does.  Thing is, truncated-print does its job
only OK, not great, so it's a hard sell to standardize on it.  You
probably do want multi-line prints sometimes...

Andy

bug#29669: repl-print - requet for iprovement

[David Pirotte]

[-- Attachment #1: Type: text/plain, Size: 1974 bytes --]

Hi Ludo,
Andy,

> > scheme@(guile-user)> ,use (ice-9 pretty-print)
> > scheme@(guile-user)> ,o print (lambda (repl obj) (truncated-print obj) (newline))
> > scheme@(guile-user)> (iota 500)
> > $20 = (0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
> > # …)
> >
> > So the question becomes: should we change the default?

Why not? then who would want multi-line prints would do the 'oppoite' of the above,
right?

	scheme@(guile-user)> ,use (ice-9 pretty-print)
	scheme@(guile-user)> ,o print (lambda (repl obj) (write obj) (newline))

Or, as I do in Guile-CV, implement a specific display method for 'their
objects' (im-display in guile-cv, and the manual says 'to be used with caution...)

> > I have a slight preference for keeping the default as it is to avoid
> > surprises, but no strong opinion.

What surprise(s)? :),
I mean, what sort of code or who's the one of us who would rely on the repl print
'layout' in any sort of way(s)?

> > Andy, Mark, others, WDYT?  

> Hoo, I don't know.  If we were to do this it should be controllable by
> REPL options, I think; we'd need the ability to go back and forth.  But
> if we have the option I think it could make sense for it to be on by
> default, like what GDB does. 

It seems to me it is a yes :)

	note that (just not to forgot), we need this to be done both for the repl
	_and_ exception printers as well ... it is as essential if not more then
	within the repl

> Thing is, truncated-print does its job  only OK, not great, so it's a hard sell to
> standardize on it.  You probably do want multi-line prints sometimes...

I'm actually using truncated-print all the time, since it is simply impossible
otherwise to use/develop guile-v, and I never spotted any 'problem' (not to say I
can guarantee there is none of course, but that did not happen in years).  Then we
could always improve when someone report a problem ...

Thanks,
David

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]