Hi Jürgen,

thank you for pointing to *DISCARD-INDENTATION.
*I even did not have it on my mind to look for such a parameter.

*SVN 1969 *  ⎕CR, ⎕FX, ∇  work as expected / configured. Thank You.

Best Regards
Hans-Peter

On 27/02/2026 19:03, Dr. Jürgen Sauermann via Bugs and suggestions for GNU APL wrote:
Hi Hans-Peter,

I double-checked the code and found the following.

1. There is a user preference named *DISCARD-INDENTATION* that
    controls whether the ∇-editor and *⎕FX* shall keep or discard leading
    and trailing blanks in the lines of defined functions. Setting this
    preference to Yes should give the IBM APL2 behavior. Not setting
    it (the default) or setting it to Yes should give the behavior that
    you expected.

2. Both the ∇-editor and *⎕FX* obey that preference.

3. However, monadic *⎕CR* did not obey that preference.
   That is, your leading blanks were discarded by *⎕CR* (and
   not by *⎕FX*).

4. I have fixed 3. in *SVN 1969*:
*
      ⎕FX 'FOO' '  bar' '    bar'
FOO
      8 ⎕CR ⎕CR 'FOO'
┌→──────┐
↓FOO    │
│  bar  │
│    bar│
└───────┘
*
Best Regards,
Jürgen



On 2/27/26 15:21, Hans-Peter Sorge wrote:
Hi Jürgen,

the ∇-editor does indeed preserve leading space for ««« ... »»»

[0]   x
[1]   t←«««
[2]       text1
[3]   longtext2
[4]             what ever
[5]   »»»
[6]   t

      x
text1
longtext2
  what ever

Could  ⍞FX use the same logic?
It would simplify more elaborate and structured help output.

Best Regards
Hans-Peter

On 27/02/2026 13:19, Dr. Jürgen Sauermann wrote:
Hi Hans-Peter,

thanks. What you see is the standard APL2 behavior
(see attached screenshot). ⎕FX and the ∇-editor
discard leading whitespace.

Best Regards,
Jürgen


On 2/26/26 20:48, Hans-Peter Sorge wrote:
Hi,

⎕FX does not maintain the multi-line string fidelity,
leading spaces get removed.:

    m←'x' 'h← «««' 'ddddd' '     lpppp'  '»»»'
     ⊃m
x
h← «««
ddddd
    lpppp
»»»
     ⎕FX m

     ⊃h
ddddd
lpppp

 ∇x[⎕]
   ∇
[0]   x
[1]   h← «««
[2]   ddddd
[3]   lpppp
[4]   »»»
   ∇


Best Regards
Hans-Peter





  • multi-line str... Hans-Peter Sorge
    • Re: multi... Dr . Jürgen Sauermann via Bugs and suggestions for GNU APL
      • Re: m... Hans-Peter Sorge
        • R... Dr . Jürgen Sauermann via Bugs and suggestions for GNU APL
          • ... Hans-Peter Sorge

Reply via email to