Re: [Bug-apl] indexing precedence

2017-02-16 Thread Jay Foad
This was a breaking change in APL2 (previously 1 2 3[2] was valid and returned 2) but I think it was the right decision for a language with stranding. It means that A[1] B[2] parses as (A[1])(B[2]) which seems sane. The alternative (ISO, Dyalog) is that A[1] B[2] parses as (A[1] B)[2], which is pr

Re: [Bug-apl] indexing precedence

2017-02-15 Thread Juergen Sauermann
Hi Xiao-Yong, I know, there are a few differences between APL2 and the ISO standard. In such cases APL2 rules because a standard which is not implemented is less useful than an implementation of it. /// Jürgen

Re: [Bug-apl] indexing precedence

2017-02-15 Thread enztec
not even 3[1] is good of even (3)[1]it has to be (,3)l1]it has to be an ARRAY so the index can work if this behavior is changed to do what apl2 has then this is no longer apl - remember apl goes right to left On Wed, 15 Feb 2017 14:44:18 -0600 Xiao-Yong Jin wrote: > ISO/IE

Re: [Bug-apl] indexing precedence

2017-02-15 Thread Xiao-Yong Jin
ISO/IEC 13751:2000(E) disagrees. On page 158, section 10.2.14 Indexed Reference, there is an example 1 2 3[2] 2 > On Feb 15, 2017, at 2:22 PM, Juergen Sauermann > wrote: > > Hi Xiao-Yong, > > IBM APL2. > > [] binds stronger than anything else. > > Thus 3[1 2] is evaluated first, giving

Re: [Bug-apl] indexing precedence

2017-02-15 Thread Juergen Sauermann
Hi Xiao-Yong, IBM APL2. [] binds stronger than anything else. Thus 3[1 2] is evaluated first, giving the RANK ERROR. /// Jürgen On 02/15/2017 09:14 PM, Xiao-Yong Jin wrote: Feature or bu

Re: [Bug-apl] indexing precedence

2017-02-15 Thread enztec
you are really trying to takeindex of just 3 with 3[2 1 3]hence the need for the ( ) On Wed, 15 Feb 2017 14:14:21 -0600 Xiao-Yong Jin wrote: > Feature or bug? > > 1 2 3[2 1 3] > RANK ERROR > 1 2 3[2 1 3] > ^^ > )sic > (1 2 3)[2 1 3] > 2 1 3 > >

[Bug-apl] indexing precedence

2017-02-15 Thread Xiao-Yong Jin
Feature or bug? 1 2 3[2 1 3] RANK ERROR 1 2 3[2 1 3] ^^ )sic (1 2 3)[2 1 3] 2 1 3