> My first clue that something is amiss is in your third line of code when the
> return skips "AA" and starts "AB, AC, AD....". That suggests to me that the
> two step assign/printf call is playing havoc with the $ anonymous variable
Missed that about the missing AA - does the same thing with a named var,
though:
raku -e 'for <AA NN> -> $alpha { for (1..14) { (state $sv = $alpha)++;
printf("d: %s\n", $sv ) } }'
d: AB
d: AC
So
raku -e 'for <AA NN> -> $alpha { for (1..14) { say (state $sv = $alpha)++;
printf("d: %s\n", $sv ) } }'
AA
d: AB
AB
d: AC
Ah, the increment happens the initial assignment.
$sv = "AA";
$sv++;
print $sv; # AB
but the
say (state $sv = $alpha)++
says the result of the assignment, then the increment. My confusion was more
about my inability to use "$" anywhere else.
raku -e 'for <AA NN> -> $alpha { for (1..14) { say (state $ = $alpha)++;
printf("d: %s\n", $ ) } }'
AA
Use of uninitialized value of type Any in string context.
Methods .^name, .raku, .gist, or .say can be used to stringify it to something
meaningful.
in block at -e line 1
Use of uninitialized value of type Any in string context.
Methods .^name, .raku, .gist, or .say can be used to stringify it to something
meaningful.
in any join at gen/moar/stage2/NQPCORE.setting line 1075
d:
AB
Use of uninitialized value of type Any in string context.
...
break it out of the parens, and it loses some "stateness":
raku -e 'for <AA NN> -> $alpha { for (1..14) { say state $ = $alpha; $++;
printf("d: %s\n", $ ) } }'
AA
Use of uninitialized value of type Any in string context.
Methods .^name, .raku, .gist, or .say can be used to stringify it to something
meaningful.
in block at -e line 1
...
AA
but the named doesn't
raku -e 'for <AA NN> -> $alpha { for (1..14) { state $sv = $alpha; say $sv;
$sv++; printf("d: %s\n", $sv ) } }'
AA
d: AB
AB
d: AC
________________________________
From: William Michels <[email protected]>
Sent: Tuesday, September 1, 2020 5:30 PM
To: Andy Bach <[email protected]>
Cc: yary <[email protected]>; perl6-users <[email protected]>
Subject: Re: print particular lines question
My first clue that something is amiss is in your third line of code when the
return skips "AA" and starts "AB, AC, AD....". That suggests to me that the two
step assign/printf call is playing havoc with the $ anonymous variable. Try
this instead:
~$ raku -e 'for <AA NN> -> $alpha { for (1..14) { printf("d: %s\n", (state $ =
$alpha)++ ) }; };'
d: AA
d: AB
d: AC
d: AD
d: AE
d: AF
d: AG
d: AH
d: AI
d: AJ
d: AK
d: AL
d: AM
d: AN
d: NN
d: NO
d: NP
d: NQ
d: NR
d: NS
d: NT
d: NU
d: NV
d: NW
d: NX
d: NY
d: NZ
d: OA
HTH, Bill.
On Tue, Sep 1, 2020 at 2:57 PM Andy Bach
<[email protected]<mailto:[email protected]>> wrote:
I'm barely hanging on with the "$" so ... so from:
raku -e 'for <AA NN> -> $alpha { for (1..14) { print (state $ = $alpha)++ ~ "
" } }'
AA AB AC AD AE AF
I tried an actual, er, non-anon var
# raku -e 'for <AA NN> -> $alpha { for (1..14) { print (state $sv = $alpha)++
~ " " } }'
AA AB AC AD AE AF ...
and then I tried
raku -e 'for <AA NN> -> $alpha { for (1..14) { (state $sv = $alpha)++;
printf("d: %s\n", $sv ) } }'
d: AB
d: AC
d: AD
d: AE
d: AF
...
but back to "$"
raku -e 'for <AA NN> -> $alpha { for (1..14) { (state $ = $alpha)++;
printf("d: %s\n", $ ) } }'
Use of uninitialized value of type Any in string context.
Methods .^name, .raku, .gist, or .say can be used to stringify it to something
meaningful.
in block at -e line 1
Use of uninitialized value of type Any in string context.
Methods .^name, .raku, .gist, or .say can be used to stringify it to something
meaningful.
in any join at gen/moar/stage2/NQPCORE.setting line 1075
d:
[27 more times]
I used printf hoping the %s context would stringify "$" as trying any of the
suggested "methods" complain of a missing "self"
raku -e 'for <AA NN> -> $alpha { for (1..14) { (state $ = $alpha)++;
printf("d: %s\n", $.raku ) } }'
===SORRY!=== Error while compiling -e
Variable $.raku used where no 'self' is available
at -e:1
------> v = $alpha)++; printf("d: %s\n", $.raku⏏ ) } }
expecting any of:
term
So I'm missing something about "$", I think
________________________________
From: William Michels via perl6-users
<[email protected]<mailto:[email protected]>>
Sent: Tuesday, September 1, 2020 3:17 PM
To: yary <[email protected]<mailto:[email protected]>>
Cc: perl6-users <[email protected]<mailto:[email protected]>>
Subject: Re: print particular lines question
I tried combining Larry's code and Yary's code, variously using
"state" or "INIT" or "BEGIN". This is what I saw:
~$ raku -e 'for <AA NN> -> $alpha { for (1..14) { print (state $ =
$alpha)++ ~ " " } }'
AA AB AC AD AE AF AG AH AI AJ AK AL AM AN NN NO NP NQ NR NS NT NU NV
NW NX NY NZ OA
~$ raku -e 'for <AA NN> -> $alpha { for (1..14) { print (INIT $ =
$alpha)++ ~ " " } }'
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
~$ raku -e 'for <AA NN> -> $alpha { for (1..14) { print (BEGIN $ =
$alpha)++ ~ " " } }'
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
Expected? --Bill.
On Tue, Sep 1, 2020 at 11:44 AM yary
<[email protected]<mailto:[email protected]>> wrote:
>
>
> Thanks, that's cool, and shows me something I was wondering about
>
> On Tue, Sep 1, 2020 at 11:36 AM Larry Wall
> <[email protected]<mailto:[email protected]>> wrote:
>>
>> If you want to re-initialize a state variable, it's probably better to make
>> it explicit with the state declarator:
>>
>> $ raku -e "for <a b> { for (1..2) { say (state $ = 'AAA')++ } }"
>> AAA
>> AAB
>> AAA
>> AAB
>
>
> $ raku -e 'for <AA OO> -> $alpha { for (1..3) { say (state $ = $alpha)++ } }'
> AA
> AB
> AC
> OO
> OP
> OQ
>
