(r) flag on scalar

(r) flag on scalar

[Han Pingtian]

I think below command line should give result "bc", but:

localhost% s=abcdefg
localhost% print $s[(r)bc]
b
localhost%

Looks like this isn't intuitional.

Re: (r) flag on scalar

[Bart Schaefer]

On Mar 9,  7:38am, Han Pingtian wrote:
} Subject: (r) flag on scalar
}
} I think below command line should give result "bc", but:
} 
} localhost% s=abcdefg
} localhost% print $s[(r)bc]
} b
} localhost%
} 
} Looks like this isn't intuitional.

$s[(r)bc] is a single index, so it's the index where the first match for
"bc" begins.

$s[(r)bc,(r)ef] is a range of indices, so it begins where "bc" begins and
ends where "ef" ends.

The only thing not quite intuitive about this is that a wildcard (either
* or with extendedglob ?# ) at the end of the ending patterrn is not
considered part of the match for (r) but is part of the match for (R).
PWS's latest doc tweak still doesn't explicitly describe that.

Re: (r) flag on scalar

[Peter Stephenson]

On Sun, 8 Mar 2015 19:56:42 -0700
Bart Schaefer <schaefer@brasslantern.com> wrote:
> The only thing not quite intuitive about this is that a wildcard (either
> * or with extendedglob ?# ) at the end of the ending patterrn is not
> considered part of the match for (r) but is part of the match for (R).
> PWS's latest doc tweak still doesn't explicitly describe that.

diff --git a/Doc/Zsh/params.yo b/Doc/Zsh/params.yo
index 7b127bc..1b175b2 100644
--- a/Doc/Zsh/params.yo
+++ b/Doc/Zsh/params.yo
@@ -197,7 +197,17 @@ example(string="abcdefghijklm"
 print ${string[+LPAR()r+RPAR()d?,+LPAR()r+RPAR()h?]})
 
 prints `tt(defghi)'.  This is an obvious generalisation of the
-rule for single-character matches.
+rule for single-character matches.  For a single subscript,
+only a single character is referenced (not the range of chracters
+covered by the match).
+
+Note that in substring operations the second subscript is handled
+differently by the tt(r) and tt(R) subscript flags: the former takes the
+shortest match as the length and the latter the longest match.  Hence
+in the former case a tt(*) at the end is redundant while in
+the latter case it matches the whole remainder of the string.  This
+does not affect the result of the single subscript case as here the
+length of the match is irrelevant.
 
 subsect(Array Element Assignment)
 

Re: (r) flag on scalar

[Mikael Magnusson]

On Mon, Mar 9, 2015 at 10:08 AM, Peter Stephenson
<p.stephenson@samsung.com> wrote:
> On Sun, 8 Mar 2015 19:56:42 -0700
> Bart Schaefer <schaefer@brasslantern.com> wrote:
>> The only thing not quite intuitive about this is that a wildcard (either
>> * or with extendedglob ?# ) at the end of the ending patterrn is not
>> considered part of the match for (r) but is part of the match for (R).
>> PWS's latest doc tweak still doesn't explicitly describe that.
>
> diff --git a/Doc/Zsh/params.yo b/Doc/Zsh/params.yo
> index 7b127bc..1b175b2 100644
> --- a/Doc/Zsh/params.yo
> +++ b/Doc/Zsh/params.yo
> @@ -197,7 +197,17 @@ example(string="abcdefghijklm"
>  print ${string[+LPAR()r+RPAR()d?,+LPAR()r+RPAR()h?]})
>
>  prints `tt(defghi)'.  This is an obvious generalisation of the
> -rule for single-character matches.
> +rule for single-character matches.  For a single subscript,
> +only a single character is referenced (not the range of chracters

characters*

-- 
Mikael Magnusson