Bug #12035: scanf suspicious results.

• Status: Rejected
• Assignee: -
• Target version: -
• ruby -v: ruby 2.2.3p173 (2015-08-18 revision 51636) [i386-cygwin], ruby 2.1.6p336 (2015-04-13 revision 50298) [i386-mingw32], and ruby 1.9.3p484 (2013-11-22) [i386-mingw32]
• Backport: 2.0.0: UNKNOWN, 2.1: UNKNOWN, 2.2: UNKNOWN, 2.3: UNKNOWN

[PeterCamilleri (Peter Camilleri)]

The code: 'a b c'.scanf('%[a] %[b] %[c]')

yields the result: ["a"] and not ["a","b","c"] as expected.

So far it seems that when sets of characters are used, only
the first one in the parse specification string actually returns
any data.

[musgravejw (John Musgrave)]

Peter Camilleri wrote:

The code: 'a b c'.scanf('%[a] %[b] %[c]')

yields the result: ["a"] and not ["a","b","c"] as expected.

So far it seems that when sets of characters are used, only
the first one in the parse specification string actually returns
any data.

I think you want: "a b c".scanf("%s %s %s")

[PeterCamilleri (Peter Camilleri)]

I think you're missing the point. Of course %s would seem to work in this case.
My question is about the %[set] (and %[^set]) format specifier that is documented as part of scanf.
It also works, but only once. That's the rub.

[musgravejw (John Musgrave)]

Peter Camilleri wrote:

I think you're missing the point. Of course %s would seem to work in this case.
My question is about the %[set] (and %[^set]) format specifier that is documented as part of scanf.
It also works, but only once. That's the rub.

Well, the sets in your string are whitespace delimited, so there is a mismatch between your input string and the format. You would need to match on the whitespace character.

"de f g".scanf("%[de]%s%[f]%s%[g]")
=> ["de", "f", "g"]

[PeterCamilleri (Peter Camilleri)]

That is what is so perplexing. The format I used was '%[a] %[b] %[c]'
Please note the explicit spaces between the formant elements. The docs
for scanf state that "White space in the format string matches any
amount of white space"

[nobu (Nobuyoshi Nakada)]

• Status changed from Open to Rejected

From http://ruby-doc.org/stdlib-2.2.4/libdoc/scanf/rdoc/Scanf.html#module-Scanf-label-Conversions

Matches a nonempty sequence of characters from the specified set of accepted characters. The usual skip of leading white space is suppressed.