[Rd] incorrect output and segfaults from sprintf with %*d (PR#13675)

maechler at stat.math.ethz.ch maechler at stat.math.ethz.ch
Sat Apr 25 15:00:13 CEST 2009


On Fri, Apr 24, 2009 at 14:40, Wacek Kusnierczyk
<Waclaw.Marcin.Kusnierczyk at idi.ntnu.no> wrote:
> maechler at stat.math.ethz.ch wrote:
>>
>> =A0 =A0 vQ> sprintf has a documented limit on strings included in the ou=
tput using the
>> =A0 =A0 vQ> format '%s'. =A0It appears that there is a limit on the leng=
th of strings included
>> =A0 =A0 vQ> with, e.g., the format '%d' beyond which surprising things h=
appen (output
>> =A0 =A0 vQ> modified for conciseness):
>> =A0 =A0 >>>
>>
>> =A0 =A0 vQ> ... and this limit is *not* documented.
>>
>> =A0 =A0 MM> well, it is basically (+ a few bytes ?)
>> =A0 =A0 MM> the same =A08192 =A0limit that *is* documented.
>>
>> indeed, I was right with that..
>>
>
> hmm, i'd guess this limit is valid for all strings included in the
> output with any format? =A0not just %s (and, as it appears, undocumentedl=
y
> %d)?

yes.

>> =A0 =A0 vQ> while snprintf would help avoid buffer overflow, it may not =
be a
>> =A0 =A0 vQ> solution to the issue of confused output.
>>
>> =A0 =A0 MM> I think it would / will. =A0We would be able to give warning=
s and
>> =A0 =A0 MM> errors, by checking the =A0snprintf() =A0return codes.
>>
>> My current working code gives an error for all the above
>> examples, e.g.,
>>
>> =A0> sprintf('%9999d', 1)
>> =A0Error in sprintf("%9999d", 1) :
>> =A0 =A0required resulting string length 9999 is > maximal 8191
>>
>> it passes =A0'make check-devel' and I am inclined to commit that
>> code to R-devel (e.g. tomorrow).
>>
>> Yes, the documentation will also have to be amended, but apart
>> from that, would people see a big problem with the "8192" limit
>> which now is suddenly of greater importance
>> {{as I said all along; =A0hence my question to Wacek (and the
>> =A0 R-develers) =A0if anybody found that limit too low}}
>>
>
> i didn't find the limit itself problematic. =A0(so far?)

ok.

> btw. (i do know what that means ;)), after your recent fix:
>
> =A0 =A0sprintf('%q%s', 1)
> =A0 =A0# Error in sprintf("%q%s", 1) :
> =A0 =A0# =A0use format %f, %e, %g or %a for numeric objects
>
> =A0 =A0sprintf('%s', 1)
> =A0 =A0# [1] "1"
>
> you may want to add '%s' (and '%x', and ...) to the error message. =A0or
> perhaps make it say sth like 'invalid format: ...'. =A0the problem is not
> that %q is not applicable to numeric, but that it is not a valid format
> at all.

yes.  As a matter of fact,  "%q%s" is dealt with as *one* format
chunk, since "%q" is
not a valid format.
The code I have just committed now gives a longer erro message, that
should be more helpful.
Thank you for the suggestion!

> there's also an issue with the additional arguments supplied after the
> format: =A0any superfluous arguments are ignored (this is not documented,
> as far as i can see),

I think we should signal an error if there are too many arguments.
Could anyone think of a case where the current behavior is desirable ?

> but they *are* evaluated nevertheless, e.g.:
>
> =A0 =A0sprintf('%d', 0, {print(1)})
> =A0 =A0# "1"
> =A0 =A0# [1] "0"
>
> it might be a good idea to document this behaviour.

actually I think it should be changed to be more strict (see above).

Thank you for the constructive feedback!
Martin



More information about the R-devel mailing list