[chbot] Weirdness in parameter interpretation in C lib function

Stuart Brown stuartbrown2 at gmail.com
Tue Jul 30 00:39:31 BST 2024


It all depends on what documentation you look at. For example
https://www.w3schools.com/c/ref_stdio_snprintf.php states it is the size of
the output buffer. The point of it is to avoid buffer overruns, so size of
output buffer makes sense.

If you want to use size of data, then you need to account for the extra
bytes you are adding in the format string. The SSID is 30, and you add
"SSID" a space and a new line. This makes 36 bytes, plus the null
terminator added by the function. so your second argument must be "sizeof
(scandata[index].ssid) + 7". But then the onus is on you to ensure that
'buf' is large enough

If instead you use 'snprintf(buf, sizeof(buf), "SSID
%s\n",scandata[index].ssid);' then you need to worry, the compiler will
catch the situation where the data will overflow the buffer.

On Tue, 30 Jul 2024 at 11:21, Robin Gilks <gb7ipd at gmail.com> wrote:

> You make the same assumption as the library coder - that 'size' is the
> length of the output buffer. The man page says otherwise
>
> *The functions snprintf() and vsnprintf() write at most size bytes
> (including the terminating null byte ('\0')) to str.*
>
>
> *This is defining the amount written, not the maximum the outbuffer can
> hold - nowhere does it say that 'size' is the length of 'str'*
>
>
>
> On Tue, Jul 30, 2024 at 11:12 AM Simeon Pilgrim <simeon.pilgrim at gmail.com>
> wrote:
>
>> ```
>> char* test(uint8_t index)
>> {
>> #define DEST_BUF_SIZE 100
>>
>>   if (index < NUMRESULTS) {
>>     static char buf[DEST_BUF_SIZE];
>>
>>     snprintf(buf, DEST_BUF_SIZE, "SSID %s\n", scandata[index].ssid);
>>     return buf;
>>   }
>>   return "INVALID INDEX";
>> }
>> ```
>>
>> On Tue, 30 Jul 2024 at 11:05, Simeon Pilgrim <simeon.pilgrim at gmail.com>
>> wrote:
>>
>>> you have provided a pointer (to memory), AND you have declared it 30
>>> bytes long. This is what YOU HAVE done. The compiler is HELPING YOU by
>>> pointing out 37 > 30, which it is.
>>>
>>> If you want the compiler to help you correctly, tell the compiler the
>>> length of the data you have, which is 100.
>>>
>>>
>>>
>>>
>>>
>>> On Tue, 30 Jul 2024 at 10:58, Robin Gilks <gb7ipd at gmail.com> wrote:
>>>
>>>> But I've provided 100 bytes, not 30!!
>>>>
>>>> On Tue, Jul 30, 2024 at 10:40 AM Simeon Pilgrim
>>>> <simeon.pilgrim at gmail.com> wrote:
>>>> >
>>>> > the second parameter should be the size of the destination buffer,
>>>> thus the code should be:
>>>> >
>>>> > snprintf (buf, 100, "SSID %s\n",
>>>> >
>>>> > or
>>>> >
>>>> > snprintf (buf, sizeof(buf), "SSID %s\n",
>>>> >
>>>> > your format string is 7 tokens (6 + zero termination) + the 30 byte
>>>> string ssid, which is larger than 30 you have provided.
>>>> >
>>>> > The snprintf functions are "safe truncation" functions, not handy
>>>> dandy truncation utilities, thus the warning that you might have truncation.
>>>> >
>>>> > On Tue, 30 Jul 2024 at 10:27, Robin Gilks <gb7ipd at gmail.com> wrote:
>>>> >>
>>>> >> Since there are a few C programmers in the group I thought I'd ask
>>>> >> this (interesting?) question
>>>> >>
>>>> >> Here is a bit of sample code to illustrate:
>>>> >>
>>>> >>
>>>> ----------------------------------------------------------------------------
>>>> >> #include <stdio.h>
>>>> >> #include <stdint.h>
>>>> >>
>>>> >>
>>>> >> typedef struct
>>>> >> {
>>>> >> char ssid[30];
>>>> >> int8_t rssi;
>>>> >> } ScanResult;
>>>> >>
>>>> >> #define NUMRESULTS 50 // how many tracked simultaneously
>>>> >>
>>>> >> static ScanResult scandata[NUMRESULTS];
>>>> >>
>>>> >> void
>>>> >> main (void)
>>>> >> {
>>>> >> }
>>>> >>
>>>> >>
>>>> >> char * test (uint8_t index)
>>>> >> {
>>>> >> static char buf[100];
>>>> >>
>>>> >> snprintf (buf, sizeof (scandata[index].ssid), "SSID %s\n",
>>>> >> scandata[index].ssid);
>>>> >> return buf;
>>>> >>
>>>> >> }
>>>> >>
>>>> ----------------------------------------------------------------------------
>>>> >> Save as test.c; compile with gcc test.c
>>>> >>
>>>> >> The warning indicates that the destination buffer may be too small
>>>> >>
>>>> >> test.c: In function ‘test’:
>>>> >> test.c:25:56: warning: ‘%s’ directive output may be truncated writing
>>>> >> up to 29 bytes into a region of size 25 [-Wformat-truncation=]
>>>> >>    25 |    snprintf (buf, sizeof (scandata[index].ssid), "SSID %s\n",
>>>> >> scandata[index].ssid);
>>>> >>       |                                                        ^~
>>>> >> test.c:25:4: note: ‘snprintf’ output between 7 and 36 bytes into a
>>>> >> destination of size 30
>>>> >>    25 |    snprintf (buf, sizeof (scandata[index].ssid), "SSID %s\n",
>>>> >> scandata[index].ssid);
>>>> >>       |
>>>> >>
>>>> >> This appear to be treating the size parameter in the snprintf as
>>>> being
>>>> >> the size of the output buffer 'buf' (which is 100 bytes long) but
>>>> >> surely it should be applying the restriction on the format string
>>>> that
>>>> >> includes  scandata[index].ssid (which is 30 bytes long)
>>>> >>
>>>> >> Am I just having brain fade or is that just fundamentally wrong?
>>>> >>
>>>> >> --
>>>> >> Robin Gilks
>>>> >>
>>>> >> _______________________________________________
>>>> >> Chchrobotics mailing list Chchrobotics at lists.ourshack.com
>>>> >> https://lists.ourshack.com/mailman/listinfo/chchrobotics
>>>> >> Mail Archives: http://lists.ourshack.com/pipermail/chchrobotics/
>>>> >> Meetings usually 3rd Monday each month. See http://kiwibots.org for
>>>> venue, directions and dates.
>>>> >> When replying, please edit your Subject line to reflect new subjects.
>>>> >
>>>> > _______________________________________________
>>>> > Chchrobotics mailing list Chchrobotics at lists.ourshack.com
>>>> > https://lists.ourshack.com/mailman/listinfo/chchrobotics
>>>> > Mail Archives: http://lists.ourshack.com/pipermail/chchrobotics/
>>>> > Meetings usually 3rd Monday each month. See http://kiwibots.org for
>>>> venue, directions and dates.
>>>> > When replying, please edit your Subject line to reflect new subjects.
>>>>
>>>> _______________________________________________
>>>> Chchrobotics mailing list Chchrobotics at lists.ourshack.com
>>>> https://lists.ourshack.com/mailman/listinfo/chchrobotics
>>>> Mail Archives: http://lists.ourshack.com/pipermail/chchrobotics/
>>>> Meetings usually 3rd Monday each month. See http://kiwibots.org for
>>>> venue, directions and dates.
>>>> When replying, please edit your Subject line to reflect new subjects.
>>>
>>> _______________________________________________
>> Chchrobotics mailing list Chchrobotics at lists.ourshack.com
>> https://lists.ourshack.com/mailman/listinfo/chchrobotics
>> Mail Archives: http://lists.ourshack.com/pipermail/chchrobotics/
>> Meetings usually 3rd Monday each month. See http://kiwibots.org for
>> venue, directions and dates.
>> When replying, please edit your Subject line to reflect new subjects.
>
> _______________________________________________
> Chchrobotics mailing list Chchrobotics at lists.ourshack.com
> https://lists.ourshack.com/mailman/listinfo/chchrobotics
> Mail Archives: http://lists.ourshack.com/pipermail/chchrobotics/
> Meetings usually 3rd Monday each month. See http://kiwibots.org for
> venue, directions and dates.
> When replying, please edit your Subject line to reflect new subjects.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ourshack.com/pipermail/chchrobotics/attachments/20240730/b0972682/attachment-0001.html>


More information about the Chchrobotics mailing list