Re: [RFC] input.cpp: Remove use of strncat(3)

[Alejandro Colomar]

Hi Steffen,

On 12/6/22 00:11, Steffen Nurpmeso wrote:
> Alejandro Colomar wrote in
>   <20221205223936.8290-1-alx@kernel.org>:
>   ...
>   |--- a/src/roff/troff/input.cpp
>   |+++ b/src/roff/troff/input.cpp
>   |@@ -7892,7 +7892,7 @@ void do_macro_source(bool quietly)
>   |    MACRO_POSTFIX, sizeof(MACRO_POSTFIX) - 1) == 0) {
>   |    char *s = new char[fnlen + sizeof(MACRO_PREFIX)];
>   |    strcpy(s, MACRO_PREFIX);
>   |-       strncat(s, fn, fnlen - sizeof(MACRO_POSTFIX) + 1);
>   |+          strlcat(s, fn, sizeof(MACRO_PREFIX));
>   |    fp = mac_path->open_file(s, &path);
>   |    delete[] s;
>
>
> To me this looks like a perfect usage of simple memcpy() since

Yes, memcpy(3) would be fine too.  Although, that hurts readability a little 
bit, since memcpy(3) doesn't tell so much that you're dealing with 
(NUL-terminated) string.  Since performance is not a big deal here, I guess 
strlcpy(3) would be just fine.  But yes, I'd very much prefer memcpy(3) over 
strncpy(3).

> both lengths are known.
> Maybe even better would be a C++ string type so you could do
> mac_path->open_file(resize().append().append().cp()), but this
> seems off topic.

Can't comment, since I don't know C++.  Keeping C-compatibility is a nice have, 
but I don't know how much of C++ strings groff already uses, so do what you have 
to do.

Cheers,

Alex

> --steffen
> |
> |Der Kragenbaer,                The moon bear,
> |der holt sich munter           he cheerfully and one by one
> |einen nach dem anderen runter  wa.ks himself off
> |(By Robert Gernhardt)

--
<http://www.alejandro-colomar.es/>

OpenPGP_signature
Description: OpenPGP digital signature