Re: tb/ban-strtok
- From
Taylor Blau <me@ttaylorr.com>
- Date
- Apr 18, 2023, 19:31 UTC
- Message-ID
- <ZD7wCYDmhiKcfhow@nand.local>
- In-Reply-To
- <xmqqo7nldt5f.fsf@gitster.g>
On Tue, Apr 18, 2023 at 09:58:20AM -0700, Junio C Hamano wrote:
Show 16 quoted lines
> Taylor Blau <me@ttaylorr.com> writes: > > >> Mark strtok() and strtok_r() to be banned. > >> > >> source: <cover.1681428696.git.me@ttaylorr.com> > > > > This one needs a reroll to remove `strtok_r()` from the proposed banned > > list, and to support slightly different semantics the new > > string_list_split_in_place_multi() function. Will send today. > > I actually do not think we would seriously miss strtok_r(); as long > as the rationale for doing so is clearly described, I think we are > OK. Just do not sell string-list as if it were the solution to > everything (which is not---it is a rather poor data structure), but > compare string_list_split() with strtok_r() and say our application > hasn't needed the flexibility of the latter.
I agree that we wouldn't miss it at all, since we have no uses of strtok_r() today.
But let's decide on banning it separately, since the much more immediate-term issues are the existing uses of strtok() that are in the tree.
Thanks, Taylor