Le 13/02/2025 à 13:43, Pavel Sanda a écrit :
I do not know how others see it, but I do not fight these as a necessary
improvements.
You squashed temporary and little of CPU time, but unless it's in time critical
code
path the savings are negligible, but you lost semantics.
When I see in the code
string const name = cmd.getArg(0);
string const state = cmd.getArg(1);
I have idea what getArg(0)/getArg(1) are supposed to mean and can even
fix bugs if something goes wrong.
Seeing just if (GuiToolbar * t = toolbar(cmd.getArg(0)))
gives me no idea what arg 0 should be.
Could we reconsider whether to follow all coverity suggestions?
I didi not follow all coverity suggestions today. I fixed 12 of them
and, according to coverity's new stats, ignored 8 additional ones (I
though I ignored more of those actually).
In this case, the nearby code is:
case LFUN_TOOLBAR_TOGGLE: {
string const name = cmd.getArg(0);
if (GuiToolbar * t = toolbar(name))
flag.setOnOff(t->isVisible());
else {
enable = false;
docstring const msg =
bformat(_("Unknown toolbar \"%1$s\""),
from_utf8(name));
flag.message(msg);
}
break;
}
This second one is not my doing, AFAIR.
Note that my main reason for doing the change here is to shut coverity
up. I could equally do that by Ignore-ing the issue if the consensus is
that it is better. In this case, I thought that using cmd.getArg(1) as a
parameter of setState() was a good enough hint.
JMarc
--
lyx-devel mailing list
lyx-devel@lists.lyx.org
https://lists.lyx.org/mailman/listinfo/lyx-devel
