On Mon, 6 May 2024 20:22:48 +0200, Daniel Kiper wrote: >You can make this check much simpler. Please take a look at the commit >ac8a37dda (net/http: Allow use of non-standard TCP/IP ports).
I did it that way in order to preserve a feature of Glenn's suggested approach: passing through any grub_errno value set by grub_strtoul(). (I merely added an additional case to update grub_errno if and only if we generate an error of our own.) Now that you mention it, though, I suppose it's okay to overwrite grub_strtoul's errno values as long as the replacement values are sensible. That function's logic seems unlikely to change in the future, after all, and we're validating the string more strictly than it does. >In case of an error here I would assume default value. I think this >behavior should be documented then. Yes, that makes sense, and would simplify the code a bit. I'll do that in the next rev. _______________________________________________ Grub-devel mailing list Grub-devel@gnu.org https://lists.gnu.org/mailman/listinfo/grub-devel
