Re: [PATCH v5 1/3] parser_json: support handle for rule positioning in explicit JSON format

[Phil Sutter <phil@nwl.cc>]

Hi Alexandre,

On Tue, Jan 20, 2026 at 03:27:58PM +0100, Alexandre Knecht wrote:
> Hi Phil,
> 
> Thanks for the comment, that's pretty straightforward to fix, I'm
> afraid to do a lot of spamming if I post again a new series, so can
> you confirm this is what you expect ?
> 
> Merged nested if-conditionals (cheap to expensive):
> - if (!(ctx->flags & CTX_F_IMPLICIT) &&
> - !json_unpack(root, "{s:I}", "handle", &h.handle.id)) {
> - if (op == CMD_INSERT || op == CMD_ADD || op == CMD_CREATE) {
> - h.position.id = h.handle.id;
> - h.handle.id = 0;
> - }
> - }
> + if (!(ctx->flags & CTX_F_IMPLICIT) &&
> +   (op == CMD_INSERT || op == CMD_ADD || op == CMD_CREATE) &&
> +   !json_unpack(root, "{s:I}", "handle", &h.handle.id)) {
> +     h.position.id = h.handle.id;
> +     h.handle.id = 0;
> + }

Yes, this looks correct!

> Reverse Christmas Tree variable declarations:
> - unsigned int i;
> - json_t *tmp;
> uint32_t old_flags;
> struct cmd *cmd;
> + unsigned int i;
> + json_t *tmp;

Also correct AFAICT.

> Or maybe there's a solution to amend this series, not kinda used to
> work with git send-email, so if I can resubmit without a new whole
> series, could be good ! Otherwise, I'll just create a new one once you
> confirm.

I can apply trivial changes to patches when applying them. With some
projects, people also just resubmit parts of the series - this causes a
bit of a mess for maintainers (and reviewers) though so it's not usually
done. With a small series like this, I would just resubmit the whole
thing. After all, becoming more familiar with git-send-email and the
whole process of amending changes and submitting a new version is very
good practice for contributing to OSS projects! And please don't forget,
we're here to help if you get stuck or don't know how to get started
with something.

> Maybe I'll wait for review on tests too before submitting everything again.

Just finished, only one more change in second patch requested.

Thanks, Phil