Toon Claes [off-list ref] writes:

karthik nayak [off-list ref] writes:

quoted

Patrick Steinhardt [off-list ref] writes:

quoted

On Fri, Dec 13, 2024 at 11:36:50AM +0100, Karthik Nayak wrote:

quoted

The `ref_transaction_add_update()` creates the `ref_update` struct. To
facilitate addition of reflogs in the next commit, the function needs to
accommodate setting the `committer_info` field in the struct. So modify
the function to also take `committer_info` as an argument and set it
accordingly.

I was wondering a bit whether we could instead pull out a
`add_update_internal()` function so that we don't need to modify all
callers of `ref_transaction_add_update()`. Because ultimately, we don't
use the field anywhere except from `ref_transaction_add_reflog_update()`
as far as I can see.

This is more of a thought than a strong opinion, so feel free to ignore.

Yes, that is a possible change, but the number of code changes are
relatively low and I didn't think it made so much difference. Also
because we'd now have one more function. But I don't mind doing it
either, if anyone feels strongly about it, I'll happily make that
change.

Yes, I agree the number of callsites isn't that large, but on the other
hand, I see various calls to this function having four `NULL`s in a row
as arguments. Personally, I think that starts to smell a bit.

I agree with your reasoning here..

Now, before you change anything. I'm not sure what Patrick was
suggesting? Would it mean we basically rename
`ref_transaction_add_update()` to `add_update_internal()` and create a
new wrapper function `ref_transaction_add_update()` that simply calls
`add_update_internal(<ARGS>..., NULL, msg)`? I don't think that's a
great solution either.

Yes, agreed with this too.

Alternively, because ref_transaction_add_update() returns the `struct
ref_update`, why not add a function `ref_update_set_committer` and call
that where we need to set the committer? I see this also will help in a
future commit where you call ref_transaction_add_update() differently
depending on reflog updates.

I think this seems like a nice way to go about it. Currently all the
logic pertaining to creating an `ref_update` struct is contained within
`ref_transaction_add_update()`. So having individual functions would
make sense, but the con here is that this doesn't enforce fields to be
set. But for `committer_info` it does make sense. I'm going to leave it
for now since the series is merged to `next`. Maybe something to do in
the future as part of #leftoverbits.

--
Toon