Re: [PATCH v4 1/3] branch: avoid unnecessary worktrees traversals

From: Jonathan Tan <jonathantanmy@google.com>
Date: Feb 27, 2023, 19:30 UTC
Message-ID: <20230227193012.2410973-1-jonathantanmy@google.com>
In-Reply-To: <4e6aeffe-98ef-b4d7-7c8f-782e0a617653@gmail.com>

Rubén Justo <rjusto@gmail.com> writes:

Show 5 quoted lines

> Reviewing this, I noticed I made a mistake here.  The original code
> doesn't stop iterating whenever refs_create_symref() fails; it continues
> trying to update the remaining worktrees.  When the iteration ends, if
> any of the updates failed, then die().  Also, the error message "HEAD of
> working tree %s is not updated" is lost.

Ah yes, I noticed this too.

Besides that, a reviewer, upon reading the commit message, might ask: why not take the worktrees as a parameter then, if we're so worried about performance? I think that the real reason for inlining is that the code being inlined needs to communicate more information to its calling function in subsequent patches; the performance improvement is only a beneficial side effect.