Re: [PATCH] revision: --include-diversions adds helpful merges

From: Junio C Hamano <gitster@pobox.com>
Date: Apr 8, 2020, 19:46 UTC
Message-ID: <xmqqzhbloj4u.fsf@gitster.c.googlers.com>
In-Reply-To: <74562b04-b1ce-cad4-da18-4af030a3dc29@gmail.com>

Derrick Stolee <stolee@gmail.com> writes:

Show 22 quoted lines

> On 4/7/2020 9:39 PM, Derrick Stolee wrote:
>> On 4/7/2020 9:30 PM, Junio C Hamano wrote:
>>> "Derrick Stolee via GitGitGadget" <gitgitgadget@gmail.com> writes:
>>>
>>>>     This --include-diversions option could use a better name.
>>>
>>> True, but I do not think of a better (or for that matter a worse)
>>> one.  
>
> Here are some alternative names:
>
> 	--audit-merges
> 	--audit-trunk
> 	--first-merges
> 	--subtle-merges
> 	--more-merges
>
> The "audit" name implies some of the intent: we are trying to
> audit which merge commits introduced these changes. The --audit-trunk
> implies we are using a trunk-based workflow where parent order is
> critical. However, "trunk" may be confusing when there are multiple
> long-lived branches.

Our features written with the intent to be useful for one purpose often end up being used for purposes other than what the feature was originally written for (the "--pickaxe" has always been a bitter example of this for me).

For that reason, I am a bit hesitant to endorse "audit" exactly because of the implication of "intent".

We usually try to give a single simplest history that explains how the current content in the path came to be. For that, commit M in the illustration in your original message does not really help when we ignore which parent chain owns the history, but in practice, many workflows treat the first parent chain as something special, so we want to show, not just the individual changes that matter, where the changes are introduced in the first-parent chain.

I wonder if there is a simple-enough phrase to convey what the latter half of above sentence says. "include" and "keep" are both good verbs---normally we discard these merges, because they do not contribute at the level of individual changes, but with the option, we "include" or "keep" these merges in the output. It's not like we keep _all_ the merges, but selected merges only. How do we decide which merges to keep?

I guess your "--first-merges" came from such a line of thought, and is the closest among the five to what I have in mind, but it drops too many words and loses too much meaning.

"--keep-first-parent-merges", perhaps?