From: Junio C Hamano <gitster@pobox.com>
Date: Thu, 25 Jan 2024 22:46:02 GMT
Subject: Re: [PATCH 1/2] index-pack: test and document --strict=<msg>
Message-ID: <xmqq8r4dt4k5.fsf@gitster.g>
In-Reply-To: <9b353aff73d6351b86cc7b55982f1565e76d08e9.1706215884.git.gitgitgadget@gmail.com>

"John Cai via GitGitGadget" <gitgitgadget@gmail.com> writes:

> From: John Cai <johncai86@gmail.com>
>
> 5d477a334a (fsck (receive-pack): allow demoting errors to warnings,
> 2015-06-22) allowed a list of fsck msg to downgrade to be passed to
> --strict. However this is a hidden argument that was not documented nor
> tested. Though true that most users would not call this option
> direction, (nor use index-pack for that matter) it is still useful to

Though it is true that ... call this option directly (nor use
index-pack, for that matter), it is still ...

or something like that, probably.

> document and test this feature.

And I agree with that.  Thanks for adding the necessary doc.

> +--strict[=<msg-ids>]::

<msg-id> in the context of "git fsck --help" seems to refer to the
left hand side of <msg-id>=<severity>.  <msg-ids> sounds as if it is
just the list of <msg-id> without saying anything about their severity,
which is not what we want to imply.

Either use a made-up word that is clearly different and can not be
mistaken as a list of <msg-id>, or spell it out a bit more
explicitly, may make it easier to follow?

	--strict[=<fsck-config>]
	--strict[=<msg-id>=<severity>...]

I dunno.

Use of <msg-id> and <severity> below looks good in the body of the
paragraph here.

> +	Die, if the pack contains broken objects or links. If `<msg-ids>` is passed,
> +	it should be a comma-separated list of `<msg-id>=<severity>` elements where
> +	`<msg-id>` and `<severity>` are used to change the severity of some possible
> +	issues, eg: `--strict="missingEmail=ignore,badTagName=error"`. See the entry
> +	for the `fsck.<msg-id>` configuration options in `linkgit:git-fsck[1] for
> +	more information on the possible values of `<msg-id>` and `<severity>`.

"eg:" -> "e.g.," probably.

> diff --git a/t/t5300-pack-object.sh b/t/t5300-pack-object.sh
> index d402ec18b79..9563372ae27 100755
> --- a/t/t5300-pack-object.sh
> +++ b/t/t5300-pack-object.sh
> @@ -441,6 +441,28 @@ test_expect_success 'index-pack with --strict' '
>  	)
>  '
>  
> +test_expect_success 'index-pack with --strict downgrading fsck msgs' '
> +	test_when_finished rm -rf strict &&
> +	git init strict &&
> +	(
> +		cd strict &&
> +		test_commit first hello &&
> +		cat >commit <<-EOF &&
> +		tree $(git rev-parse HEAD^{tree})
> +		parent $(git rev-parse HEAD)
> +		author A U Thor
> +		committer A U Thor
> +
> +		commit: this is a commit wit bad emails

"wit" -> "with"; as this typo does not contribute anything to
the badness we expect index-pack to notice, it would pay to
make sure we do not have it, to avoid distracting readers.

> +		EOF
> +		git hash-object --literally -t commit -w --stdin <commit >commit_list &&
> +		PACK=$(git pack-objects test <commit_list) &&
> +		test_must_fail git index-pack --strict "test-$PACK.pack" &&
> +		git index-pack --strict="missingEmail=ignore" "test-$PACK.pack"
> +	)
> +'
> +
>  test_expect_success 'honor pack.packSizeLimit' '
>  	git config pack.packSizeLimit 3m &&
>  	packname_10=$(git pack-objects test-10 <obj-list) &&