From: Junio C Hamano <gitster@pobox.com>
Date: Sat, 09 May 2020 17:33:30 GMT
Subject: Re: [PATCH v3 2/3] multi-pack-index: respect repack.packKeptObjects=false
Message-ID: <xmqqlfm1ui6t.fsf@gitster.c.googlers.com>
In-Reply-To: <20200509161159.GA15146@danh.dev>

Đoàn Trần Công Danh  <congdanhqx@gmail.com> writes:

> On 2020-05-09 14:24:29+0000, Derrick Stolee via GitGitGadget <gitgitgadget@gmail.com> wrote:
>> From: Derrick Stolee <dstolee@microsoft.com>
>> 
>> +test_expect_success 'repack respects repack.packKeptObjects=false' '
>> +	test_when_finished rm -f dup/.git/objects/pack/*keep &&
>> +	(
>> +		cd dup &&
>> +		ls .git/objects/pack/*idx >idx-list &&
>
> I think ls(1) is an overkill.
> I think:
>
> 	echo .git/objects/pack/*idx
>
> is more efficient.

When there is no file whose name ends with idx, what happens?

    $ ls *idx && echo OK
    ls: cannot access '*idx': No such file or directory
    $ echo *idx && echo OK
    *idx
    OK

>> +		test_line_count = 5 idx-list &&
>> +		ls .git/objects/pack/*.pack | sed "s/\.pack/.keep/" >keep-list &&
>
> Likewise.

Likewise.

>> +		for keep in $(cat keep-list)
>> +		do
>> +			touch $keep || return 1
>
> Is this intended?
> Since touch(1) accepts multiple files as argument.

Good suggestion, but doesn't .keep file record why the pack is kept
in real life (i.e. not an empty file)?

>> +		done &&
>> +		git multi-pack-index repack --batch-size=0 &&
>> +		ls .git/objects/pack/*idx >idx-list &&
>> +		test_line_count = 5 idx-list &&
>> +		test-tool read-midx .git/objects | grep idx >midx-list &&
>> +		test_line_count = 5 midx-list &&
>> +		THIRD_SMALLEST_SIZE=$(test-tool path-utils file-size .git/objects/pack/*pack | sort -n | head -n 3 | tail -n 1) &&
>
> This line is overly long.
> Should we write test-tool's output to temp file and process it?
>
> And I think either
>
> 	sed -n '3{p;q}'
>
> or:
>
> 	sed -n 3p
>
> is cleaner than
>
> 	head -n 3 | tail -n 1

"sed -n 3p" is the only valid way to write it ;-)