From: Junio C Hamano <gitster@pobox.com>
Date: Fri, 16 May 2014 17:36:15 GMT
Subject: Re: [PATCH 1/2] config: be strict on core.commentChar
Message-ID: <xmqqsio9aank.fsf@gitster.dls.corp.google.com>
In-Reply-To: <20140516162508.GA12314@google.com>

Jonathan Nieder <jrnieder@gmail.com> writes:

> Nguyễn Thái Ngọc Duy wrote:
>
>> --- a/config.c
>> +++ b/config.c
>> @@ -826,8 +826,12 @@ static int git_default_core_config(const char *var, const char *value)
>>  	if (!strcmp(var, "core.commentchar")) {
>>  		const char *comment;
>>  		int ret = git_config_string(&comment, var, value);
>> -		if (!ret)
>> -			comment_line_char = comment[0];
>> +		if (!ret) {
>> +			if (comment[0] && !comment[1])
>> +				comment_line_char = comment[0];
>> +			else
>> +				return error("core.commentChar should only be one character");
>> +		}
>
> Perhaps, to decrease indentation a little:
>
> 		if (ret)
> 			return ret;
> 		if (comment[0] && !comment[1])
> 			comment_line_char = comment[0];
> 		else
> 			return error(...);
> 		return 0;
>
> [...]
>> --- a/t/t7508-status.sh
>> +++ b/t/t7508-status.sh
>> @@ -1348,12 +1348,6 @@ test_expect_success "status (core.commentchar with submodule summary)" '
>>  	test_i18ncmp expect output
>>  '
>>  
>> -test_expect_success "status (core.commentchar with two chars with submodule summary)" '
>> -	test_config core.commentchar ";;" &&
>> -	git -c status.displayCommentPrefix=true status >output &&
>> -	test_i18ncmp expect output
>
> Could keep the test to avoid regressions:
>
> 	test_config core.commentchar ";;" &&
> 	test_must_fail git -c status.displayCommentPrefix=true status

All good points, including your other review message.