From: Junio C Hamano <gitster@pobox.com>
Date: Mon, 29 Feb 2016 19:47:51 GMT
Subject: Re: [PATCH v5 3/3] git: submodule honor -c credential.* from command line
Message-ID: <xmqqa8mj9u3c.fsf@gitster.mtv.corp.google.com>
In-Reply-To: <CA+P7+xoNdsDH0jBz6bcikpeMa-qMR6795U+hKqGDSfu52dJFSw@mail.gmail.com>

Jacob Keller <jacob.keller@gmail.com> writes:

> On Mon, Feb 29, 2016 at 10:20 AM, Junio C Hamano <gitster@pobox.com> wrote:
>> Jacob Keller <jacob.e.keller@intel.com> writes:
>>
>>> +static int sanitize_submodule_config(const char *var, const char *value, void *data)
>>> +{
>>> +     struct strbuf quoted = STRBUF_INIT;
>>> +     struct strbuf *out = data;
>>> +
>>> +     if (submodule_config_ok(var)) {
>>> +             if (out->len)
>>> +                     strbuf_addch(out, ' ');
>>> +
>>> +             sq_quotef(out, "%s=%s", var, value);
>>
>> Can a configuration variable that comes from the original command
>> line be a boolean true that is spelled without "=true", i.e. can
>> value be NULL here?
>>
>
> Wouldn't it just be the empty string?

I was talking about the "not even an equal sign" true, i.e.

    $ git -c random.what -c random.false=no -c random.true=yes \
      config --bool --get-regexp 'random.*'
    random.what true
    random.false false
    random.true true

What would you see for random.what in the above function (if the
callchain allowed you to pass it through)?