Re: [PATCH v2 2/5] doc: git-diff: apply format changes to diff-options

From: Jean-Noël Avila <jn.avila@free.fr>
Date: Nov 12, 2024, 09:04 UTC
Message-ID: <b0a894fc-911d-49f5-a10c-19d441dc7133@free.fr>
In-Reply-To: <xmqqh68d8e9h.fsf@gitster.g>

Le 12/11/2024 à 01:52, Junio C Hamano a écrit :

Show 8 quoted lines

> "Jean-Noël Avila via GitGitGadget" <gitgitgadget@gmail.com> writes:
>  
>> @@ -39,28 +39,28 @@ endif::git-format-patch[]
>>  ifdef::git-log[]
>>  -m::
> 
> Shouldn't this and all others like -c/--cc be also quoted as
> `literal` options?

As stated in the commit message, only the parts of the files which appear in git-diff man page are converted (like was done for git-clone and git-init)

Thinking again about it, I don't find it wise, because other man pages will be half-converted anyway, at least for the common parts of the included files, and we are going to introduce several commits for the same file.

So, better convert all the file in on run.

Show 6 quoted lines

> 
>> @@ -73,33 +73,33 @@ The following formats are supported:
>>  off, none::
> 
> Shouldn't this and other option values like on, first-parent, etc.,
> that are literals be marked-up specially?

This is to be converted, and will be with the conversion of the full file.

Show 9 quoted lines

> 
>> --U<n>::
>> ---unified=<n>::
>> -	Generate diffs with <n> lines of context instead of
>> +`-U<n>`::
>> +`--unified=<n>`::
>> +	Generate diffs with _<n>_ lines of context instead of
> 
> Understandable.  Shouldn't <n> part be italicised?

No need: the processing of back tick quotes already treats each part according to its semantics and applies the corresponding format.