Hi Duy, On Fri, 4 May 2018, Duy Nguyen wrote:
> On Thu, May 3, 2018 at 10:32 PM, Johannes Schindelin > <johannes.schinde...@gmx.de> wrote: > > > > On Thu, 3 May 2018, Johannes Schindelin wrote: > > > >> On Thu, 3 May 2018, Duy Nguyen wrote: > >> > >> > On Thu, May 3, 2018 at 5:30 PM, Johannes Schindelin > >> > <johannes.schinde...@gmx.de> wrote: > >> > > diff --git a/command-list.txt b/command-list.txt > >> > > index a1fad28fd82..c89ac8f417f 100644 > >> > > --- a/command-list.txt > >> > > +++ b/command-list.txt > >> > > @@ -19,6 +19,7 @@ git-archive mainporcelain > >> > > git-bisect mainporcelain info > >> > > git-blame ancillaryinterrogators > >> > > git-branch mainporcelain > >> > > history > >> > > +git-branch-diff mainporcelain info > >> > > >> > Making it part of "git help" with the info keywords at this stage may > >> > be premature. "git help" is about _common_ commands and we don't know > >> > (yet) how popular this will be. > >> > >> Makes sense. I removed the `mainporcelain` keyword locally. > > > > On second thought, I *think* you meant to imply that I should remove that > > line altogether. Will do that now. > > Actually I only suggested to remove the last word "info". That was > what made this command "common". Classifying all commands in this file > is definitely a good thing, and I think mainporcelain is the right > choice. Oh, okay. It was not at all clear to me what the exact format and role of these lines are... So that's what `info` does: it influences whether/where the command is listed in `git help`'s output... Interesting. I thought the lines here were trying to automate parts of the tab completion or something. I re-added the line, this time without `info` and verified that `branch-diff` does not show up in `git help`'s output. Ciao, Dscho
