Re: [PATCH 1/2] help: introduce option --command-only

From: Junio C Hamano <gitster@pobox.com>
Date: Aug 19, 2016, 15:53 UTC
Message-ID: <xmqqinuw4uww.fsf@gitster.mtv.corp.google.com>
In-Reply-To: <alpine.DEB.2.20.1608190954461.4924@virtualbox>

Johannes Schindelin <Johannes.Schindelin@gmx.de> writes:

> - configure help.format = html (for "man", the current code would always
>   add $(prefix)/share/man to the MANPATH when testing, not what we want,
>   and hacking this code *just* for testing is both ugly and unnecessary).

A very constructive suggestion to show a good direction. Because we are not in the business of verifying the "man" works as expected and how it wants the files in MANPATH are structured, it is a very good idea to do our testing with the HTML format, which gives us tighter control of what we actually use for our testing with help.browser configuration variable. I really like it.

> - configure help.htmlpath to point to a subdirectory that is created and
>   populated in the same test script.

Yup!

> - configure help.browser to point to a script that is created in the same
>   script and whose output we can verify, too.

Yup, yup! The "browser" can be something that parrots its command line to the standard output and does not even have to care if the file pointed at is HTML at all.

> The last point actually requires a patch that was recently introduced into
> Git for Windows [*1*] (and that did not make it upstream yet) which
> reverts that change whereby web--browse was sidestepped. That sidestepping
> was well-intentioned but turned out to cause more harm than good.

Good.