Re: Regression testing for org-mode

From: "Eric Schulte" <schulte.eric@gmail.com>
To: Robert Goldman <rpgoldman@sift.info>
Cc: emacs-orgmode@gnu.org
Subject: Re: Regression testing for org-mode
Date: Thu, 23 Oct 2008 16:57:22 -0700	[thread overview]
Message-ID: <87wsfyx3tp.fsf@gmail.com> (raw)
In-Reply-To: <4900A06B.8020605@sift.info> (Robert Goldman's message of "Thu, 23 Oct 2008 11:03:55 -0500")

I've added a "testing tools" section and a simple example of an org-mode
test using ert to Sebastian's org testing page in worg.  I don't mean to
advocate for ert, it is simply the only elisp testing framework that I
have any experience with.

I'm pessimistic that the colossal amount of work involved in wrapping
all of org-modes extensive and varied functionality into a test
framework can/should actually be completed.  Would it be sufficient to
simply begin dealing with bug reports in a test-first manner?

Also, should we start tracking bug reports somewhere (worg), so that
they can be claimed, tested against, and repaired?

Thanks -- Eric

Robert Goldman <rpgoldman@sift.info> writes:

> There's a page on the EmacsWiki describing schemes for unit/regression
> testing in emacs lisp:
>
> http://www.emacswiki.org/cgi-bin/wiki/UnitTesting
>
> Perhaps we could pick up one of these?  I haven't evaluated any of them.
>
> In response to Ben Alexander's question, yes, sometimes the screen
> output is what you care about, but a big advantage of emacs is that the
> screen output is a big array of text, so we could check it (it's a lot
> worse, if it's just a bit map...).  Sure, we can't always do that, but I
> think in many cases we could just grab up a buffer's contents and
> compare it against a "right answer."
>
> best,
> r
>
>
>
> _______________________________________________
> Emacs-orgmode mailing list
> Remember: use `Reply All' to send replies to the list.
> Emacs-orgmode@gnu.org
> http://lists.gnu.org/mailman/listinfo/emacs-orgmode