Re: Call for contributions: test suite

From: Lorenzo <>
Date: Mon, 30 Mar 2015 17:14:19 +0200

On 03/30/2015 04:00 PM, Laurent Bercot wrote:
> On 30/03/2015 00:51, Lorenzo wrote:
>> A part of me still wishes I never sent that email, you know... ;p
> Why ? Is there anything wrong with knowledge of test suites ?
> If it's just that you hate testing as much as I do, then relax,
> you can know forget everything you just learned about it, if you
> so desire. :P
>> TLDR: I vote for tap - who said that nothing good came out of perl?
> Thanks a lot for the pointers and analysis. I really appreciate it.
> At first sight, TAP looks good; I'll have a deeper look into it
> (and other candidates) when I get the time and the courage.
Getting started is the worst part.
Make it as painless as possible to write new tests - ideally, just write
a C file and let make (or a script) notice it.

If you end up choosing tap, remember to make it easy to get a list of
all tests, either t/*.t or an autogenerated file listing them all.
Preferably a filelist, because while t/*.t is obvious, it's built around
the assumption of a scripting language - should "make clean" remove
them? what if you have a mix of compiled binaries and handwritten shell
(or you could do something like what you already do with src/ vs compile/)
> (You have to treat perl, and other similar software, as you would
> manure. Doing things directly with it stinks and will only get you
> dirty hands; but it's excellent fertilizer that can help grow
> beautiful ideas.)
"Dai diamanti non nasce niente, dal letame nascono i fiori" ;)

It's very light on details, but sounds encouraging - especially since
you were targeting raspberry pi, which I've just learned is more powerful.
Received on Mon Mar 30 2015 - 15:14:19 UTC

This archive was generated by hypermail 2.3.0 : Sun May 09 2021 - 19:38:49 UTC