Use warnings

[openstrike]

This is a trivial PR to use warnings in the module as recommended by CPANTS. There doesn't appear to be any particular reason not to use warnings for this module and it passes tests cleanly (and quietly) when warnings are used on my test systems.

[karenetheridge]

I looked at adding the use of warnings a while back, but it would be really risky to introduce new warnings to production systems that are not expecting it -- this can fill up logs quickly.

[openstrike]

The reticence is understandable and I agree with it to a certain extent.

However, as a user I would be inclined towards the use of warnings as by their very nature they should serve to highlight potential problems or unexpected scenarios in either my code or within the module which could then be addressed. If we had an extensive release test (yes, I might be volunteering here) which would cover as many of the combinations of inputs as possible and thereby determine that normal operation were silent even when warnings are used would that be a sufficient enabler for using warnings?

[karenetheridge]

Yes, potentially :)

On Fri, Aug 21, 2015 at 3:44 AM, Pete Houston notifications@github.com
wrote:

The reticence is understandable and I agree with it to a certain extent.

However, as a user I would be inclined towards the use of warnings as by
their very nature they should serve to highlight potential problems or
unexpected scenarios in either my code or within the module which could
then be addressed. If we had an extensive release test (yes, I might be
volunteering here) which would cover as many of the combinations of inputs
as possible and thereby determine that normal operation were silent even
when warnings are used would that be a sufficient enabler for using
warnings?

—
Reply to this email directly or view it on GitHub
#25 (comment)
.

[oalders]

Reworked this in #69. Thanks, @openstrike!