[prev in list] [next in list] [prev in thread] [next in thread] 

List:       gump
Subject:    Re: svn commit: r164483 -
From:       Leo Simons <mail () leosimons ! com>
Date:       2005-04-25 7:57:28
Message-ID: BE926F88.26A7C%mail () leosimons ! com
[Download RAW message or body]

On 25-04-2005 00:31, "Adam R. B. Jack" <ajack@apache.org> wrote:
> Leo wrote:
> 
>> * pygump/python/gump/plugins/builder.py: commenting out this existence
> test makes the associated testcase fail under posix. Error log:
> 
> Sorry, I should've run/fixed the test case.

Hey -- no worries!

> I suspect that failure to find a
> script file is not more severe an error than failure to execute one, or it
> failing to execute, so (1) I wonder why we special case it and (2) I don't
> think we ought exit the plugin on that eventuality. Surely we ought continue
> and set appropriate properties, with failure to run just being one. No?

Well, failure to find a script file is a problem in the gump metadata
(having a <script/> tag in there to me is stating that said script file
exists), whereas a failure to complete the script successfully is likely a
problem with the script.

My take is that plugins should only handle that which they fully understand
and let everything else bubble up. Don't make it look intelligent if it
really is not. Any stuff that you would set in properties you can store in
an exception as well.

These lines really said something like "Okay, I know there's a problem but I
don't know what caused it. Whatever. I give up. You handle it". Making a lot
of your code lazy and naïve is a good way to not end up with lots of
redundant code.

WDYT?

(...)

However, all that is just blah blah I'm making up right now; I didn't think
about that when I made this change this change. TDD requires that the tests
run, they didn't, so I fixed that in, for me, the easiest way possible
(which is an XP requirement :-)). If there is something wrong with the code
(ie like you're proposing it should be recovering gracefully), fix the
tests, and fix 'em first!

Cheers,

LSD



---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@gump.apache.org
For additional commands, e-mail: general-help@gump.apache.org


[prev in list] [next in list] [prev in thread] [next in thread]