Re: Detecting unbound options

2009/5/27 Norman Walsh <>:
> "Henry S. Thompson" <> writes:
>> Gee, I'm still not happy with that idea.  There are lots of cases
>> where not specifying a value is importantly different from specifying
>> the empty string.

Again learning from Apache ant.
I test for both unspecified and no value specified.
Note, I'm talking about variables here.
Please disregard if you're only thinking of options/params etc.

> I don't think all of those are problematic. Options that aren't
> strings (undeclare-prefixes) or are QNames (initial-mode, template-name)
> or that don't have a meaningful empty-string value (doctype-public,
> media-type) are all ok.
> But you're right that it would be a problem for some (base-uri, from,
> to, ...)
> Yuck! But can we really live with this...

What of your file utilities?
create file ${varName}

would leave me with ??? What? A file named ${varName} ?

How about a configuration option to report unspecified variables?
I'd certainly like to have xproc fall over 'variable XYZ unspecified'.


Dave Pawson
Docbook FAQ.

Received on Wednesday, 27 May 2009 12:43:25 UTC