W3C home > Mailing lists > Public > xproc-dev@w3.org > October 2013

Re: No more parameter ports

From: David Cramer <david@thingbag.net>
Date: Tue, 08 Oct 2013 11:41:50 -0500
Message-ID: <525435CE.10502@thingbag.net>
To: Norman Walsh <ndw@nwalsh.com>, XProc Dev <xproc-dev@w3.org>
Hash: SHA1

On 10/08/2013 11:15 AM, Norman Walsh wrote:
> David Cramer <david@thingbag.net> writes:
>> Would you have to specifiy every parameter in the pipeline that
>> could be passed in to the step? Would you also have to specify a
>> defaut value for it? Thinking of the DocBook xslts, which have
>> many params with sensible defaults set, you don't want to HAVE to
>> repeat the full list of allowed params and their defaults in the
>> xpl. I suppose you could generate part of your pipeline from your
>> xslts, but that's just moving the work and complexity to another
>> place.
> If we do nothing else, then yes, it means you'd have to name them
> all. I think we might be able to manage a little something, but the
> heart of the proposal is "drop this feature on the floor".

Ok, then there are to things that are a drag about that:
1. That you have to declare the params at all.
2. That you have to provide a value to pass in.

#2 seems to be the worse of the two. What if you have in your xslt:

<xsl:param name="blah">
  <!-- some logic here that only happens if nothing is passed in -->

Now you have to move that logic into the xpl as well? Or refactor the
xslt so that you take in a dummy param and then test the dummy param
and do the logic if it's not the default value?

I used the DocBook xslts as an example, but I don't think it's fair to
say "the DocBook xslts are unusually complex". This is likely to come
up especially if you're using 3rd party xslts and don't want to modify
them so you can call them from XProc.

Version: GnuPG v1.4.12 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

Received on Tuesday, 8 October 2013 16:42:20 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:03:11 UTC