W3C home > Mailing lists > Public > xproc-dev@w3.org > March 2010

RE: Storing a document mid sequence. Can it be done?

From: Christopher Ball <christopher.r.ball@gmail.com>
Date: Sat, 6 Mar 2010 11:16:42 -0500
To: "'Geert Bormans'" <geert@gbormans.telenet.be>, <xproc-dev@w3.org>
Message-ID: <003801cabd48$6acd60e0$958313ac@cgifederal.com>
Geert,

Very interesting - it would be great to see an example of your code.

It also makes me think of 'patterns' (common solutions to common problems)
and how it powerful it could be to have a place to post 'xproc patterns' - a
place to start identifying some common utilizations and giving them
intuitive names to help share ideas, understand some best practices, and get
the most from xproc. 

Anyone interested?

Christopher

-----Original Message-----
From: xproc-dev-request@w3.org [mailto:xproc-dev-request@w3.org] On Behalf
Of Geert Bormans
Sent: Thursday, March 04, 2010 4:07 PM
To: xproc-dev@w3.org
Subject: Re: Storing a document mid sequence. Can it be done?

Interesting,

Assuming you really need a "store", I won't be 
able to add too much more to what was said already.

I have a tendency though to name most of my steps.
Since most of my pipelines have multiple input ports and multiple output
ports
and some forking and joining in the middle, I 
often like to have debug points in the middle of the sequences.
For that I simply use extra output ports for 
debugging purposes, having a p:pipe referencing the named steps

This way I get my debug states without the extra cost of an identity step.
And I can make the results more readable by giving serialization
instructions
The Java around the processor (calabash in this 
case) binds the port in debugging mode and just drops it in production

Just in the process of learning myself, I wonder,
would that not be a more flexible approach over the using p:store?
Or do I miss something?

Cheers

Geert



At 15:09 4/03/2010, Alex Muir wrote:
>Thanks, got it working.. all too logical.
>
>On Thu, Mar 4, 2010 at 1:53 PM, 
><<mailto:Toman_Vojtech@emc.com>Toman_Vojtech@emc.com> wrote:
>
>Absolutely:
>
>
>
>.
>
><p:identity name="before-store"/>
>
>
>
><p:store href="."/>
>
>
>
><p:identity name="after-store"/>
>
>   <p:input port="source">
>
>     <p:pipe step="before-store" port="result"/>
>
>   </p:input>
>
></p:identity>
>
>.
>
>
>
>
>
>Regards,
>
>Vojtech
>
>
>
>
>
>From: 
><mailto:xproc-dev-request@w3.org>xproc-dev-request@w3.org 
>[mailto:xproc-dev-request@w3.org] On Behalf Of Alex Muir
>Sent: Thursday, March 04, 2010 2:49 PM
>To: XProc Dev
>Subject: Storing a document mid sequence. Can it be done?
>
>
>
>Hi,
>
>I was reading a post here
>
><http://markmail.org/search/?q=list%3Axproc-dev+p%3Astore#query:list%3Axpro
c-dev%20p%3Astore+page:1+mid:xzoj5hpzbz5jt2k2+state:results>http://markmail.
org/search/?q=list%3Axproc-dev+p%3Astore#query:list%3Axproc-dev%20p%3Astore+
page:1+mid:xzoj5hpzbz5jt2k2+state:results
>
>That stated:
>
>"If you stick a p:store in the middle of a 
>sequence, odds are good you aren't going to get 
>the result you expected. So by making the output 
>port not-primary, the pipeline author is forced 
>to be explicit about the bindings desired."
>
>Is there a way to save a copy of what the 
>current document looks like mid sequence and continue processing?
>
>Regards
>--
>Alex
><https://sites.google.com/a/utg.edu.gm/alex>https://sites.google.com/a/utg.
edu.gm/alex
>
>Some Good Music
><http://sites.google.com/site/greigconteh/>http://sites.google.com/site/gre
igconteh/
>
>
>
>
>--
>Alex
><https://sites.google.com/a/utg.edu.gm/alex>https://sites.google.com/a/utg.
edu.gm/alex
>
>Some Good Music
><http://sites.google.com/site/greigconteh/>http://sites.google.com/site/gre
igconteh/
Received on Monday, 8 March 2010 13:24:13 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 8 March 2010 13:24:14 GMT