W3C home > Mailing lists > Public > public-xml-processing-model-wg@w3.org > August 2007

Re: a component with * inputs

From: Innovimax SARL <innovimax@gmail.com>
Date: Wed, 1 Aug 2007 19:21:15 +0200
Message-ID: <546c6c1c0708011021h276f2e3cr9d4e1888718e5f4@mail.gmail.com>
To: "Norman Walsh" <ndw@nwalsh.com>
Cc: public-xml-processing-model-wg@w3.org

On 8/1/07, Norman Walsh <ndw@nwalsh.com> wrote:
> / Innovimax SARL <innovimax@gmail.com> was heard to say:
> | On 8/1/07, Norman Walsh <ndw@nwalsh.com> wrote:
> |> / "Innovimax SARL" <innovimax@gmail.com> was heard to say:
> |> | and the result is
> |> |
> |> | <wrapper><doc><a/></doc><doc><A/></doc><alpha/></wrapper>
> |> | <wrapper><doc><b/></doc><doc><B/></doc><beta/></wrapper>
> |> | <wrapper><doc><c/></doc><doc><C/></doc><gamma/></wrapper>
> |>
> |> Ah, ok.
> |>
> |> |> And what's the value again?
> |> |
> |> | being able to give it to a for-each to simulate a p:equal-sequence
> |>
> |> And why not just iterate over the first sequence, using the
> |> p:iteration-count to exract the right element out of the other
> |> sequence?
> |
> | It is not efficient at all !!
>
> I never said it was.
>
> |> Is this really a common use case?
> |
> | Well, equal-sequence is the most obvious one
>
> Do you know of any actual use cases where you need to compare two
> sequences of documents?


>
> For comparing two sequences of documents, you could simply wrap each
> sequence in a "wrapper" and compare the two "wrapper" documents.

Good point, but I need to point which one is different, and I won't be
able to do it your way
Furthermore it will fail if the the first is different and never tells
me the rest are identical

>
> | The other one is when you have two flows of left pages and right pages
> | and what to merge them
>
> Do you know of any XML process that produces independent sequences of
> left and right pages?

Yep, bilingual books
I must admit that there is some syncho between both but for fast
snapshot you use some rough approximation
(The weird part are the foot notes)

>
> | More generally when you need to manipulate multiple flows you will use
> | something like this
>
> Indeed. I just don't know of any processes that produce such flows.
> I'm not saying that it couldn't be done, I just wonder if it's common
> enough to justify a standard specialty component.

In fact, this component would be perfectly symetric to how

<p:input port="source">
  <p:pipe port="seq1"/>
  <p:pipe port="seq2"/>
  <p:pipe port="seq3"/>
</p:input>

will work, but instead of consuming, the full seq1 first, then the
full seq2, you will be able to consume them in parallel


Mohamed

-- 
Innovimax SARL
Consulting, Training & XML Development
9, impasse des Orteaux
75020 Paris
Tel : +33 9 52 475787
Fax : +33 1 4356 1746
http://www.innovimax.fr
RCS Paris 488.018.631
SARL au capital de 10.000 
Received on Wednesday, 1 August 2007 17:21:43 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:21:53 GMT