[
Lists Home |
Date Index |
Thread Index
]
three questions before i leave you to formulate the gory details:
you wrote:
>
>
> Structurally, the three part morphology
>
> xmlmap : n1 n2
> n1 : URI
> n2 : e1 l1 e2 l2 ....
>
> is an association of the URI with the list (interpretable as a hash), like
> in Perl:
>
> { ...
> URI => { e1 => l1
> e2 => l2
> ...
> }
> }
>
1. how to encode the mapping specification?
what is the advantage of
resolver == URI -> local-name -> external-name
over
resolver == local-name -> (URI x external-name)
since the local name must be unique in any given context. this would suggest an alternative morphology, and thereby, encoding:
xmlmap : n1 n2
n1 : (URI1 e1 l1) (URI2 e2 l2)
whereby the URIs may be identical
2. ontology, taxonomy, or name set
> |>| please do it in advance. please do not repeat the sins of "namespaces".
> |>
> |>I have no idea what you're talking about.
> |
> | the uncertainty as to whether what matters is the set of names or where
> | they come from and or are used.
>
> The sins of Namespaces are on the heads of its (their?) advocates. It
> always matters where the names we use *in processing* came from. Names
> actually occuring in markup can be arbitrary. They are merely a scaffold
> for the arch.
why does "where the names came from" matter for a renaming transformation?
3.
> |>| does the "element" scope entail contained elements?
> |>
> |> I "guess" so, but then again, maybe not. Seriously, I have no idea why
> |> your notion of "scope" matters to you.
> |
> | it matters because, to the extent that XMap is intended to subsume
> | namespaces in XML, under certain conditions it will behave the same,
> | which means that while performing the transformation which XMap
> | describes, the pairs in the control attributes are equivalent to variable
> | bindings and the transformation itself is equivalent to beta-conversion.
>
> I don't see how. I'd say XMap is an analog of alpha renaming.
it is alpha-conversion only if one posits an implicit binding for each name. an implementation looks a lot more like beta-conversion.
taking the analog to the original example expression, reformulated in the alternative encoding:
<img src="some.gif"
xmlmap="yy"
yy="{http://www.w3.org/1999/xlink}href {}src
{http://www.w3.org/1999/xlink}type {}foo
{http://www.w3.org/1999/xlink}role {}bar
{http://www.w3.org/1999/xlink}title {}quux"
foo="simple"
bar="http://example.org/some/role";
quux="Sleeping Kitty"
>
this can be represented with a model which has the following, simple, readable abstract syntax as
? (defParameter *form_* '(({}img ({}src . "some.gif")
({}xmlmap . "{}yy")
({}yy . "{http://www.w3.org/1999/xlink}href {}src
{http://www.w3.org/1999/xlink}type {}foo
{http://www.w3.org/1999/xlink}role {}bar
{http://www.w3.org/1999/xlink}title {}quux")
({}foo . "simple")
({}bar . "http://example.org/some/role";)
({}quux . "Sleeping Kitty"))
"some content or other"))
*FORM_*
;; which permits the following definition for XMap, (i hope this isn't wrapping too badly, and
;; neglect the special case for GI mapping and the question of scoping)
? (defGeneric XMap (form)
(:method ((form list))
(labels ((read-names (string)
(let ((names nil) (name nil))
(with-input-from-string (stream string)
(loop (unless (setf name (read stream nil nil)) (return (nreverse names)))
(push name names)))))
(make-alist (mapping)
(when mapping (acons (second mapping) (first mapping) (make-alist (cddr mapping))))))
(xmla:with-abstract-element ((gi &rest attributes) &rest content) form
(xmla:with-abstract-attributes ((xmlmap-attr {}xmlmap)) attributes
;; collect the mappings from the attributes
(let ((mappings
(make-alist
(apply #'append (mapcar #'read-names
(mapcar #'(lambda (attr-name)
(xmla:get-abstract-property-value attributes attr-name))
(read-names xmlmap-attr)))))))
(setf form (apply #'xmla:make-element gi (copy-alist attributes) (mapcar #'XMap content)))
(if mappings ;; perform the transformation if a mapping spec is present
(nsublis mappings form)
form))))))
(:method ((form t)) form))
#<STANDARD-GENERIC-FUNCTION XMAP #x63EAE96>
;; which, when applied to the example form, yields
? (pprint (xmap *form_*))
(({}img ({http://www.w3.org/1999/xlink}href . "some.gif") ({}xmlmap . "{}yy")
({}yy . "{http://www.w3.org/1999/xlink}href {}src
{http://www.w3.org/1999/xlink}type {}foo
{http://www.w3.org/1999/xlink}role {}bar
{http://www.w3.org/1999/xlink}title {}quux")
({http://www.w3.org/1999/xlink}type . "simple") ({http://www.w3.org/1999/xlink}role . "http://example.org/some/role";)
({http://www.w3.org/1999/xlink}title . "Sleeping Kitty"))
"some content or other")
;; or alternatively, reserialized
? (write-node (xmap *form_*) *trace-output*)
<img nsp-1:href='some.gif' xmlmap='{}yy' yy='{http://www.w3.org/1999/xlink}href {}src
{http://www.w3.org/1999/xlink}type {}foo
{http://www.w3.org/1999/xlink}role {}bar
{http://www.w3.org/1999/xlink}title {}quux' nsp-1:type='simple'
nsp-1:role='http://example.org/some/role'
nsp-1:title='Sleeping Kitty' xmlns:nsp-1='http://www.w3.org/1999/xlink'>some content or other</img>
NIL
...
|
