[
Lists Home |
Date Index |
Thread Index
]
> I thought namespaces were a pain to implement.
>
> I have a question about "Validity constraint: Proper Declaration/PE
> Nesting" [1] I understand that this is saying the following is violates
> the VC:
>
> Doc1.dtd
> ========
> <!ENTITY % e "<!-- a ">
> <!ENTITY % f "%e;comment -->">
>
>
> Doc2.dtd
> ========
> <!ENTITY % e "<!-- a ">
I agree.
>
> But what about this:
>
> Doc3.dtd
> ========
> <!ENTITY % e "<foo>">
> <!ENTITY % f "%e;</foo>">
> <!ENTITY g "%g;">
>
> I think that it is legal but who knows anymore. As near as I can figure
> it doesn't violate this VC because it is not a markup decl specifically.
> Also, it is not a violation of "Validity constraint: Proper Group/PE
> Nesting" [2]. I assume it is valid because of this text in 4.3.2 "An
> internal general parsed entity is well-formed if its replacement text
> matches the production labeled content. All internal parameter entities
> are well-formed by definition."
Yes, I agree.
> Bonus questions: Which wins out: "Well-formedness constraint: In DTD"
> [3] or "Not Recognized" [4]
>
> [1] http://www.w3.org/TR/REC-xml/#vc-PEinMarkupDecl
> [2] http://www.w3.org/TR/REC-xml/#vc-PEinGroup
> [3] http://www.w3.org/TR/REC-xml/#indtd
> [4] http://www.w3.org/TR/REC-xml/#not-recognized
I have always found "Well-formedness constraint: In DTD" redundant,
as due to "Not Recognized", you cannot define PE references outside of the DTD.
So, "Not Recognized" wins, and the above entity declaration is legal.
Karl
|
