On 4/23/06, P T Withington <[EMAIL PROTECTED]> wrote:
I think it is something more obscure. The switch certainly looks
like it will not process the library, yet we get all sorts of
warnings about things in the library. Oh! Maybe when the switch
asks the xml parser to get the children, it tries to greedily parse
the library, even though it will be ignored, and it barfs because
none of the auto-included libraries are in the schema yet.
Is there some way to ask the xml parser to give me a child node
without it validating the contents of that node?
On 2006-04-23, at 19:14 EDT, Henry Minsky wrote:
> Some sort of autoinclude failure maybe?
>
> On 4/23/06, P T Withington <[EMAIL PROTECTED]> wrote:
>>
>> On 2006-04-23, at 14:55 EDT, P T Withington wrote:
>>
>>> components/debugger/library: use <switch> tag to not include
>>> component debugger for DHTML.
>>
>> This doesn't actually work. I was wondering if you (Henry) had any
>> insight.
>>
>> When I write the switch thusly:
>>
>> <library>
>> <switch>
>> <when runtime="dhtml">
>> </when>
>> <otherwise>
>> <include href="">>> </otherwise>
>> </switch>
>> </library>
>>
>> The compiler issues all sorts of complaints
>>
>>> 21 compilation warnings:
>>>
>>> debugger/debugger.lzx:74:31: element "resizelayout" not allowed in
>>> this context. Check whether it is spelled correctly, and whether a
>>> class with this name exists.
>>> debugger/debugger.lzx:87:36: element "stableborderlayout" not
>>> allowed in this context. Check whether it is spelled correctly, and
>>> whether a class with this name exists.
>>> debugger/debugger.lzx:460:30: element "dragstate" not allowed in
>>> this context. Check whether it is spelled correctly, and whether a
>>> class with this name exists.
>>
>> (&c.)
>>
>> I suspect there is something going on where the parser is parsing the
>> debugger file before the switch can decide not to compile it.
>> Perhaps the switch 'compilation' needs to actually happen at parse
>> time.
>>
>> The question remains whether we want or need to support this... it is
>> a primitive requires/provides, so seemed like a logical place to
>> start.
>> _______________________________________________
>> Laszlo-dev mailing list
>> [email protected]
>> http://www.openlaszlo.org/mailman/listinfo/laszlo-dev
>>
>
>
>
> --
> Henry Minsky
> Software Architect
> [EMAIL PROTECTED]
--
Henry Minsky
Software Architect
[EMAIL PROTECTED]
_______________________________________________ Laszlo-dev mailing list [email protected] http://www.openlaszlo.org/mailman/listinfo/laszlo-dev
