: If neither of these is the case, then we still might copy the value to one
: or more other fields due to a wildcard match in a copyField.

Darren, glancing at your patch, I see you added a simple test for copying
from *_t to text -- but *_t is an existing dynamicField .. could you
perhaps add some more test cases covering these new situations you've
added support...

   <copyField source="*_notADeclaredDynamicField" dest="text"/>
   <copyField source="notADeclaredField"          dest="text"/>


Also; the paranoid part of my brain wonders if the <copyField> directive
should be left as it is, and a new <dynamicCopyField> directive should be
added to support wildcards in the source ... i worry that some weird sick
individual out there might acctually want a field name that contains a "*"
in it, and we've not eliminated the ability for them to use that field
with a copyField

(this was the same motivation for the <field> <dynamicField> distinction)

: The syntax you described below, where the destination contains a wildcard,
: is not supported by this implementation.  The destination must be a an
: explicit field, meeting the conditions above.

: If I have a <copyField source="*_a" dest="*_b"/>

I'm acctually in favor of this approach .. Yonik's idea once upon a time
of supporting <copyField source="*_a" dest="*_b"/> allways scared me, not
only does it raise hte question:

  is that legal as schema creation even if i don't have a <dynamicField
  name="*_b" ... /> but i do have a lot of filds whose names end with _b?

but it opens the door for <copyField source="*_a" dest="b_*"/> ... how
should that be treated if in addition to a dynamicField with the prefix
b_* i also have an explicitly field named "b_foo" .. normal matching rules
say that if i add a doc with a field named "b_foo" it's subject to the
FieldType of the explicitly declared field, but is that the "right" thing
to do when we're in a copyField relationship where the source was dynamic?

...like i said, i'm paranoid.



-Hoss

Reply via email to