DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
<http://nagoya.apache.org/bugzilla/show_bug.cgi?id=25803>.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://nagoya.apache.org/bugzilla/show_bug.cgi?id=25803

[PATCH] PropertyList: Fixes to convertAttributeToProperty and findBaseProperty





------- Additional Comments From [EMAIL PROTECTED]  2003-12-31 10:06 -------
I agree with you about the requirements. But the patch does not what you
describe. When the attribute leader-length.maximum is processed, the attribute
leader-length is retrieved from the list of attributes and processed first, out
of order. What I am ignoring is when the attribute is processed again, in the
order of the attributes.

I should add that the patch only works correctly when the issue in item 2 is
also addressed. Without it, the out-of-order processing fails and a default base
property is constructed. This would indeed result in the situation you depict,
with the specification of leader-length being ignored.

A more conservative patch would be to construct a member done in the
propertyList object, which is an array of booleans and keeps track of which
attributes have been processed, so that reprocessing of an attribute is
prevented. The logic of such a solution is more direct and less vulnerable to
logical errrors.

I think that in any solution, the base attribute must be processed before the
compound attribute, as is done now, and that reprocessing of the base attribute
must be prevented. Any method to prevent that is OK.

Reply via email to