http://www.w3.org/Bugs/Public/show_bug.cgi?id=1705 ------- Comment #10 from jmdyck@ibiblio.org 2006-10-05 01:02 ------- (In reply to comment #6) > > The working groups have decided to adopt the following resolution for it: > > (1) adopt your suggested two phase processing for the prolog which splits > export context processing from static context processing. > (2) solves the issue with static analysis of global variables without > a type declaration by relying on the existing constraint in XQuery which > disallow circularity in variable definitions These two points are somewhat contradictory. If [err:XQST0054] is the only safeguard against circularity, it sounds like ECP of a type-decl-less variable decl involves a full STA of the initializing expression, which requires the appropriate static environment, which normally results from SCP. So how can you separate ECP from SCP?Received on Thursday, 5 October 2006 01:02:18 GMT
This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:14:17 GMT