[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: i115 side-note; possible editorial issue
Lines 545-547 of the very lovely http://www.oasis-open.org/apps/org/workgroup/ws-rx/download.php/17836/wsrm-1.1-spec-wd-12.pdf state:
If a
non-mustUnderstand fault occurs when
processing an RM Header that was piggy-backed on another message, a fault
MUST be generated, but the processing of
the original message MUST NOT be affected.
Lines 575-577 have the exact same text.
Firstly, I'm not sure if the mention of the mustUnderstand fault
is relevant. If a message carrying a piggy-backed RM Header contained
another header with the soap:mU attribute and that header wasn't supported by
that SOAP node (I'm assuming that it isn't the RM Header that is marked mU), it
doesn't seem that anything anyone did or didn't do at the RM layer would make
the tiniest bit of difference. The SOAP processor is going to generate a fault
and the RM layer will never see the message. It seems like we are directing RM
implementers to worry about a case that isn't their problem.
Secondly, if the part about "a non-mustUnderstand fault"
is relevant, the text only says what
should happen in one of the cases. If the RM layer were involved in the
handling a mustUnderstand fault on a message containing a piggy-backed RM Header
what is it supposed to do? One guess would be to process the fault "normally"
(stop processing and return a fault). Another, equally valid, guess would be to
drop the fault on the floor and continue processing the
message.
-
gp
|
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]