![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
Hello again, Carlisle Adams wrote: > Hi Johan, > > > ---------- > > From: Johan Otterstrvm[SMTP:jo at sectra.se] > > Sent: Tuesday, May 05, 1998 11:16 AM > > To: cat > > Subject: Bad target names > > > > Hello, > > > > continuing my work with the IDUP-API I have a problem with the > "hiding" > > of the bad target names in the target info. It seems to me that a > more > > straighforward solution is to use a SET OF INTERNAL NAME as input > and > > SET OF Bad_Target_Name as output. Is there a reason to combine these > > > two? > > > The concept of parameter bundles has been used in this API simply to > tie > together parameters that are logically / conceptually related in some > way. > This is the case here: "target names" and "bad target names" don't > *have* > to be tied together (i.e., they could be separated as you suggest), > but they > seem so related that tying them together makes sense. For example, if > a > calling application wishes to sign a document (only) using IDUP, then > it > doesn't need the target names parameter. By extension, therefore, it > doesn't need the bad target names parameter either. > > Parameter bundles are not used to "hide" parameters; they are used to > "group" parameters. It seems to me that this particular grouping is > reasonable here. > The grouping make sence in that the bad target names are a subset of the target names but still the target names are input and bad target names are output. But as with your example there are other parameters that are not used for some cases, for example the pidu-buffer in clear-signing. If the grouping is a fact I would suggest that the parameter is in- and output. The target-info may also be changed to be a set of bad target name only (a new name would be prefered) were the names are used as input and the status as output for each name. It is not necessary to name the names twice. In any case the bad_targ_count is unnecessary since bad_target_names is a set. regards /jo -- ---------------------------------------------------------------- Johan Otterström Sectra AB Phone +46 13 23 52 46 Teknikringen 2 Telefax +46 13 21 21 85 S-583 30 Linköping Mail mailto:jo at sectra.se SWEDEN WWW http://www.sectra.se ----------------------------------------------------------------
Note Well: Messages sent to this mailing list are the opinions of the senders and do not imply endorsement by the IETF.