| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 1665.1 | You might try | IOSG::SHOVE | Dave Shove -- REO-D/3C | Tue Oct 27 1992 11:08 | 17 |
| You're right. For performance reasons, ALL-IN-1 only reads the values
of logicals once, during initialisation.
You _may_ be able to get it to re-read the value of oa$dds_prime by
clong down the mail subsystem and re-initialising it:
MAIL CLOSE_SYSTEM
MAIL INIT
I haven't tried this; without looking at the code I can't tell you if
it'll work.
Someone might be able to suggest something else (the v3.0 changes to
this area are extensive and deep in the code - there's no way you could
retrofit them to v2.4).
Dave.
|
| 1665.2 | Or doesn't this include OA$DDS_PRIME? | SCOTTC::MARSHALL | | Tue Oct 27 1992 11:45 | 3 |
| How about using OA$INI_LOGICALS to re-read logicals?
Scott
|
| 1665.3 | | EWBV51::NAGURA | Yukiji Nagura /ALL-IN-1 CSC/Tokyo/Japan | Wed Oct 28 1992 02:18 | 10 |
|
You can utilize an undocumented special symbol OA$DDS_SEARCH_SCOPE,
though I myself haven't used it. It is somewhere in the named data
of DIR SMD releated forms, and it can take a value from 0 to 2.
0 - Local Search Scope, 1 - World Search Scope, 2 - Escalated Search
I'm not so sure about its functionality, but it's a writable one.
Yukiji
|
| 1665.4 | I know a little about its functionality | AIMTEC::WICKS_A | Liverpool 4 Norwich 1 | Wed Oct 28 1992 04:44 | 22 |
| Yukiji,
Oh no you can't!!!!!!!!!!! (:==:)
It's set when you do a MAIL INITIALIZE and is then stored internally
to the SUBSCRIBER DSAB. You might think you're changing it but trust me
that the DSAB is far too smart to be faked out by some application
programmer writing to it.
Also the Escalated Search code never got finished, and still isn't
implemented in v3.0 so that doesn't work either.
I'll give you a clue you need to do something with OA$DDS_NODE_TYPE
to even get 0 and 1 to behave themselves and that's a readable symbol
only!
sorry nice try - but it's undocumented for a good reason
regards,
Andrew.D.Wicks (Not the DDS developer anymore)
|
| 1665.5 | I've tried the above and... | TAV02::SHAPIRA | | Wed Oct 28 1992 07:20 | 22 |
| Thanks for all the quick replys.
rep. 1:
Dave - MAIL CLOSE_SYSTEM ??? Is that an undocumented function ? What does that
function do ?
MAIL INIT - Didn't help to set os$dds_prime.
rep. 2:
Scott - We have tried the funcion oa$ini_logicals, but it seemed that it does
nothing. Even when we traced the function we couldn't see what it is
doing. Is it a regular function (just like oa$ini_init) ?
rep. 3 and 4:
Since Andrew suggested to forget about oa$dds_search_scope, I've forgot it !!
Thanks,
Yariv
|
| 1665.6 | | SCOTTC::MARSHALL | | Wed Oct 28 1992 08:33 | 9 |
| Hi,
>> Is it a regular function (just like oa$ini_init) ?
Yes, it's in the documentation on the page after oa$ini_init. However, I've
never looked at the code so I don't know whether it does what the book claims
it does! :-)
Scott
|
| 1665.7 | CLOSE_SYSTEM doesn't work | AIMTEC::WICKS_A | Liverpool 4 Norwich 1 | Wed Oct 28 1992 16:39 | 11 |
| For people still at v2.4 both MAIL CLOSE_SYSTEM and MAIL INIT are
undocumented.
For those at v3.0 MAIL INIT is now documented but CLOSE_SYSTEM isn't
It's supposed to undo all the work that MAIL INIT did but looking at
the code it doesn't look to have been updated since oh about 1986 so
It's no wonder it doesn't undo the work of MAIL INIT
Regards,
Andrew.D.Wicks
|
| 1665.8 | Waiting for V3.0 migration | TAV02::SHAPIRA | | Sun Nov 01 1992 06:35 | 18 |
|
Hi Andrew,
The customer will use either System Nicknames or NETWORK.DAT until they will
migrate to V3.0. In 3.0 they will use DDS (hopefully...). The performance of
the DDS search was VERY poor in compare to the other methods.
As for MAIL CLOSE SYSTEM to be updated to 1986 - that's probably explains why
it did nothing...
"Liverpool 4 Norwich 1" (taken out from your header) - are you a Liverpool
supporter ? Regards from an old time Manchester City supporter. Any way, you
will probably never walk alone..
Thanks for everybody's quick help,
Yariv
>
|