| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 3858.1 | No Starlet bindings upgrades planed; yet... | KMOOSE::CMCCUTCHEON | Charlie McCutcheon | Wed Apr 09 1997 10:51 | 12 |
| We are not currently planning any updates to our Starlet bindings
for VMS 7.1. That could change, if we get enough requests.
The user can make their own bindings, assuming they can determine what
they need, and compile them into a local library, or directly into the
Ada Predefined library. Note that we won't support these. ;-)
Note that we do not specifically support VMS 7.1 yet. If you find any
7.1 only problems, we are interested in them for a future release, but
will not give the same sort of support as for VMS 6.1-7.0 on Alpha VMS.
Charlie
|
| 3858.2 | customer is asking for a commitment of some kind | CUJO::SAMPSON | | Wed Apr 09 1997 22:28 | 11 |
| Charlie,
I guess it's a lot to ask, since (I presume) SDL does not
generate Ada, but how does the future look for having the compiler
installation procedure extract Ada STARLET definitions from the
target operating system, or perhaps having the target operating
system supply the Ada STARLET definitions? This would be more in
keeping with some of the other compilers.
Thanks for listening,
Bob Sampson
|
| 3858.3 | | KMOOSE::CMCCUTCHEON | Charlie McCutcheon | Thu Apr 10 1997 10:42 | 11 |
| SDL does generate Ada, but we've found that we need to "edit" it to make the output
useful. That's a fair amount of tedious work (I've done it before).
I don't think we can commit to doing a full upgrade of all bindings to 7.1 at this
time (since know what we have on our plate for the next release), but perhaps we can
look at the bindings your user is concerned about. Who is your customer? ;-)
(Also, if the list of routines is short, I could probably run SDL on the V7.1 Starlet
bindings out currently, and give a UNsupported copy to you as a "guideline"...)
Charlie
|
| 3858.4 | .-1 edited to fit 80 columns | CUJO::SAMPSON | | Fri Apr 11 1997 00:02 | 15 |
| <<< Note 3858.3 by KMOOSE::CMCCUTCHEON "Charlie McCutcheon" >>>
SDL does generate Ada, but we've found that we need to "edit" it to make the
output useful. That's a fair amount of tedious work (I've done it before).
I don't think we can commit to doing a full upgrade of all bindings to 7.1
at this time (since know what we have on our plate for the next release),
but perhaps we can look at the bindings your user is concerned about.
Who is your customer? ;-)
(Also, if the list of routines is short, I could probably run SDL on the
V7.1 Starlet bindings out currently, and give a UNsupported copy to you
as a "guideline"...)
Charlie
|
| 3858.5 | "no" answer is acceptable | CUJO::SAMPSON | | Fri Apr 11 1997 00:32 | 21 |
| Charlie,
My technical lead, Al Meier, asked me to ask you about a commitment.
In this case, I believe the customer can live with a "no" answer, and just
declare a "kludge" constant IO_M_ALLOWFAST with value 4000 hex.
>Who is your customer? ;-)
He's an application developer for a "major" aerospace company.
>(Also, if the list of routines is short, I could probably run SDL on the
>V7.1 Starlet bindings out currently, and give a UNsupported copy to you
>as a "guideline"...)
That won't be necessary, since the customer really needs just the
one new constant IO_M_ALLOWFAST. Also, he isn't contractually obligated
to use only the Ada language, so he can always resort to C or Fortran,
as needed.
Thanks,
Bob Sampson
|
| 3858.6 | Workaround for user | KMOOSE::CMCCUTCHEON | Charlie McCutcheon | Fri Apr 11 1997 10:11 | 22 |
| > That won't be necessary, since the customer really needs just the
>one new constant IO_M_ALLOWFAST. Also, he isn't contractually obligated
>to use only the Ada language, so he can always resort to C or Fortran,
>as needed.
Based on this, I'll say "no" to commiting to updating all bindings to 7.1.
If your customer wants something more, I'm open to taking more requirments.
You can mail me offline, I'd want to know if they're one of our known
enthusiastic customers whose names I'm familiar with.... ;-)
If your customer just needs one constant declared. They can edit starlet
themselves if they want:
$ acs
>set library ada$predefined
>extract source starlet
$ edit starlet_.ada
$ ada /nodebug starlet_.ada
$ acs set library ...
Charlie
|
| 3858.7 | need is isolated to one or two modules | CUJO::SAMPSON | | Sun Apr 13 1997 15:41 | 11 |
| Charlie,
Thanks, but he can just define and use the constant in one or two
Ada modules that actually perform skipfile operations on tape, can't he?
He just noticed that the definition was missing from STARLET after upgrading
to OpenVMS V7.1 and re-installing Ada V3.3, and he just wanted to know why.
I promised him I'd try to find out when STARLET would get updated. If you
happen to know a general, non-commital answer, that would help too.
Thanks,
Bob Sampson
|
| 3858.8 | | KMOOSE::CMCCUTCHEON | Charlie McCutcheon | Mon Apr 14 1997 10:54 | 8 |
| >I promised him I'd try to find out when STARLET would get updated. If you
>happen to know a general, non-commital answer, that would help too.
No plans in place to upgrade it. Requests can change that. Especially requests
saying what portions are wanted. I'm less likely to do a bit tedious upgrade for
one constant. ;-)
Charlie
|