[Info-vax] VSI roadmap

[Johnny Billquist]

On 2023-08-20 19:44, Jan-Erik Söderholm wrote:
> Den 2023-08-20 kl. 19:30, skrev Johnny Billquist:
>> On 2023-08-20 18:00, Dave Froble wrote:
>>> On 8/20/2023 11:01 AM, Johnny Billquist wrote:
>>>> On 2023-08-18 22:39, Dave Froble wrote:
>>>> [...]
>>>>
>>>>> Ok, understand what you are doing.
>>>>>
>>>>> Caviet, I've never used Rdb.
>>>>>
>>>>> Another, I've only used SQL2000 from Microsoft.
>>>>>
>>>>> Regardless, consider the following:
>>>>>
>>>>> SELECT * From CustomerTable Where Country is "USA"
>>>>>
>>>>> And where country is not a key.
>>>>>
>>>>> In RMS this cannot be done, without scanning all records, and use 
>>>>> some type of
>>>>> mapping of the record definition, CDD, include file, in program 
>>>>> definition.
>>>>> However, if the database contains the record definitions, then the 
>>>>> database
>>>>> would return a recordset of the records with Country equal to USA.
>>>>
>>>> Maybe I missed something. I didn't see RMS in there anywhere.
>>>> So why are you bringing that up?
>>>>
>>>> Rdb is not the same thing as RMS.
>>>>
>>>>   Johnny
>>>>
>>>
>>> The topic was the VSI replacement for CDD.  CDD was/is used to 
>>> provide record definitions for RMS and anything else that needed 
>>> such.  My opinion is that the lack of record definitions in any data 
>>> store is poor design.
>>
>> Well, in this case it was Rdb, wasn't it? But sure, you need record 
>> definitions for anything/everything. And it's nice if that comes from 
>> the same source. Which is what I think is what Jan-Erik was 
>> describing/doing.
> 
> Well, I was just describing how to get record defs for your Rdb tables
> in a format that can be used directly by your programming language.

Right. Which means the record definitions are implicitly accessible 
through the data store. Which was what David was complaining about not 
being the case. I became confused because what you were explaining was 
something done against Rdb, and David then complained about RMS, which I 
didn't think you talked about at all. And David then talked about CDD, 
which you also didn't mention either.

>> Data stored in Rdb, record definitions extracted with RMU to whatever 
>> language used, then eventually included in the compilation.
>>
>> Now, where did RMS come into this, and it would certainly seem that 
>> the record definitions are available along with the data store.
>>
>> How this is related to CDD is another question. But that apparently 
>> came from the fact that Rdb is owned by Oracle, and so is CDD. And VSI 
>> have a replacement for CDD, but Oracle also seem to be going to port 
>> Rdb to x86-64. Unclear if they also care about CDD. But since VSI have 
>> a replacement, that seems non-critical. Is CDD/VDD tied in with that 
>> RMU tool?
>>
>>    Johnny
>>
> 
> Not sure I follow... Why should CDD be "tied in" with RMU?
> The "Rdb Management Utility".

Never claimed it was. I did ask if there was a tie in. And by your 
comment I think the answer is "no".

> If you have CDD, the compilers read records defs directly from CDD
> and RMU has nothing to do with that.
> 
> If you do not have CDD, RMU can create record defs for you to be
> later included by the compiler.

Right.

So again - what did this have to do with RMS? You explained how you were 
using Rdb, and somehow David slid into CDD and RMS from your explanation.

But maybe we can close that thread. If people want to talk about CDD, 
then let's do that. :-)

   Johnny