Subject | Re: [Firebird-Architect] Question about limits in fb 2.5 or 3.0 |
---|---|
Author | Jim Starkey |
Post date | 2009-02-24T14:55:56Z |
Dmitry Yemanov wrote:
it's the normal case.
considering doing it automatically in other circumstances, but probably
won't for the first release.
--
Jim Starkey
President, NimbusDB, Inc.
978 526-1376
> Jim Starkey wrote:No, there is no problem truncating segments of a compound key. In fact,
>
>>> Yep, it's already in the tracker and seems quite easy to offer. However,
>>> I foresee some problems for compound indices.
>>>
>> What problem? They're implemented in Falcon and it was straightforward.
>>
>
> It's not a technical problem, more like a support one :-)
>
> Obviously, you cannot truncate the every segment of the compound index
> before constructing the key, it can be truncated as a whole only. With
> the first segment being long enough, it could mean that other segments
> are not indexed at all (with a performance impact), regardless of the
> user intentions specified in DDL. Not a big problem for sure, just a
> possible confusion.
>
it's the normal case.
> BTW, does Falcon truncate the keys always behind the scene or does itMySQL has syntax in create index for truncated segments. Falcon is
> offer a custom CREATE INDEX extension for that purpose? We were thinking
> about the latter.
>
>
considering doing it automatically in other circumstances, but probably
won't for the first release.
--
Jim Starkey
President, NimbusDB, Inc.
978 526-1376