[dev] Redundant index in table mnemo_memos?

[Duck]

On Saturday 19 August 2006 00:56, Chuck Hagenbuch wrote:
> Yup, sure looks like it. What we should have is an index on the
> memo_id field. I think I'd prefer to leave the memo_owner index and to
> make the primary key just memo_id. This would go along with something
> else I want to do for Nag, Kronolith, and Turba as well as Mnemo,
> which is to make all of the _id fields integers again (they became
> long strings when we thought we needed them to be uids, but then we
> realized the uids had to be seperate anyway).

There are many other inconsistencies in sql scripts and the actual api usages. 
Another example is Kronolith. Where the occurrence type is always in integer. 
But in the table structure is defined as varchar(11). This should be fixed in 
the next releases, as not requires any programming. Because in general the 
structure is just copied in and now one bothers about it.

Duck