Re: 1.2 -> 1.4 data port missing tables & constraints not in DB?

[..hidden..]

> Hi;
>
> I think I understand what difficulties you are getting at.
>
> The jobs table can be ignored.  It is there for asynchronous processing of
> processing jobs, not for job costing.  it's a bad name for the table and
> we
> should probably change it.
>
> Am I right in assuming that you need to be able to track a "generic
> project
> part?"  We could add an add-on script to allow a specific part number to
> be
> duplicated for generic projects, or you could just let it autogenerate (or
> use the project code).
>
No not really I dont think. My assignment now is QA to map the 1.4
database schema with the 1.2 and look for data that might be lost. With I
am assuming sql/upgrades/1.2-1.4.sql?

One issue so far is stock 1.2 has a parts table with a column of
project_id's (that other users will have) that is silently dropped by 1.4
yet still used in the code. bin/ic.pl line 577 and in Ledgersmb/IC.pm line
243 (Unchanged 1.2 - 1.4) there is a block of code that tries to retrieve
the project_id from the db.
However that project_id is gone from the parts table? Do you not use ic.pl
or IC.pm anymore?

In general I think ledgersmb upgrades to a data set should preserve
historical data users might have since they may have other applications
that have begun to access that data.

An assembly I think can have a one to one relation with a project in 1.2,
in the 1.4 system you intend (I think)to have a many to many association
with parts and assembles? Or Any number of assemblies can be part of any
number of projects etc?

I am just not wrapping my head around how to port existing data to 1.4
without loosing any
yet :) but i will dont worry.

Cheers Turtle