Are there any reasons why the data migration plugin always converts the type
of the field to INT8? This can be a bit annoying if there are foreign keys
of type INT4 which refers to the fields. The default behaviour in PostgreSQL
is to create new serial columns of type int4.
I suggest you keep the type of the field, and only change it only according
to the type map.
Another small annoyance is that the plugin uses "_key" to name the
sequences. Could this be changed to "_seq" to be consistent with the SERIAL
type in PostgreSQL?
-Brage
*** frmWizard.frm.orig Wed Feb 12 15:52:35 2003
--- frmWizard.frm Wed Feb 12 15:48:21 2003
***************
*** 1621,1627 ****
' Matthew MacSuga Auto Increment Fix
If auto_increment_on = 1 Then
If LCase(newColumnArray(Y)) = LCase(auto_increment_field_name) Then
! szTemp2 = "int8 DEFAULT nextval('" & QUOTE & auto_increment_table & "_"
& auto_increment_field_name & "_key" & QUOTE & "')"
End If
End If
--- 1621,1627 ----
' Matthew MacSuga Auto Increment Fix
If auto_increment_on = 1 Then
If LCase(newColumnArray(Y)) = LCase(auto_increment_field_name) Then
! szTemp2 = szTemp2 & " DEFAULT nextval('" & QUOTE & auto_increment_table
& "_" & auto_increment_field_name & "_key" & QUOTE & "')"
End If
End If
---------------------------(end of broadcast)---------------------------
TIP 6: Have you searched our list archives?
http://archives.postgresql.org
