* Make more compatible with existing Oracle Schema
* TODO: Splitting of XML file
This commit is contained in:
parent
08a3110d44
commit
f7fe785407
File diff suppressed because it is too large
Load Diff
|
@ -1,17 +1,23 @@
|
|||
1. adempiere-model.xml
|
||||
XML File
|
||||
* adempiere-model.xml
|
||||
- DB independent adempiere schema
|
||||
|
||||
2. adempiere-data.jar
|
||||
* adempiere-data.jar
|
||||
- DB independent adempiere seed data
|
||||
|
||||
Changes from existing schema
|
||||
* Use SQL standard varchar instead of Oracle proprietary nvarchar datatype.
|
||||
* Use Integer datatype for _ID field instead of Number(10)
|
||||
Datatype difference between Oracle and PostgreSQL
|
||||
* ID Field map to Integer in postgreSQL and Number(10) in Oracle.
|
||||
* Integer field map to Integer in postgreSQL, Number(10) in Oracle
|
||||
( Oracle's Integer datatype map to number(38) which is not standard)
|
||||
* PostgreSQL have no native support for CLOB and BLOB, BYTEA is use instead.
|
||||
* Timestamp field map to Timestamp in PostgreSQL, DATE in Oracle.
|
||||
|
||||
Issue
|
||||
* Global Temporary Table is different between Oracle and PostgreSQL
|
||||
* Integer datatype size is different between Oracle and Postgresql
|
||||
- 4 byte in PostgreSQL, Number(38) in Oracle
|
||||
* Fyracle script need to be tested by someone with access to the developer kit.
|
||||
* Integer datatype for primary key is too small in Java(sqlj) and PostgreSQL
|
||||
- 32 bit signed, probably should use bigint in PostgreSQL and long in Java (sqlj)
|
||||
* XML model and data file to be separated by table to ease maintenance.
|
||||
|
||||
Note
|
||||
* SQL standard varchar datatype is use in place of Oracle's non standard nvarchar
|
||||
datatype. This changes means you must create your database to use UTF8 as default
|
||||
character set.
|
||||
* Integer instead of BigInt will be use as datatype for ID field to maintain compatibility
|
||||
with existing schema.
|
Loading…
Reference in New Issue