I just experienced something scary with SQL Developer and the “Database Copy” utility. One table in production was truncated by a mistake. I have always been a little against using the SQL Developer functionality “Database Copy …” in production, but for this one task during deployment of a new release, this customer (including me) have always used this feature. The customer have two somewhat static tables in a meta database (in a test zone), where they store table and columns comments (don’t ask me why they are not using the COMMENT feature in the Oracle database). During a release the customer copy the data from these two tables into two production databases. By accident or some kind of bug SQL Developer started to truncate ALL tables in the schema.
Continue Reading »
We have just been through a relatively tricky situation, trying to copy our test schemas from our Oracle 12c database to the new Oracle 19c test database. We were trying to use data pump to copy the schemas. The copy was ordered from operation, and the schemas was exported and imported as ordered. Before importing both editions and the schemas was created manually. The schemas was also granted privileges to use the newly created editions. So check … test database was ready to be used? But no. Things failed in test.
Continue Reading »
Lately I have seen projects using the NUMBER datatype for table columns without setting precision and scale. When I see this I always question “Is this intentional or did they just happen to forget to set the values?”. This is the problem with using implicit configuration/settings. You don’t really know if it was a mistake or intentional. So I usually go for the explicit definitions (but of course … it depends).
Continue Reading »
I often see DDLs with NOT NULL definitions which is really redundant. When I see such DDLs I really question “do the writer really know what he or she is doing”. They might not hurt you application, but why set them if they are redundant? (If you know I reason please comment)
Example:
CREATE TABLE post (
id NUMBER(19,0) GENERATED ALWAYS AS IDENTITY NOT NULL ,
title VARCHAR2(255 CHAR),
version NUMBER(10,0) NOT NULL,
PRIMARY KEY (id));
In this case I can see two reasons why I would NOT define the ID column with a NOT NULL constraint.
Continue Reading »
In the last article we looked at Spring-boot and accessed the database using the JdbcTemplate class.Today we’ll rewrite the same application but by using JPA and Hibernate instead. The Spring Initializr web page have changed a little since the last time, so the first part will be a little different.
Continue Reading »
I just came over these slides and nice little overview of the PL/SQL language by Morten Braten. The slides covers many good tips to follow when coding PL/SQL. It’s not ment to be an introduction to the PL/SQL language, so many topics are just mention briefly and would need to be followed by further reading and/or investigation. But it is definitely worth a look!
There is only one thing I used to miss from my Windows environment – Notepad++. After switching to Mac and OSX I have been looking for an alternative. For a while I was using Atom, but did not really find it efficient. So I often ended up using the default TextEdit utility.
The other day I got a tip from a Java developer at Bouvet. He was using Visual Studio Code 2 (https://code.visualstudio.com/) . I have used this for two weeks now, and is very very happy with it. Finally a text editor which beets Notepad++.
Today we’ll be looking at Spring-boot and a tools which developers use to migrate databases, FlywayDB. We’ll also look at how you can automate grants towards our two roles (which we created in part 3) whenever new tables are created.
Continue Reading »
