I've just wasted the past two hours of my life trying to create a table with an auto incrementing primary key bases on this tutorial, The tutorial is great the issue I've been encountering is that the Create Target fails if I have a column which is a timestamp and a table that is called timestamp in the same table...
Why doesn't oracle flag this as being an issue when I create the table?
Here is the Sequence of commands I enter:
Creating the Table:
CREATE TABLE myTable
(id NUMBER PRIMARY KEY,
field1 TIMESTAMP(6),
timeStamp NUMBER,
);
Creating the Sequence:
CREATE SEQUENCE test_sequence
START WITH 1
INCREMENT BY 1;
Creating the trigger:
CREATE OR REPLACE TRIGGER test_trigger
BEFORE INSERT
ON myTable
REFERENCING NEW AS NEW
FOR EACH ROW
BEGIN
SELECT test_sequence.nextval INTO :NEW.ID FROM dual;
END;
/
Here is the error message I get:
ORA-06552: PL/SQL: Compilation unit analysis terminated
ORA-06553: PLS-320: the declaration of the type of this expression is incomplete or malformed
Any combination that does not have the two lines with a the word "timestamp" in them works fine. I would have thought the syntax would be enough to differentiate between the keyword and a column name.
As I've said I don't understand why the table is created fine but oracle falls over when I try to create the trigger...
CLARIFICATION
I know that the issue is that there is a column called timestamp which may or may not be a keyword. MY issue is why it barfed when I tried to create a trigger and not when I created the table, I would have at least expected a warning.
That said having used Oracle for a few hours, it seems a lot less verbose in it's error reporting, Maybe just because I'm using the express version though.
If this is a bug in Oracle how would one who doesn't have a support contract go about reporting it? I'm just playing around with the express version because I have to migrate some code from MySQL to Oracle.
There is a note on metalink about this (227615.1) extract below:
# symptom: Creating Trigger fails
# symptom: Compiling a procedure fails
# symptom: ORA-06552: PL/SQL: %s
# symptom: ORA-06553: PLS-%s: %s
# symptom: PLS-320: the declaration of the type of this expression is incomplete or malformed
# cause: One of the tables being references was created with a column name that is one of the datatypes (reserved key word). Even though the field is not referenced in the PL/SQL SQL statements, this error will still be produced.
fix:
Workaround:
1. Rename the column to a non-reserved word.
2. Create a view and alias the column to a different name.
TIMESTAMP is not listed in the Oracle docs as a reserved word (which is surprising).
It is listed in the V$RESERVED_WORDS data dictionary view, but its RESERVED flag is set to 'N'.
It might be a bug in the trigger processing. I would say this is a good one for Oracle support.
You've hinted at the answer yourself. You're using timestamp as a column name but it's also a keyword. Change the column name to something else (eg xtimestamp) and the trigger compiles.
Well, I'm not totally sure about it, but I think this happens because the SQL code used to manipulate and access database objects is interpreted by some interpreter different form the one used to interpret PL/SQL code.
Have in mind that SQL an PL/SQL are different things, and so they are processed differently. So, I think there is some error in one interpreter, just not sure which one is.
Instead of having Oracle maintain a view, use EXECUTE IMMEDIATE (i.e. if 'Rename the column to a non-reserved word' is not an option.
You can execute via EXECUTE IMMEDIATE. IT's not better way but work's and avoid column rename.
In my case rename column will be a caotic way
Related
CREATE TABLE test ( id serial primary key, name text );
CREATE OR REPLACE PROCEDURE test_insert_data( "name" text)
LANGUAGE SQL
AS $$
INSERT INTO public.test values("name")
$$;
Error & Hint:
column "id" is of type integer but expression is of type character varying
LINE 4: INSERT INTO public.test values("name")
^
HINT: You will need to rewrite or cast the expression.
I followed this tutorial: https://www.enterprisedb.com/postgres-tutorials/10-examples-postgresql-stored-procedures.
Obviously, I don't need to attach the column id for inserting.
There is no quoting issue, like comments would suggest.
And the linked tutorial is not incorrect. (But still bad advise.)
The missing target column list is the problem.
This would work:
CREATE OR REPLACE PROCEDURE test_insert_data("name" text)
LANGUAGE sql AS
$proc$
INSERT INTO public.test(name) -- !! target column list !!
VALUES ("name");
$proc$;
(For the record, since "name" is a valid identifier, all double quotes are just noise and can (should) be omitted.)
If you don't specify the target column(s), Postgres starts to fill in columns from left to right, starting with id in your case - which triggers the reported error message.
(The linked tutorial also provides an ID value, so it does not raise the same exception.)
Even if it would work without explicit target column list, it's typically still advisable to add one for persisted INSERT commands. Else, later modifications to the table structure can break your code silently. With any bad luck in a way you'll only notice much later - like filling in the wrong columns without raising an error.
See:
SQL INSERT without specifying columns. What happens?
Inserting into Postgres within a trigger function
Aside: I would never use "name" as column name. Not even in a generic tutorial. That's not helpful. Any column name is a "name". Use meaningful identifiers instead.
I've created a DB2 sql script that populates a static table and then does a rename to swap out the live table with the newly updated one. Its a fairly large SQL script so I'm only including the areas that Im having a an error on.
I'm getting the error: "[IBM][CLI Driver][DB2/NT64] SQL0104N An unexpected token "RENAME" was found following "D_HOLIDAY_LOG_OLD; ". Expected tokens may include: "TRUNCATE". LINE NUMBER=382. SQLSTATE=42601".
I suspect, its a syntax issue with the RENAME commands. If I need to add the whole query, I can. Thanks in advance
CREATE OR REPLACE PROCEDURE NSD_HOLIDAY_LOG_SPROC()
LANGUAGE SQL
SPECIFIC SP_NSD_HOLIDAY_LOG_SPROC
DYNAMIC RESULT SETS 1
BEGIN
COMMIT;
TRUNCATE TABLE TMWIN.NSD_HOLIDAY_LOG immediate;
DROP TABLE NSD_HOLIDAY_LOG_OLD;
RENAME TABLE TMWIN.NSD_HOLIDAY_LOG_LIVE TO NSD_HOLIDAY_LOG_OLD;
RENAME TABLE TMWIN.NSD_HOLIDAY_LOG TO NSD_HOLIDAY_LOG_LIVE;
RENAME TABLE TMWIN.NSD_HOLIDAY_LOG_OLD TO NSD_HOLIDAY_LOG;
END#
This is frequently asked.
As you are using static SQL in an SQL PL stored procedure, you must follow the documented rules for blocks of Compound SQL (Compiled) statements.
On of those rules is that static SQL has a restricted set of statements that can appear in such a block of code.
For example, with current versions of Db2-LUW, you cannot use any of the following statically (including rename table) :
ALTER , CONNECT,CREATE, DESCRIBE, DISCONNECT, DROP, FLUSH EVENT MONITOR, FREE LOCATOR, GRANT, REFRESH TABLE, RELEASE (connection only), RENAME TABLE, RENAME TABLESPACE, REVOKE, SET CONNECTION, SET INTEGRITY, SET PASSTHRU, SET SERVER OPTION ,TRANSFER OWNERSHIP
Other Db2 platforms (Z/OS, i-series) might have different restrictions but the same principle.
To achieve what you need you can use dynamic SQL instead of Static-SQL (as long as you understand the implications).
In other words, instead of writing:
RENAME TABLE TMWIN.NSD_HOLIDAY_LOG_LIVE TO NSD_HOLIDAY_LOG_OLD;
you could instead use:
execute immediate('RENAME TABLE TMWIN.NSD_HOLIDAY_LOG_LIVE TO NSD_HOLIDAY_LOG_OLD' );
or equivalent.
You can also use two statements, one to PREPARE and the other to EXECUTE , whichever suits the design. Refer to the documentation for execute immediate.
The same is true for other statements that your version of Db2 disallows in static compound-SQL (compiled) blocks (for example, DROP, or CREATE etc.).
I would like to keep data which is already saved into a Table field varchar(32) and to convert it to BLOB in Firebird database.
I am using a software: IBExpert ....
If it is possible, how to do that ?
Let's consider you have table TEST with one column NAME:
create table test (name varchar(32));
insert into test values('test1');
insert into test values('test2');
insert into test values('test3');
commit;
select * from test;
It is possible to change the column from varchar to BLOB by the following script:
alter table test add name_blob blob;
update test set name_blob = name;
commit;
alter table test drop name;
alter table test alter column name_blob to name;
commit;
select * from test;
Specifically in IBExpert it easy to do with Firebird 2.1 and Firebird 2.5 via direct modifications of system tables (this "obsolete" method was prohibited in Firebird 3 but nothing was introduced to replace it).
This works in both directions, VARCHAR to BLOB and BLOB to VARCHAR.
You have to have a DOMAIN - that is, a named data type, in SQL, and that domain should be of BLOB type - then IBExpert itself would issue the command, that Firebird 2.x executes, if you set Firebird 2.x in database options.
If you don't have IB Expert then the following command you have to issue:
CREATE DOMAIN T_TEXT_BLOB AS -- prerequisite
BLOB SUB_TYPE 1;
update RDB$RELATION_FIELDS -- the very action
set RDB$FIELD_SOURCE = 'T_TEXT_BLOB'
/* the name of the blob domain to be set as the field's new data type */
where RDB$FIELD_SOURCE = '.....' /* name of the column */
and RDB$RELATION_NAME = '....' /* name of the table */
See also:
https://firebirdsql.org/file/documentation/reference_manuals/fblangref25-en/html/fblangref25-ddl-domn.html
https://firebirdsql.org/file/documentation/reference_manuals/fblangref25-en/html/fblangref-appx04-relfields.html
When you create a column or change its datatype without explicitly naming the type (like just varchar(10)) then under the hood Firebird creates automatically-managed one-per-column user domains with names like RDB$12345, so while you can do that too, perhaps having an explicit named domain would be more practical and more safe.
This method however fails on Firebird 3, where you do have to copy the whole table, as shown by Maxim above.
http://tracker.firebirdsql.org/browse/CORE-6052
FB devs warn about some "bugs" and "it never worked properly" however refuse to give details.
UPDATE I finally managed to reproduce the bug mentioned in the tracker.
At least in Firebird 2.1 reference counting is broken w.r.t. BLOB payloads.
So the trick seems to be to instantly re-write implicit blobs into explicit, tricking Firebird to think we supplied new content for all the BLOB values.
Assuming the names from Maxim's answer above...
To trigger and demonstrate the bug take the "vanilla" VarChar database, apply the conversion described above, and issue the following commands:
update test set name = name -- trying to force data conversion, but instead forcing Firebird reference counting bug
select cast( name as VarChar(200) ) from test - or actually any command that would try to actually read the field contents - any such attempt would be shot down with the infamous invalid BLOB ID Firebird error.
To work around the bug we must prevent Firebird from (broken) reference counting. So, we must do a fake update, invoking expression evaluator, so that Firebird optimizer would loose track of value sources and would fail to realize the data was not really changed.
update test set name = '' || name || '' -- really forcing data over-writing and conversion, bypassing BLOB reference counting.
select cast( name as VarChar(200) ) from test - now works like a charm (albeit 200 was too short a string and you would be stuck with "overflow" error
The update command can be any other triggering expression evaluator, for example update test set name = cast( name as VarChar( NNN ) ) - but you would need to devise large enough NNN for specific column of specific table. So, string concatenations with empty string is universal and does the work, at least on Firebird 2.1.
The above stands for Firebird 2.1.7 Win32. I did not manage to trigger "invalid BLOB id" with Firebird 2.5.8 Win64 - it "just worked".
At least using the single-connection schema update script, which is anyway the intended way to do database upgrades.
Maybe if I would do schema upgrades while simultaneously there would be users actively working - the FB 2.5 would get broken too, don't know.
Whether to use this shortcut risky way disregarding FB developers' hints, or to use "official" Maxim's answer, possibly dismounting and then re-creating half the database which happened to have "dependencies" upon the field to be dropped, stays up to the reader.
Is there a column comment syntax that allows me to specify a column comment directly where I declare the column in the create table statement (i.e. inline)? The 11g spec does not mention anything, on another page something is mentioned but I could not get it to work. There is a way to specify comments after creating the table, but I think it is annoying that the comment is separated from the field definition. I am looking for something like this (which does not work):
create table whatever (
field number(15,0) primary key comment 'primary key generated from sequence pkseq',
...
)
I'm afraid the "annoying" COMMENT ON syntax is the only way of doing this. SQL Server, PostgreSQL and DB2 use the same syntax (even though, as far as I know, there is no ANSI standard syntax for adding comments to database objects).
MySQL supports the way you would like it to work. I agree it would be a nicer mechanism, but in my experience so few people use comments at all that I doubt Oracle will ever change it.
I'm afraid it can only be done after table creation, using the comment on column ... is '' syntax.
A workaround to this annoying syntax is also to view and edit the tables in Oracles SQLExplorer. It contains a wizard that allows you to edit the comments right next to the columns. It even allows easy creation of alter table scripts.
My procedure when editing tables is to enter the changes in the wizard without actually executing them, then go to its DDL tab and retrieve the SQL from there (as update, not full create script) and press cancel on the wizard. Then I put the created SQL into the SQL script I am writing. Only when I am finished with the script I execute everything; I do never make any changes with the wizard itself.
Test on sqlplus (or similar), but the syntax is as follows:
-- assuming you have privileges
COMMENT ON COLUMN SCHEMA1.TABLE1.COL1
IS 'My comment'
-- then you can double check like this
SELECT * FROM all_col_comments WHERE
(OWNER, TABLE_NAME, COLUMN_NAME)
IN (('SCHEMA1','TABLE1','COL1'));
Note that the comment will now show in SQLDeveloper (or Toad or whatever env you have) until you reopen said table's properties.
Similar syntax can be used to annotate tables, indexes and materialized views. [source: https://docs.oracle.com/cd/B19306_01/server.102/b14200/statements_4009.htm]
I understand similar syntax exists for MySQL and others, but it is not proper ANSI. It's very useful, though.
Is it possible to force a column in a SQL Server 2005 table to a certain value regardless of the value used in an insert or update statement is? Basically, there is a bug in an application that I don't have access to that is trying to insert a date of 1/1/0001 into a datetime column. This is producing a SqlDateTime overflow exception. Since this column isn't even used for anything, I'd like to somehow update the constraints on the columns or something in the database to avoid the error. This is obviously just a temporary emergency patch to avoid the problem... Ideas welcome...
How is the value being inserted? If it's through a stored proc... you could just modify the Sproc to ignore that input parameter.
if it's through client-side generated SQL, or an ORM tool, otoh, then afaik, the only option is a "Before" Trigger that "replaces" the value with an acceptable one...
If you're using SQL 2005 you can create an INSTEAD OF trigger.
The code in this trigger wil run in stead of the original insert/update
-Edoode
I'd create a trigger to check and change the value
If it is a third party application then I will assume you don't have access to the Stored Procedure, or logic used to generate and insert that value (it is still worth checking the SPs for the application's database though, to see if you can modify them).
As Charles suggested, if you don't have access to the source, then you need to have a trigger on the insert.
The Microsoft article here will give you some in depth information on creating triggers.
However, SQL Server doesn't have a true 'before insert' trigger (to my knowledge), so you need to try INSTEAD OF. Have a look here for more information. In that article, pay particular note of section 37.7, and the following example (again from that article):
CREATE TRIGGER T_InsertInventory ON CurrentInventory
INSTEAD OF INSERT AS
BEGIN
INSERT INTO Inventory (PartNumber, Description, QtyOnOrder, QtyInStock)
SELECT PartNumber, Description, QtyOnOrder, QtyInStock
FROM inserted
END
Nick.
the simplest hack would be to make it a varchar, and let it insert that as a string into the column.
The more complicated answer is, you can massage the data with a trigger, but it would still have to be valid in the first place. For instance I can reset a fields value in an update/insert trigger, but it would still have to get through the insert first.