I am running npgsql v3.7 with .NetCore on Ubuntu.
When I execute a select query and a cell in any row in the results is null, an exception is thrown with the error message "Column is null".
I am having to work around this by putting every column in the select clause inside a case statement which tests for NULL
"CASE WHEN " + fieldName + " IS NULL THEN '' ELSE " + fieldName + " END "
This seems a bit extreme and should not be necessary. Has anyone else come across this.
Thanks.
You are probably trying to read the column like this:
using (var reader = cmd.ExecuteReader()) {
reader.Next();
var o = reader.GetString(0); // Or any other of the Get methods on reader
...
}
This code will fail if the column contains a null, and is the expected behavior. In ADO.NET, you need to check for a null value with reader.IsDBNull(0) before actually getting the value. That's just how the database API works.
I don't know why NULL values are giving you errors, but you can do away with the ugly CASE statement in favor of using COALESCE:
"COALESCE(" + fieldName + ", '')"
Ideally you should make a configuration change such that NULL values do not cause this problem.
Related
I have a JPA/Springboot application backed by a Postgres database. I need to get a records that is equal to a substring passed back to the server.
For example:
Select * from dp1_attachments where TRIM(RIGHT(dp1_submit_date_dp1_number::text, 5)) ='00007'
This query works in PgAdmin, but not in the JPA #Query statement.
#Query("SELECT a.attachmentsFolder as attachmentsFolder, a.attachmentNumber as attachmentNumber, a.attachmentName as attachmentName, a.dp1SubmitDateDp1Number as dp1SubmitDateDp1Number,a.attachmentType as attachmentType, a.attachmentDate as attachmentDate, a.attachmentBy as attachmentBy "
+ "FROM DP1Attachments a WHERE TRIM(SUBSTRING(a.dp1SubmitDateDp1Number::text, 5 )) = :dp1Number")
I've also tried CASTing the parameter like this:
#Query("SELECT a.attachmentsFolder as attachmentsFolder, a.attachmentNumber as attachmentNumber, a.attachmentName as attachmentName, a.dp1SubmitDateDp1Number as dp1SubmitDateDp1Number,a.attachmentType as attachmentType, a.attachmentDate as attachmentDate, a.attachmentBy as attachmentBy "
+ "FROM DP1Attachments a WHERE TRIM(SUBSTRING(CAST(a.dp1SubmitDateDp1Number as string, 5 ))) = :dp1Number")
but the application won't even run, and returns an error that the query isn't valid.
If I make no attempt to cast it, I get an error that function pg_catalog.substring(numeric, integer) does not exist
UPDATE
I've also tried creating a native query instead but that also doesn't seem to work.
List<DP1AttachmentsProjection> results = em.createNativeQuery("Select * FROM dp1_attachments WHERE TRIM(RIGHT(CAST(dp1_submit_date_dp1_number as varchar),5)) =" + dp1Number).getResultList();
In place of varchar I have also tried string and text.
Errors come back similar to ERROR: operator does not exist: text = integer. Its like the CAST is being ignored and I'm not sure why.
I also tried the following as a native query:
em.createNativeQuery("Select * FROM dp1_attachments WHERE TRIM(RIGHT(dp1_submit_date_dp1_number::varchar),5)) =" + dp1Number).getResultList();
and get ERROR: syntax error at or near ":"
FINAL SOLUTION
Thanks to #Nenad J I altered the query to get the final working solution:
#Query(value = "SELECT a.attachments_Folder as attachmentsFolder, a.attachment_Number as attachmentNumber, a.attachment_Name as attachmentName, a.dp1_Submit_Date_Dp1_Number as dp1SubmitDateDp1Number,a.attachment_Type as attachmentType, a.attachment_Date as attachmentDate, a.attachment_By as attachmentBy FROM DP1_Attachments a WHERE TRIM(RIGHT(CAST(a.dp1_Submit_Date_Dp1_Number as varchar ), 5 )) = :dp1Number", nativeQuery = true)"
Default substring returns a string, so substring(integer data,5) returns a string. Thus no need for the cast.
#Query("SELECT * FROM DP1Attachments a WHERE TRIM(SUBSTRING(a.dp1SubmitDateDp1Number, 5)) = :dp1Number")
But I recommend in this case use native query like this:
Put this code in your attachment repository.
#Query(value="SELECT * FROM DP1Attachments AS a WHERE TRIM(SUBSTRING(a.dp1SubmitDateDp1Number, 5 )) = :dp1Number", nativeQuery=true)
Be careful with the column's name.
I'm getting SQL exception while executing the query which contains
split_part() method as split_part(value::TEXT,':', 1).
String queryStr = " select split_part(value::TEXT,':', 1) from table";
Query query = entityManager.createNativeQuery(queryStr);
List results = query.getResultList();
ERROR 2020-02-10 14:54:37,926 [http-nio-7070-exec-1] 142 - ERROR: syntax error at or near ":"
Position: 855
Your obfuscation layer probably chokes on the :: operator. Use the cast() operator instead:
String queryStr = " select split_part(cast(value as text),':', 1) from table";
But why do you think you need the cast to begin with? If you are storing : characters that column, it is most probably a text (or varchar) column anyway and you don't need a cast at all.
I am trying to update jsonb column in java with mybatis.
Following is my mapper method
#Update("update service_user_assn set external_group = external_group || '{\"service_name\": \"#{service_name}\" }' where user=#{user} " +
" and service_name= (select service_name from services where service_name='Google') " )
public int update(#Param("service_name")String service_name,#Param("user") Integer user);
I am getting the following error while updating the jsonb (external_group) cloumn.
### Error updating database. Cause: org.postgresql.util.PSQLException: The column index is out of range: 2, number of columns: 1.
### The error may involve com.apds.mybatis.mapper.ServiceUserMapper.update-Inline
I am able to update with the same way for non-jsonb columns.
Also if I am putting hardcoded value it's working for jsonb columns.
How to solve this error while updating jsonb column?
You should not enclose #{} in single quotes because it will become part of a literal rather than a placeholder. i.e.
external_group = external_group || '{"service_name": "?"}' where ...
So, there will be only one placeholder in the PreparedStatement and you get the error.
The correct way is to concatenate the #{} in SQL.
You may also need to cast the literal to jsonb type explicitly.
#Update({
"update service_user_assn set",
"external_group = external_group",
"|| ('{\"service_name\": \"' || #{service_name} || '\" }')::jsonb",
"where user=#{user} and",
"service_name= (select service_name from services where service_name='Google')"})
The SQL being executed would look as follows.
external_group = external_group || ('{"service_name": "' || ? || '"}')::jsonb where ...
I'm changing queries from an Oracle Database to PostgreSQL, and in this query I am getting this error:
ERROR: syntax error at or near "SET"
the query is:
{call UPDATE alarm_instance SET last_update_time=default, wait_expire_time=null, core_number=nextval(SEQ_ALRM_NUMBR)
where wait_time <= current_date RETURNING alarm_instance_id bulk collect INTO ?}
I am using JDBC to connect to the database and here is the call code
try (CallableStatement cs = super.prepareCall_(query)) {
cs.registerOutParameter(1, Types.ARRAY);
cs.execute();
...
I have taken a long look at Postgres documentation and cannot find what is wrong and didn't find any answer to this specific situation
An UPDATE statement can't be executed with a CallableStatement. A CallableStatement is essentially only intended to call stored procedures. In case of Oracle that includes anonymous PL/SQL blocks.
And bulk collect is invalid in Postgres to begin with.
It seems you want something like this:
String sql =
"UPDATE alarm_instance " +
" SET last_update_time=default, " +
" wait_expire_time=null, "
" core_number=nextval('SEQ_ALRM_NUMBR') " +
" where wait_time <= current_date RETURNING alarm_instance_id";
Statement stmt = connection.createStatement();
stmt.execute(sql);
int rowsUpdated = stmt.getUpdateCount();
ResultSet rs = stmt.getResultSet();
while (rs.next() {
// do something with the returned IDs
}
I'm doing a little application for the work and it includes a form. When the person using the form doesn't put a value in a data type "numeric" (lets say a PIN number) or even a date in one of my "date" Data Type fields (lets say his anniverssary), it returns me with the errors as follow:
Error Type – Type d'erreur
_Microsoft OLE DB Provider for ODBC Drivers (0x80004005)_
ERROR: invalid input syntax for type numeric: ""; Error while executing the query
Error Type – Type d'erreur
_Microsoft OLE DB Provider for ODBC Drivers (0x80004005)_
ERROR: syntax error at or near ")"; Error while executing the query
So it seems that when the person using the form enters nothing, it returns the string "empty" that is "".
Why can't the numeric type and data type read that as a NULL entry? These fields are not mandatory and so I need to have them be sometimes blank.
How can I make it possible? Is their a way to keep using "numeric" and "date" type so that when the user enters nothing in those fields, the table fills with a blank case insted of giving me this error?
Here is my sql statement :
trsConn.EXECUTE "INSERT INTO ""TRS"".employeetbl ( "& _
"""firstName"", " & _
"""lastName"" , " & _
"""detContract"", " & _
"sle, " & _
"""posNumber"" "& _
") VALUES (" & _
"'" & Request.Form("empFirst") & "', " & _**
"'" & Replace(Request.Form("empLast"), "'", "`") & "', " & _
"'" & Request.Form("dateContract") & "', " & _
"'" & Request.Form("sle") & "', " & _
"'" & Request.Form("posNum") & "');"
(The posNum and dateContract are both respectivly of type "numeric" and "date"
Thanks so much for the help. Looking forward for hearing what you geniuses have to say.
The concept of NULL in SQL is pretty muddled and inconsistent ... but it's very clear that '' is distinct from NULL.
'' isn't NULL, it's '', the empty string. You can't convert it to a date, number, etc:
regress=# SELECT CAST('' AS DATE);
ERROR: invalid input syntax for type date: ""
LINE 1: SELECT CAST('' AS DATE);
^
regress=# SELECT CAST('' AS NUMERIC);
ERROR: invalid input syntax for type numeric: ""
LINE 1: SELECT CAST('' AS NUMERIC);
^
Some products - notably Microsoft Access and old versions of MySQL - are confused about that matter. NULL is NULL, '' is the empty string; they aren't the same thing. You can't convert one to the other.
So it seems that when the person using the form enters nothing, it
returns the string "empty" that is "". Why can't the numeric type and
data type read that as a NULL entry? These fields are not mandatory
and so I need to have them be sometimes blank.
That's your application's job. When your app it sees the empty string come in on a form field for a numeric, date, or similar, it should send NULL to the database, not ''. That's normally a routine part of converting data from user input before it's supplied to the database. It is vital that you do such conversion; you should never just send values from the user straight to the database.
A quick search suggests that asp classic uses null or undefined as its null values; you should be able to pass them into your prepared statements when something is null.
The fact that you get a syntax error after the error about '' suggests that you're building your SQL statements as strings, not using prepared statements with placeholders. (Thanks JayC for the SO question ref). This is begging for SQL injection; in other words your application is critically insecure. Imagine what happens if the user enters the "date":
2012-01-01'); DROP SCHEMA public;--
and your app happily turns that into
INSERT INTO sometable (blah, blah, blah) VALUES (1, 2, DATE '2012-01-01'); DROP SCHEMA public;--');
The DROP SCHEMA then merrily executes and whoops, splat, there goes your database. That's just the dumbest, simplest kind of SQL injection attack too.