Do you have any idea on how to fix some errors with have a isnull(integer, integer)() function in EDB PostgreSQL? I convert databases from SQLServer to PostgreSQL. it returns errors like this:
"SQLSTATE= 42883. ERROR function ISNULL(integer, integer) does not exist".
and
"SQLSTATE= 42883. ERROR function ISNULL(bigint, integer) does not exist"
I believe this should be fix with create the function instead, but i have no idea how to create working solution.
I need to replace in database ISNULL() to COALESCE() function
Please, Help!
Maybe something like this:
create function public.isnull(integer, integer) returns ?
stable language sql as coalesce(?)';
Could You please help me with the correct syntax for this function?
Related
I have created a UDF function to process some content based on dates in Azure cosmos database. The function looks like below
function userDefinedFunction(array,dateString){
var moment = require('moment');
const startDate = moment(dateString);
const endDate = moment(dateString).add(1,'days');
// filter the array by the dates and return a value
}
When the above UDF is used inside a query the following error message is thrown .
Encountered exception while executing Javascript. Exception = ReferenceError: 'require' is not defined
This error is seemed to be an error orginating from incorrect import of the moment node module.
i have already tried checking on the microsoft offcial docs about the UDFs and using of node module inside UDF.
I tried surfing through the internet about this issue, but both the methods did not provide me with satisfactory answer.
so i would like to know how to import a node module and use it inside an UDF function . Thanks a lot in advance .
Importing modules is not supported for any of the service-side features including stored procedures, triggers and user-defined functions.
This is not currently documented. Will ask for this to be updated.
I a trying to use ARRAY_AGG function in Spark SQL. When I use it, it throws error
<<Undefined function: 'array_agg'. This function is neither a registered temporary function nor a permanent function registered in the database 'default>>
Dataset<Row> finalDS1 = sparkSession.sql("select array_agg(company_private_id) from TEMP_COMPANY_PRIVATE_VIEW");
Anyone know how to solve it? I am trying to compare one array with another column. For that I am using ARRAY_AGG.
"select cp.array_column & (select array_agg(int_column) from getCompanyPrivateDS ds1) as filtered_data from getCompanyPrivateDS cp"
I think this is a documentation error by Spark. They clearly show array_agg() in their function list: https://spark.apache.org/docs/latest/api/sql/index.html#array_agg
but I have also experienced that this function doesn't work on Spark 3.1.2
Collect_set() and collect_list() should work for your purposes: the former dedupes results, while the latter doesn't.
I am trying to run a parameterised query using the npm module #google-cloud/bigquery.
Something like this:
SELECT * FROM myTable WHERE id IN (#ids);
I have no idea how bigQuery is expecting the parameter ids formatted.
My options.params look like something like this:
{ ids: '"1234", "4567"'}
But I don't get any result back. I know there are results, I can see them in bigquery and if I remove the parameter and just inject the string works just fine.
It seem pretty easy, but I can't figure out why it doesn't work, anyone who is willing to help me out?
Thank you in advance
Of course I found the solution as soon as I posted the question...
Thanks to this thread! Need to do some gymnastic...
So provided that the parameter is a string like:
'1234,5678'
We need to do:
WHERE id IN UNNEST(REGEXP_EXTRACT_ALL(#ids,"[0-9a-zA-Z]+"))
REGEXP_EXTRACT_ALL - returns an array
UNNEST - flattens the array for the IN clause as stated in the link above.
Working with KnexJS on a project and have been using the .raw() method throughout the project with out a single issue.
I now have one case where the .raw is just not being built into the SQL, and so I end up with the syntax error at end of input error. If I dump the SQL string, I can see why as it's failing
'update "mytable" set "a" = ? returning '
I can see why it's having an issue, the problem I'm having is that the returning is a Raw value as such, and so I just can't figure out why it's not being compiled with the SQL.
knexQuery.update(data).into('mytable').returning( knexQuery.raw('mytable::json') );`
If I use a string, in place of the raw in the returning() method, it will compile the string into it.
If I console out the knexQuery.raw('mytable::json') part, it shows as a Raw object, with the right object data...
Raw {
client:
Client_PG {
... },
sql: 'mytable::json',
bindings: undefined,
_wrappedBefore: undefined,
_wrappedAfter: undefined,
_debug: undefined } }
I know that the SQL works, as I've filled in the missing part and it works, but the SQL is not the issue, I just can't figure out why the raw() method is not being compiled with the SQL string.
I also have another piece of code (and INSERT one) somewhere else in my code that is using the same returning( knexQuery.raw(....) ) and that works perfectly fine.
I'm starting to think this is a bug in the code, but after spending an hour going through the KnexJS library code, I can't see any reason why it would not work.
So why does this code not build the raw into the query, while my other one has no problem and works?
I'm getting to the point where I just want to use something else to get around the problem, but it's just not possible without using this method.
This is bug in knex https://runkit.com/mikaelle/592412c3a631940012a51928 please open issue in knex github. Looks like handling raw input just haven't been implemented to returning builder method.
Anyways looks like this works as a workaround:
knex('TestTable')
.insert({ foo: 'bar' })
.returning([knex.raw('mytable::json')]) // raw wrapped in array
.toSQL();
I'm trying to make something like this:
int count = new Select().From(tblSchema).Where("Type & 1").IsEqualTo("1").GetRecordCount();
And the error message is:
Incorrect syntax near '&'.
Must declare the scalar variable "#Deleted".
Is it possible to do that with SubSonic?
Must declare the scalar variable
"#Deleted"
The second error would be caused by using logical deletes on the table you are querying (the table has an isDeleted or Deleted column).
But I'm looking through the code, I'm not sure how that parameter is getting in there. The SqlQuery.GetRecordCount method doesn't call CheckLogicalDelete(), from what I can tell. Is that error message unrelated?
This seems to be a bug in the way SubSonic is naming it's parameters when it generates the SQL to be executed.
What's happening is that SubSonic is looking at "Type & 1" and then creating a parameter to compare against called #Type&10 which is not a valid SQL parameter name. So you'll end up with the following SQL from your original query. You should submit a bug to http://code.google.com/p/subsonicproject/
Meanwhile you can workaround the bug for now by using an inline query:
http://subsonicproject.com/docs/Inline_Query_Tool
It is a little fuzzy as to what you are trying to accomplish but here is a best guess.
int count = new Select().From(tbl.Schema).Where(tbl.TypeColumn).IsEqualTo(true).GetRecordCount();