I want to encrypt a field in my table. I have function to do this
CREATE FUNCTION baz.encrypt
(
p BIGINT
)
RETURNS CHAR(16)
....
Now I want to update all rows to be encrypted
UPDATE FOO SET BAR = baz.encrypt(BAR)
does not work, either does
UPDATE FOO f SET BAR = SELECT baz.encrypt(BAR) FROM FOO where pk = f.pk
Just solved it. The problem is the my function expects a BIGINT however my column has alpha chars in it. So the CAST does not work
Related
When I create an object in a constructor,
myTable property is constructed on the fly, for example, from csv file. I want to have some validation for that property as I have for myInt.
Do you happen to know how to do that? In the future, I guess I will need a per-column validation...
classdef BigObject
properties
myInt (1,1) double {mustBeNonnegative, mustBeInteger} = 0;
myTable (?,?) table {mustBeGreaterThan(table{1}.Variables,1), mustBeLessThan(table{2} ? ,2)} = table;
I tried selecting columns
myTable table {mustBeGreaterThan(table{1},1)} = table;
myTable table {mustBeGreaterThan(table{2}.Variables,2)} = table;
myTable table {mustBeGreaterThan(table.Variables,3)} = table;
The best way to achieve this is to write your own simple property validation function. You can build this on top of existing functions. So, you might do something like this:
%% Base case - passes
t = table((1:10).', 'VariableNames',{'MyVar'});
someFcn(t);
%% Fails because value is negative
t2 = table(-(1:10).', 'VariableNames',{'MyVar'});
someFcn(t2)
% Example function that uses validation
function someFcn(t)
arguments
t {mustHavePositiveVar(t, 'MyVar')}
end
end
% Custom validation function
function mustHavePositiveVar(t, varName)
try
mustBeA(t, 'table');
mustBeMember({varName}, t.Properties.VariableNames)
mustBePositive(t.(varName));
catch E
error('Input must be a table with a variable named %s that is all positive', varName)
end
end
Note there's a nuance if you use the class constraint syntax at the same time as a property validation function - namely that the property gets assigned and type-coerced before the validation functions run. So if you're going to validate the type of an input, you need to omit the class constraint (otherwise the input might simply be coerced to the correct type anyway).
Wanted to create the multiple parameter of function but it gives me this error:
CREATE FUNCTION failed because a column name is not specified for
column 1.
Code below:
create function dmt.Impacted(
#nameOfColumn varchar , #nameOfParam varchar)
returns table
as
return
(select
case when '['+#nameOfColumn+']' is null or len(rtrim('['+#nameOfColumn+']')) = 0
then Convert(nvarchar(2),0)
else
#nameOfParam end from employee) ;
As the error message clearly said, the column in the returned result need a name. Either give it an alias in the SELECT like
SELECT CASE
...
END a_column_name
...
or define it in the declaration of the return type as in
...
RETURNS TABLE
(a_column_name nvarchar(max)
...
As you can see in the second form you have to specify a data type. As your current code doesn't make much sense now I cannot figure out what is the right one there. You'd need to amend it.
Note, that len(rtrim('['+#nameOfColumn+']')) = 0 is never true as len(rtrim('['+#nameOfColumn+']')) is either NULL, when #nameOfColumn is NULL or at least 2 because of the added brackets.
If #nameOfColumn is supposed to be a column name you shouldn't use varchar (especially without a length specified for it) but sysname which is a special type for object names.
Either way you should define a length for #nameOfColumn and #nameOfParam as just varchar without any length means varchar(1), which is probably not what you want. And maybe instead of varchar you want nvarchar.
You may also want to look into quotename().
Define name of column in SELECT statement :
(select case when '['+#nameOfColumn+']' is null or
len(rtrim('['+#nameOfColumn+']')) = 0
then Convert(nvarchar(2),0)
else #nameOfParam
end as name_column -- define column name
from employee)
Also, your function parameter has no data length, by default it will accept only 1 character #nameOfColumn varchar , #nameOfParam varchar & rest will trim.
I'm sure this could be a duplicate but I can't seem to find the right search phrase.
Given a table in a named schema (i.e. not dbo) requires you include the schema name in the statement. So previously I'd have simply written it as so:
UPDATE [Schema].[Table1]
SET [AColumn] =
(
SELECT [SomeColumn]
FROM [Schema].[Table2]
WHERE [Schema].[Table2].[SameColumnName] = [Schema].[Table1].[SameColumnName]
);
But since More than two-part column name is deprecated, I need to find a new way to do this which is future proof. I have come up with 2 options, firstly using an alias:
UPDATE [Alias1]
SET [AColumn] =
(
SELECT [SomeColumn]
FROM [Schema].[Table2] [Alias2]
WHERE [Alias2].[SameColumnName] = [Alias1].[SameColumnName]
)
FROM [Schema].[Table1] [Alias1];
The second way is the one I'm really having trouble finding out if it's truly VALID T-Sql:
UPDATE [Schema].[Table1]
SET [AColumn] =
(
SELECT [SomeColumn]
FROM [Schema].[Table2]
WHERE [Table2].[SameColumnName] = [Table1].[SameColumnName]
);
I have tested both and they work, so my question is, is the second completely valid and normal to use just the table name without the Schema in this sense or should I rather opt for the slightly more verbose Alias?
As I said in my comment, alias your objects.
SELECT MT.MyColumn,
YT.MyColumn
FROM dbo.MyTable MT
JOIN so.YourTable YT ON MT.ID = YT.fID
WHERE YT.[name] = N'Jane';
If you're performing an UPDATE, then specify the alias of the object to Update:
UPDATE MT
SET MyColumn = YT.MyColumn --Column on the left side of the SET will always reference the table being updated
FROM dbo.MyTable MT
JOIN so.YourTable YT ON MT.ID = YT.fID
WHERE YT.[name] = N'Jane';
I need to update a jsonb column which is called "verticals" and the array of values it holds are like HOM, BFB etc. There are no keys in the array.
Table: Product(verticals jsonb, code int)
sample value stored in "verticals" column is
[HOM,rst,NLF,WELSAK,HTL,TRV,EVCU,GRT]
I need to update the value 'HOM' to 'XXX' in the column "verticals" where code =1
My expected output is
[XXX,rst,NLF,WELSAK,HTL,TRV,EVCU,GRT]
Because you chose to store your data in a de-normalized way, updating it is more complicated then it has to be.
You need to first unnest the array (essentially normalizing the data), replace the values, then aggregate them back and update the column:
update product p
set verticals = t.verticals
from (
select jsonb_agg(case when x.v = 'HOM' then 'XXX' else x.v end order by idx) as verticals
from product p2, jsonb_array_elements_text(p2.verticals) with ordinality as x(v,idx)
where code = 1
) t
where p.code = t.code;
This assumes that product.code is a primary (or unique) key!
Online example: http://rextester.com/KZQ65481
If the order of the array elements is not important, this gets easier:
update product
set verticals = (verticals - 'HOM')||'["XXX"]'
where code = 1;
This removes the element 'HOM' from the array (regardless of the posisition) and then appends 'XXX' to the end of the array.
You should use jsonb_set(target jsonb, path text[], new_value jsonb[, create_missing boolean]) and array_position() OR array_replace(anyarray, anyelement, anyelement)
https://www.postgresql.org/docs/9.5/static/functions-json.html
https://www.postgresql.org/docs/10/static/functions-array.html
I have a query in postgres
insert into c_d (select * from cd where ak = '22019763');
And I get the following error
ERROR: column "region" is of type integer but expression is of type character varying
HINT: You will need to rewrite or cast the expression.
An INSERT INTO table1 SELECT * FROM table2 depends entirely on order of the columns, which is part of the table definition. It will line each column of table1 up with the column of table2 with the same order value, regardless of names.
The problem you have here is whatever column from cd with the same order value as c_d of the table "region" has an incompatible type, and an implicit typecast is not available to clear the confusion.
INSERT INTO SELECT * statements are stylistically bad form unless the two tables are defined, and will forever be defined, exactly the same way. All it takes is for a single extra column to get added to cd, and you'll start getting errors about extraneous extra columns.
If it is at all possible, what I would suggest is explicitly calling out the columns within the SELECT statement. You can call a function to change type within each of the column references (or you could define a new type cast to do this implicitly -- see CREATE CAST), and you can use AS to set the column label to match that of your target column.
If you can't do this for some reason, indicate that in your question.
Check out the PostgreSQL insert documentation. The syntax is:
INSERT INTO table [ ( column [, ...] ) ]
{ DEFAULT VALUES | VALUES ( { expression | DEFAULT } [, ...] ) | query }
which here would look something like:
INSERT INTO c_d (column1, column2...) select * from cd where ak = '22019763'
This is the syntax you want to use when inserting values from one table to another where the column types and order are not exactly the same.