How to define an array variable in Snowflake - arrays

How to define an array variable in snowflake worksheet?
set columns = (SELECT array_agg(COLUMN_NAME) FROM INFORMATION_SCHEMA.COLUMNS
where table_name='MEMBERS');
I get this error:
Unsupported feature 'assignment from non-constant source expression'.

2022 update:
Now it's possible with Snowflake Scripting:
declare
tmp_array ARRAY default ARRAY_CONSTRUCT();
rs_output RESULTSET;
begin
for i in 1 to 20 do
tmp_array := array_append(:tmp_array, OBJECT_CONSTRUCT('c1', 'a', 'c2', i));
end for;
rs_output := (select value:c1, value:c2 from table(flatten(:tmp_array)));
return table(rs_output);
end;
https://stackoverflow.com/a/71231621/132438
Previously:
Instead of storing an array, aggregate in a comma separated string:
set x = (SELECT listagg(COLUMN_NAME, ',') FROM INFORMATION_SCHEMA.COLUMNS WHERE COLUMN_NAME LIKE 'TABLE_S%');
However: "The size of string or binary variables is limited to 256 bytes" according to https://docs.snowflake.com/en/sql-reference/session-variables.html.
Which means that even if you could store an array in a variable, it would probably exceed the limits. Instead store the result in [temp] tables or so.

I found if you encapsulate with to_json it will work:
set x = (SELECT to_json(array_agg(COLUMN_NAME)) FROM INFORMATION_SCHEMA.COLUMNS
where table_name='CUSTOMER');
select $x;

Related

How to add new elements to an array of a user-defined type?

I have converted some stored procedure from Oracle to PostgreSQL but faced the below issue:
I have a user-defined type in PostgreSQL:
CREATE TYPE ut_merci_row AS (
SGLCNTNR varchar(100),
CODEST_MERCEVARIA varchar(50),
CODCICLO smallint
);
CREATE TYPE ut_merci_table AS (ut_merci_table UT_MERCI_ROW[]);
In Oracle:
retTable UT_MERCI_TABLE := UT_MERCI_TABLE();
.....
retTable.extend;
retTable(retTable.last) := UT_MERCI_ROW(rec.SGLCNTNR,rec.CODEST_MERCEVARIA,rec.CODCICLO);
Could you advise me how to convert the 2 last below code lines to PostgreSQL?
retTable.extend;
retTable(retTable.last) := UT_MERCI_ROW(rec.SGLCNTNR,rec.CODEST_MERCEVARIA,rec.CODCICLO);
Someone said that we don't need to "extend" the collection in PostgreSQL.
You can simply append the value to the array using the || concatenation operator. Make sure you initialize the variable to an empty array (otherwise it would null)
retTable ut_merci_row[] := '{}';
Then in the loop, just append a value of the type to the array:
retTable := rettable || row(rec.SGLCNTNR,rec.CODEST_MERCEVARIA,rec.CODCICLO)::ut_merci_row;
Trying to mimic Oracle's complicated way of returning tables with a cursor and custom types is not the right way to do this in Postgres.
But a much better solution would be to get rid of all that cursor processing and array handling and return a set of the base type:
CREATE OR REPLACE FUNCTION uf_getstatomerce1()
returns setof ut_merci_row
AS $body$
SELECT DISTINCT row(m.SGLCNTNR, m.CODEST_MERCEVARIA, 1 as CODCICLO)::ut_merci_row
from merce.DELIVERY_ORDER_RIF rif
inner join merce.DELIVERY_ORDER_PARTITA pm on rif.DO_ID = pm.DO_ID
inner join merce.vPARTITEMERCE m on pm.CODPARTITAMERCE = m.CODPARTITAMERCE;
$body$
LANGUAGE sql
STABLE;
In fact, in Postgres you also don't need a custom type for this, just declare the function as returns table and forget all the overhead:
CREATE OR REPLACE FUNCTION uf_getstatomerce1()
returns table (SGLCNTNR varchar, CODEST_MERCEVARIA varchar, CODCICLO smallint )
AS $body$
SELECT DISTINCT row(m.SGLCNTNR, m.CODEST_MERCEVARIA, 1 as CODCICLO)::ut_merci_row
from merce.DELIVERY_ORDER_RIF rif
inner join merce.DELIVERY_ORDER_PARTITA pm on rif.DO_ID = pm.DO_ID
inner join merce.vPARTITEMERCE m on pm.CODPARTITAMERCE = m.CODPARTITAMERCE;
$body$
LANGUAGE sql
STABLE;
The SQL function that simply returns a query will be much more efficient than the cursor loop.
In both cases, you can use it like this:
select *
from uf_getstatomerce1() ;
The second type definition is pretty redundant, just use ut_merci_row[].
You don't have to “extend” an array in PostgreSQL, just assign to an index that is not yet used.

Dynamic SQL operator IN with multiple parameters [duplicate]

How do I parameterize a query containing an IN clause with a variable number of arguments, like this one?
SELECT * FROM Tags
WHERE Name IN ('ruby','rails','scruffy','rubyonrails')
ORDER BY Count DESC
In this query, the number of arguments could be anywhere from 1 to 5.
I would prefer not to use a dedicated stored procedure for this (or XML), but if there is some elegant way specific to SQL Server 2008, I am open to that.
You can parameterize each value, so something like:
string[] tags = new string[] { "ruby", "rails", "scruffy", "rubyonrails" };
string cmdText = "SELECT * FROM Tags WHERE Name IN ({0})";
string[] paramNames = tags.Select(
(s, i) => "#tag" + i.ToString()
).ToArray();
string inClause = string.Join(", ", paramNames);
using (SqlCommand cmd = new SqlCommand(string.Format(cmdText, inClause))) {
for(int i = 0; i < paramNames.Length; i++) {
cmd.Parameters.AddWithValue(paramNames[i], tags[i]);
}
}
Which will give you:
cmd.CommandText = "SELECT * FROM Tags WHERE Name IN (#tag0, #tag1, #tag2, #tag3)"
cmd.Parameters["#tag0"] = "ruby"
cmd.Parameters["#tag1"] = "rails"
cmd.Parameters["#tag2"] = "scruffy"
cmd.Parameters["#tag3"] = "rubyonrails"
No, this is not open to SQL injection. The only injected text into CommandText is not based on user input. It's solely based on the hardcoded "#tag" prefix, and the index of an array. The index will always be an integer, is not user generated, and is safe.
The user inputted values are still stuffed into parameters, so there is no vulnerability there.
Edit:
Injection concerns aside, take care to note that constructing the command text to accomodate a variable number of parameters (as above) impede's SQL server's ability to take advantage of cached queries. The net result is that you almost certainly lose the value of using parameters in the first place (as opposed to merely inserting the predicate strings into the SQL itself).
Not that cached query plans aren't valuable, but IMO this query isn't nearly complicated enough to see much benefit from it. While the compilation costs may approach (or even exceed) the execution costs, you're still talking milliseconds.
If you have enough RAM, I'd expect SQL Server would probably cache a plan for the common counts of parameters as well. I suppose you could always add five parameters, and let the unspecified tags be NULL - the query plan should be the same, but it seems pretty ugly to me and I'm not sure that it'd worth the micro-optimization (although, on Stack Overflow - it may very well be worth it).
Also, SQL Server 7 and later will auto-parameterize queries, so using parameters isn't really necessary from a performance standpoint - it is, however, critical from a security standpoint - especially with user inputted data like this.
Here's a quick-and-dirty technique I have used:
SELECT * FROM Tags
WHERE '|ruby|rails|scruffy|rubyonrails|'
LIKE '%|' + Name + '|%'
So here's the C# code:
string[] tags = new string[] { "ruby", "rails", "scruffy", "rubyonrails" };
const string cmdText = "select * from tags where '|' + #tags + '|' like '%|' + Name + '|%'";
using (SqlCommand cmd = new SqlCommand(cmdText)) {
cmd.Parameters.AddWithValue("#tags", string.Join("|", tags);
}
Two caveats:
The performance is terrible. LIKE "%...%" queries are not indexed.
Make sure you don't have any |, blank, or null tags or this won't work
There are other ways to accomplish this that some people may consider cleaner, so please keep reading.
For SQL Server 2008, you can use a table valued parameter. It's a bit of work, but it is arguably cleaner than my other method.
First, you have to create a type
CREATE TYPE dbo.TagNamesTableType AS TABLE ( Name nvarchar(50) )
Then, your ADO.NET code looks like this:
string[] tags = new string[] { "ruby", "rails", "scruffy", "rubyonrails" };
cmd.CommandText = "SELECT Tags.* FROM Tags JOIN #tagNames as P ON Tags.Name = P.Name";
// value must be IEnumerable<SqlDataRecord>
cmd.Parameters.AddWithValue("#tagNames", tags.AsSqlDataRecord("Name")).SqlDbType = SqlDbType.Structured;
cmd.Parameters["#tagNames"].TypeName = "dbo.TagNamesTableType";
// Extension method for converting IEnumerable<string> to IEnumerable<SqlDataRecord>
public static IEnumerable<SqlDataRecord> AsSqlDataRecord(this IEnumerable<string> values, string columnName) {
if (values == null || !values.Any()) return null; // Annoying, but SqlClient wants null instead of 0 rows
var firstRecord = values.First();
var metadata= new SqlMetaData(columnName, SqlDbType.NVarChar, 50); //50 as per SQL Type
return values.Select(v =>
{
var r = new SqlDataRecord(metadata);
r.SetValues(v);
return r;
});
}
Update
As Per #Doug
Please try to avoid var metadata = SqlMetaData.InferFromValue(firstRecord, columnName);
It's set first value length, so if first value is 3 characters then its set max length 3 and other records will truncated if more then 3 characters.
So, please try to use: var metadata= new SqlMetaData(columnName, SqlDbType.NVarChar, maxLen);
Note: -1 for max length.
The original question was "How do I parameterize a query ..."
This is not an answer to that original question. There are some very good demonstrations of how to do that, in other answers.
See the first answer from Mark Brackett (the first answer starting "You can parameterize each value") and Mark Brackett's second answer for the preferred answer that I (and 231 others) upvoted. The approach given in his answer allows 1) for effective use of bind variables, and 2) for predicates that are sargable.
Selected answer
I am addressing here the approach given in Joel Spolsky's answer, the answer "selected" as the right answer.
Joel Spolsky's approach is clever. And it works reasonably, it's going to exhibit predictable behavior and predictable performance, given "normal" values, and with the normative edge cases, such as NULL and the empty string. And it may be sufficient for a particular application.
But in terms generalizing this approach, let's also consider the more obscure corner cases, like when the Name column contains a wildcard character (as recognized by the LIKE predicate.) The wildcard character I see most commonly used is % (a percent sign.). So let's deal with that here now, and later go on to other cases.
Some problems with % character
Consider a Name value of 'pe%ter'. (For the examples here, I use a literal string value in place of the column name.) A row with a Name value of `'pe%ter' would be returned by a query of the form:
select ...
where '|peanut|butter|' like '%|' + 'pe%ter' + '|%'
But that same row will not be returned if the order of the search terms is reversed:
select ...
where '|butter|peanut|' like '%|' + 'pe%ter' + '|%'
The behavior we observe is kind of odd. Changing the order of the search terms in the list changes the result set.
It almost goes without saying that we might not want pe%ter to match peanut butter, no matter how much he likes it.
Obscure corner case
(Yes, I will agree that this is an obscure case. Probably one that is not likely to be tested. We wouldn't expect a wildcard in a column value. We may assume that the application prevents such a value from being stored. But in my experience, I've rarely seen a database constraint that specifically disallowed characters or patterns that would be considered wildcards on the right side of a LIKE comparison operator.
Patching a hole
One approach to patching this hole is to escape the % wildcard character. (For anyone not familiar with the escape clause on the operator, here's a link to the SQL Server documentation.
select ...
where '|peanut|butter|'
like '%|' + 'pe\%ter' + '|%' escape '\'
Now we can match the literal %. Of course, when we have a column name, we're going to need to dynamically escape the wildcard. We can use the REPLACE function to find occurrences of the % character and insert a backslash character in front of each one, like this:
select ...
where '|pe%ter|'
like '%|' + REPLACE( 'pe%ter' ,'%','\%') + '|%' escape '\'
So that solves the problem with the % wildcard. Almost.
Escape the escape
We recognize that our solution has introduced another problem. The escape character. We see that we're also going to need to escape any occurrences of escape character itself. This time, we use the ! as the escape character:
select ...
where '|pe%t!r|'
like '%|' + REPLACE(REPLACE( 'pe%t!r' ,'!','!!'),'%','!%') + '|%' escape '!'
The underscore too
Now that we're on a roll, we can add another REPLACE handle the underscore wildcard. And just for fun, this time, we'll use $ as the escape character.
select ...
where '|p_%t!r|'
like '%|' + REPLACE(REPLACE(REPLACE( 'p_%t!r' ,'$','$$'),'%','$%'),'_','$_') + '|%' escape '$'
I prefer this approach to escaping because it works in Oracle and MySQL as well as SQL Server. (I usually use the \ backslash as the escape character, since that's the character we use in regular expressions. But why be constrained by convention!
Those pesky brackets
SQL Server also allows for wildcard characters to be treated as literals by enclosing them in brackets []. So we're not done fixing yet, at least for SQL Server. Since pairs of brackets have special meaning, we'll need to escape those as well. If we manage to properly escape the brackets, then at least we won't have to bother with the hyphen - and the carat ^ within the brackets. And we can leave any % and _ characters inside the brackets escaped, since we'll have basically disabled the special meaning of the brackets.
Finding matching pairs of brackets shouldn't be that hard. It's a little more difficult than handling the occurrences of singleton % and _. (Note that it's not sufficient to just escape all occurrences of brackets, because a singleton bracket is considered to be a literal, and doesn't need to be escaped. The logic is getting a little fuzzier than I can handle without running more test cases.)
Inline expression gets messy
That inline expression in the SQL is getting longer and uglier. We can probably make it work, but heaven help the poor soul that comes behind and has to decipher it. As much of a fan I am for inline expressions, I'm inclined not use one here, mainly because I don't want to have to leave a comment explaining the reason for the mess, and apologizing for it.
A function where ?
Okay, so, if we don't handle that as an inline expression in the SQL, the closest alternative we have is a user defined function. And we know that won't speed things up any (unless we can define an index on it, like we could with Oracle.) If we've got to create a function, we might better do that in the code that calls the SQL statement.
And that function may have some differences in behavior, dependent on the DBMS and version. (A shout out to all you Java developers so keen on being able to use any database engine interchangeably.)
Domain knowledge
We may have specialized knowledge of the domain for the column, (that is, the set of allowable values enforced for the column. We may know a priori that the values stored in the column will never contain a percent sign, an underscore, or bracket pairs. In that case, we just include a quick comment that those cases are covered.
The values stored in the column may allow for % or _ characters, but a constraint may require those values to be escaped, perhaps using a defined character, such that the values are LIKE comparison "safe". Again, a quick comment about the allowed set of values, and in particular which character is used as an escape character, and go with Joel Spolsky's approach.
But, absent the specialized knowledge and a guarantee, it's important for us to at least consider handling those obscure corner cases, and consider whether the behavior is reasonable and "per the specification".
Other issues recapitulated
I believe others have already sufficiently pointed out some of the other commonly considered areas of concern:
SQL injection (taking what would appear to be user supplied information, and including that in the SQL text rather than supplying them through bind variables. Using bind variables isn't required, it's just one convenient approach to thwart with SQL injection. There are other ways to deal with it:
optimizer plan using index scan rather than index seeks, possible need for an expression or function for escaping wildcards (possible index on expression or function)
using literal values in place of bind variables impacts scalability
Conclusion
I like Joel Spolsky's approach. It's clever. And it works.
But as soon as I saw it, I immediately saw a potential problem with it, and it's not my nature to let it slide. I don't mean to be critical of the efforts of others. I know many developers take their work very personally, because they invest so much into it and they care so much about it. So please understand, this is not a personal attack. What I'm identifying here is the type of problem that crops up in production rather than testing.
You can pass the parameter as a string
So you have the string
DECLARE #tags
SET #tags = ‘ruby|rails|scruffy|rubyonrails’
select * from Tags
where Name in (SELECT item from fnSplit(#tags, ‘|’))
order by Count desc
Then all you have to do is pass the string as 1 parameter.
Here is the split function I use.
CREATE FUNCTION [dbo].[fnSplit](
#sInputList VARCHAR(8000) -- List of delimited items
, #sDelimiter VARCHAR(8000) = ',' -- delimiter that separates items
) RETURNS #List TABLE (item VARCHAR(8000))
BEGIN
DECLARE #sItem VARCHAR(8000)
WHILE CHARINDEX(#sDelimiter,#sInputList,0) <> 0
BEGIN
SELECT
#sItem=RTRIM(LTRIM(SUBSTRING(#sInputList,1,CHARINDEX(#sDelimiter,#sInputList,0)-1))),
#sInputList=RTRIM(LTRIM(SUBSTRING(#sInputList,CHARINDEX(#sDelimiter,#sInputList,0)+LEN(#sDelimiter),LEN(#sInputList))))
IF LEN(#sItem) > 0
INSERT INTO #List SELECT #sItem
END
IF LEN(#sInputList) > 0
INSERT INTO #List SELECT #sInputList -- Put the last item in
RETURN
END
I heard Jeff/Joel talk about this on the podcast today (episode 34, 2008-12-16 (MP3, 31 MB), 1 h 03 min 38 secs - 1 h 06 min 45 secs), and I thought I recalled Stack Overflow was using LINQ to SQL, but maybe it was ditched. Here's the same thing in LINQ to SQL.
var inValues = new [] { "ruby","rails","scruffy","rubyonrails" };
var results = from tag in Tags
where inValues.Contains(tag.Name)
select tag;
That's it. And, yes, LINQ already looks backwards enough, but the Contains clause seems extra backwards to me. When I had to do a similar query for a project at work, I naturally tried to do this the wrong way by doing a join between the local array and the SQL Server table, figuring the LINQ to SQL translator would be smart enough to handle the translation somehow. It didn't, but it did provide an error message that was descriptive and pointed me towards using Contains.
Anyway, if you run this in the highly recommended LINQPad, and run this query, you can view the actual SQL that the SQL LINQ provider generated. It'll show you each of the values getting parameterized into an IN clause.
If you are calling from .NET, you could use Dapper dot net:
string[] names = new string[] {"ruby","rails","scruffy","rubyonrails"};
var tags = dataContext.Query<Tags>(#"
select * from Tags
where Name in #names
order by Count desc", new {names});
Here Dapper does the thinking, so you don't have to. Something similar is possible with LINQ to SQL, of course:
string[] names = new string[] {"ruby","rails","scruffy","rubyonrails"};
var tags = from tag in dataContext.Tags
where names.Contains(tag.Name)
orderby tag.Count descending
select tag;
In SQL Server 2016+ you could use STRING_SPLIT function:
DECLARE #names NVARCHAR(MAX) = 'ruby,rails,scruffy,rubyonrails';
SELECT *
FROM Tags
WHERE Name IN (SELECT [value] FROM STRING_SPLIT(#names, ','))
ORDER BY [Count] DESC;
or:
DECLARE #names NVARCHAR(MAX) = 'ruby,rails,scruffy,rubyonrails';
SELECT t.*
FROM Tags t
JOIN STRING_SPLIT(#names,',')
ON t.Name = [value]
ORDER BY [Count] DESC;
LiveDemo
The accepted answer will of course work and it is one of the way to go, but it is anti-pattern.
E. Find rows by list of values
This is replacement for common anti-pattern such as creating a dynamic SQL string in application layer or Transact-SQL, or by using LIKE operator:
SELECT ProductId, Name, Tags
FROM Product
WHERE ',1,2,3,' LIKE '%,' + CAST(ProductId AS VARCHAR(20)) + ',%';
Addendum:
To improve the STRING_SPLIT table function row estimation, it is a good idea to materialize splitted values as temporary table/table variable:
DECLARE #names NVARCHAR(MAX) = 'ruby,rails,scruffy,rubyonrails,sql';
CREATE TABLE #t(val NVARCHAR(120));
INSERT INTO #t(val) SELECT s.[value] FROM STRING_SPLIT(#names, ',') s;
SELECT *
FROM Tags tg
JOIN #t t
ON t.val = tg.TagName
ORDER BY [Count] DESC;
SEDE - Live Demo
Related: How to Pass a List of Values Into a Stored Procedure
Original question has requirement SQL Server 2008. Because this question is often used as duplicate, I've added this answer as reference.
This is possibly a half nasty way of doing it, I used it once, was rather effective.
Depending on your goals it might be of use.
Create a temp table with one column.
INSERT each look-up value into that column.
Instead of using an IN, you can then just use your standard JOIN rules. ( Flexibility++ )
This has a bit of added flexibility in what you can do, but it's more suited for situations where you have a large table to query, with good indexing, and you want to use the parametrized list more than once. Saves having to execute it twice and have all the sanitation done manually.
I never got around to profiling exactly how fast it was, but in my situation it was needed.
We have function that creates a table variable that you can join to:
ALTER FUNCTION [dbo].[Fn_sqllist_to_table](#list AS VARCHAR(8000),
#delim AS VARCHAR(10))
RETURNS #listTable TABLE(
Position INT,
Value VARCHAR(8000))
AS
BEGIN
DECLARE #myPos INT
SET #myPos = 1
WHILE Charindex(#delim, #list) > 0
BEGIN
INSERT INTO #listTable
(Position,Value)
VALUES (#myPos,LEFT(#list, Charindex(#delim, #list) - 1))
SET #myPos = #myPos + 1
IF Charindex(#delim, #list) = Len(#list)
INSERT INTO #listTable
(Position,Value)
VALUES (#myPos,'')
SET #list = RIGHT(#list, Len(#list) - Charindex(#delim, #list))
END
IF Len(#list) > 0
INSERT INTO #listTable
(Position,Value)
VALUES (#myPos,#list)
RETURN
END
So:
#Name varchar(8000) = null // parameter for search values
select * from Tags
where Name in (SELECT value From fn_sqllist_to_table(#Name,',')))
order by Count desc
This is gross, but if you are guaranteed to have at least one, you could do:
SELECT ...
...
WHERE tag IN( #tag1, ISNULL( #tag2, #tag1 ), ISNULL( #tag3, #tag1 ), etc. )
Having IN( 'tag1', 'tag2', 'tag1', 'tag1', 'tag1' ) will be easily optimized away by SQL Server. Plus, you get direct index seeks
I would pass a table type parameter (since it's SQL Server 2008), and do a where exists, or inner join. You may also use XML, using sp_xml_preparedocument, and then even index that temporary table.
In my opinion, the best source to solve this problem, is what has been posted on this site:
Syscomments. Dinakar Nethi
CREATE FUNCTION dbo.fnParseArray (#Array VARCHAR(1000),#separator CHAR(1))
RETURNS #T Table (col1 varchar(50))
AS
BEGIN
--DECLARE #T Table (col1 varchar(50))
-- #Array is the array we wish to parse
-- #Separator is the separator charactor such as a comma
DECLARE #separator_position INT -- This is used to locate each separator character
DECLARE #array_value VARCHAR(1000) -- this holds each array value as it is returned
-- For my loop to work I need an extra separator at the end. I always look to the
-- left of the separator character for each array value
SET #array = #array + #separator
-- Loop through the string searching for separtor characters
WHILE PATINDEX('%' + #separator + '%', #array) <> 0
BEGIN
-- patindex matches the a pattern against a string
SELECT #separator_position = PATINDEX('%' + #separator + '%',#array)
SELECT #array_value = LEFT(#array, #separator_position - 1)
-- This is where you process the values passed.
INSERT into #T VALUES (#array_value)
-- Replace this select statement with your processing
-- #array_value holds the value of this element of the array
-- This replaces what we just processed with and empty string
SELECT #array = STUFF(#array, 1, #separator_position, '')
END
RETURN
END
Use:
SELECT * FROM dbo.fnParseArray('a,b,c,d,e,f', ',')
CREDITS FOR: Dinakar Nethi
The proper way IMHO is to store the list in a character string (limited in length by what the DBMS support); the only trick is that (in order to simplify processing) I have a separator (a comma in my example) at the beginning and at the end of the string. The idea is to "normalize on the fly", turning the list into a one-column table that contains one row per value. This allows you to turn
in (ct1,ct2, ct3 ... ctn)
into an
in (select ...)
or (the solution I'd probably prefer) a regular join, if you just add a "distinct" to avoid problems with duplicate values in the list.
Unfortunately, the techniques to slice a string are fairly product-specific.
Here is the SQL Server version:
with qry(n, names) as
(select len(list.names) - len(replace(list.names, ',', '')) - 1 as n,
substring(list.names, 2, len(list.names)) as names
from (select ',Doc,Grumpy,Happy,Sneezy,Bashful,Sleepy,Dopey,' names) as list
union all
select (n - 1) as n,
substring(names, 1 + charindex(',', names), len(names)) as names
from qry
where n > 1)
select n, substring(names, 1, charindex(',', names) - 1) dwarf
from qry;
The Oracle version:
select n, substr(name, 1, instr(name, ',') - 1) dwarf
from (select n,
substr(val, 1 + instr(val, ',', 1, n)) name
from (select rownum as n,
list.val
from (select ',Doc,Grumpy,Happy,Sneezy,Bashful,Sleepy,Dopey,' val
from dual) list
connect by level < length(list.val) -
length(replace(list.val, ',', ''))));
and the MySQL version:
select pivot.n,
substring_index(substring_index(list.val, ',', 1 + pivot.n), ',', -1) from (select 1 as n
union all
select 2 as n
union all
select 3 as n
union all
select 4 as n
union all
select 5 as n
union all
select 6 as n
union all
select 7 as n
union all
select 8 as n
union all
select 9 as n
union all
select 10 as n) pivot, (select ',Doc,Grumpy,Happy,Sneezy,Bashful,Sleepy,Dopey,' val) as list where pivot.n < length(list.val) -
length(replace(list.val, ',', ''));
(Of course, "pivot" must return as many rows as the maximum number of
items we can find in the list)
If you've got SQL Server 2008 or later I'd use a Table Valued Parameter.
If you're unlucky enough to be stuck on SQL Server 2005 you could add a CLR function like this,
[SqlFunction(
DataAccessKind.None,
IsDeterministic = true,
SystemDataAccess = SystemDataAccessKind.None,
IsPrecise = true,
FillRowMethodName = "SplitFillRow",
TableDefinintion = "s NVARCHAR(MAX)"]
public static IEnumerable Split(SqlChars seperator, SqlString s)
{
if (s.IsNull)
return new string[0];
return s.ToString().Split(seperator.Buffer);
}
public static void SplitFillRow(object row, out SqlString s)
{
s = new SqlString(row.ToString());
}
Which you could use like this,
declare #desiredTags nvarchar(MAX);
set #desiredTags = 'ruby,rails,scruffy,rubyonrails';
select * from Tags
where Name in [dbo].[Split] (',', #desiredTags)
order by Count desc
I think this is a case when a static query is just not the way to go. Dynamically build the list for your in clause, escape your single quotes, and dynamically build SQL. In this case you probably won't see much of a difference with any method due to the small list, but the most efficient method really is to send the SQL exactly as it is written in your post. I think it is a good habit to write it the most efficient way, rather than to do what makes the prettiest code, or consider it bad practice to dynamically build SQL.
I have seen the split functions take longer to execute than the query themselves in many cases where the parameters get large. A stored procedure with table valued parameters in SQL 2008 is the only other option I would consider, although this will probably be slower in your case. TVP will probably only be faster for large lists if you are searching on the primary key of the TVP, because SQL will build a temporary table for the list anyway (if the list is large). You won't know for sure unless you test it.
I have also seen stored procedures that had 500 parameters with default values of null, and having WHERE Column1 IN (#Param1, #Param2, #Param3, ..., #Param500). This caused SQL to build a temp table, do a sort/distinct, and then do a table scan instead of an index seek. That is essentially what you would be doing by parameterizing that query, although on a small enough scale that it won't make a noticeable difference. I highly recommend against having NULL in your IN lists, as if that gets changed to a NOT IN it will not act as intended. You could dynamically build the parameter list, but the only obvious thing that you would gain is that the objects would escape the single quotes for you. That approach is also slightly slower on the application end since the objects have to parse the query to find the parameters. It may or may not be faster on SQL, as parameterized queries call sp_prepare, sp_execute for as many times you execute the query, followed by sp_unprepare.
The reuse of execution plans for stored procedures or parameterized queries may give you a performance gain, but it will lock you in to one execution plan determined by the first query that is executed. That may be less than ideal for subsequent queries in many cases. In your case, reuse of execution plans will probably be a plus, but it might not make any difference at all as the example is a really simple query.
Cliffs notes:
For your case anything you do, be it parameterization with a fixed number of items in the list (null if not used), dynamically building the query with or without parameters, or using stored procedures with table valued parameters will not make much of a difference. However, my general recommendations are as follows:
Your case/simple queries with few parameters:
Dynamic SQL, maybe with parameters if testing shows better performance.
Queries with reusable execution plans, called multiple times by simply changing the parameters or if the query is complicated:
SQL with dynamic parameters.
Queries with large lists:
Stored procedure with table valued parameters. If the list can vary by a large amount use WITH RECOMPILE on the stored procedure, or simply use dynamic SQL without parameters to generate a new execution plan for each query.
May be we can use XML here:
declare #x xml
set #x='<items>
<item myvalue="29790" />
<item myvalue="31250" />
</items>
';
With CTE AS (
SELECT
x.item.value('#myvalue[1]', 'decimal') AS myvalue
FROM #x.nodes('//items/item') AS x(item) )
select * from YourTable where tableColumnName in (select myvalue from cte)
If we have strings stored inside the IN clause with the comma(,) delimited, we can use the charindex function to get the values. If you use .NET, then you can map with SqlParameters.
DDL Script:
CREATE TABLE Tags
([ID] int, [Name] varchar(20))
;
INSERT INTO Tags
([ID], [Name])
VALUES
(1, 'ruby'),
(2, 'rails'),
(3, 'scruffy'),
(4, 'rubyonrails')
;
T-SQL:
DECLARE #Param nvarchar(max)
SET #Param = 'ruby,rails,scruffy,rubyonrails'
SELECT * FROM Tags
WHERE CharIndex(Name,#Param)>0
You can use the above statement in your .NET code and map the parameter with SqlParameter.
Fiddler demo
EDIT:
Create the table called SelectedTags using the following script.
DDL Script:
Create table SelectedTags
(Name nvarchar(20));
INSERT INTO SelectedTags values ('ruby'),('rails')
T-SQL:
DECLARE #list nvarchar(max)
SELECT #list=coalesce(#list+',','')+st.Name FROM SelectedTags st
SELECT * FROM Tags
WHERE CharIndex(Name,#Param)>0
I'd approach this by default with passing a table valued function (that returns a table from a string) to the IN condition.
Here is the code for the UDF (I got it from Stack Overflow somewhere, i can't find the source right now)
CREATE FUNCTION [dbo].[Split] (#sep char(1), #s varchar(8000))
RETURNS table
AS
RETURN (
WITH Pieces(pn, start, stop) AS (
SELECT 1, 1, CHARINDEX(#sep, #s)
UNION ALL
SELECT pn + 1, stop + 1, CHARINDEX(#sep, #s, stop + 1)
FROM Pieces
WHERE stop > 0
)
SELECT
SUBSTRING(#s, start, CASE WHEN stop > 0 THEN stop-start ELSE 512 END) AS s
FROM Pieces
)
Once you got this your code would be as simple as this:
select * from Tags
where Name in (select s from dbo.split(';','ruby;rails;scruffy;rubyonrails'))
order by Count desc
Unless you have a ridiculously long string, this should work well with the table index.
If needed you can insert it into a temp table, index it, then run a join...
For a variable number of arguments like this the only way I'm aware of is to either generate the SQL explicitly or do something that involves populating a temporary table with the items you want and joining against the temp table.
Another possible solution is instead of passing a variable number of arguments to a stored procedure, pass a single string containing the names you're after, but make them unique by surrounding them with '<>'. Then use PATINDEX to find the names:
SELECT *
FROM Tags
WHERE PATINDEX('%<' + Name + '>%','<jo>,<john>,<scruffy>,<rubyonrails>') > 0
Use the following stored procedure. It uses a custom split function, which can be found here.
create stored procedure GetSearchMachingTagNames
#PipeDelimitedTagNames varchar(max),
#delimiter char(1)
as
begin
select * from Tags
where Name in (select data from [dbo].[Split](#PipeDelimitedTagNames,#delimiter)
end
Here is another alternative. Just pass a comma-delimited list as a string parameter to the stored procedure and:
CREATE PROCEDURE [dbo].[sp_myproc]
#UnitList varchar(MAX) = '1,2,3'
AS
select column from table
where ph.UnitID in (select * from CsvToInt(#UnitList))
And the function:
CREATE Function [dbo].[CsvToInt] ( #Array varchar(MAX))
returns #IntTable table
(IntValue int)
AS
begin
declare #separator char(1)
set #separator = ','
declare #separator_position int
declare #array_value varchar(MAX)
set #array = #array + ','
while patindex('%,%' , #array) <> 0
begin
select #separator_position = patindex('%,%' , #array)
select #array_value = left(#array, #separator_position - 1)
Insert #IntTable
Values (Cast(#array_value as int))
select #array = stuff(#array, 1, #separator_position, '')
end
return
end
In ColdFusion we just do:
<cfset myvalues = "ruby|rails|scruffy|rubyonrails">
<cfquery name="q">
select * from sometable where values in <cfqueryparam value="#myvalues#" list="true">
</cfquery>
Here's a technique that recreates a local table to be used in a query string. Doing it this way eliminates all parsing problems.
The string can be built in any language. In this example I used SQL since that was the original problem I was trying to solve. I needed a clean way to pass in table data on the fly in a string to be executed later.
Using a user defined type is optional. Creating the type is only created once and can be done ahead of time. Otherwise just add a full table type to the declaration in the string.
The general pattern is easy to extend and can be used for passing more complex tables.
-- Create a user defined type for the list.
CREATE TYPE [dbo].[StringList] AS TABLE(
[StringValue] [nvarchar](max) NOT NULL
)
-- Create a sample list using the list table type.
DECLARE #list [dbo].[StringList];
INSERT INTO #list VALUES ('one'), ('two'), ('three'), ('four')
-- Build a string in which we recreate the list so we can pass it to exec
-- This can be done in any language since we're just building a string.
DECLARE #str nvarchar(max);
SET #str = 'DECLARE #list [dbo].[StringList]; INSERT INTO #list VALUES '
-- Add all the values we want to the string. This would be a loop in C++.
SELECT #str = #str + '(''' + StringValue + '''),' FROM #list
-- Remove the trailing comma so the query is valid sql.
SET #str = substring(#str, 1, len(#str)-1)
-- Add a select to test the string.
SET #str = #str + '; SELECT * FROM #list;'
-- Execute the string and see we've pass the table correctly.
EXEC(#str)
In SQL Server 2016+ another possibility is to use the OPENJSON function.
This approach is blogged about in OPENJSON - one of best ways to select rows by list of ids.
A full worked example below
CREATE TABLE dbo.Tags
(
Name VARCHAR(50),
Count INT
)
INSERT INTO dbo.Tags
VALUES ('VB',982), ('ruby',1306), ('rails',1478), ('scruffy',1), ('C#',1784)
GO
CREATE PROC dbo.SomeProc
#Tags VARCHAR(MAX)
AS
SELECT T.*
FROM dbo.Tags T
WHERE T.Name IN (SELECT J.Value COLLATE Latin1_General_CI_AS
FROM OPENJSON(CONCAT('[', #Tags, ']')) J)
ORDER BY T.Count DESC
GO
EXEC dbo.SomeProc #Tags = '"ruby","rails","scruffy","rubyonrails"'
DROP TABLE dbo.Tags
I have an answer that doesn't require a UDF, XML
Because IN accepts a select statement
e.g. SELECT * FROM Test where Data IN (SELECT Value FROM TABLE)
You really only need a way to convert the string into a table.
This can be done with a recursive CTE, or a query with a number table (or Master..spt_value)
Here's the CTE version.
DECLARE #InputString varchar(8000) = 'ruby,rails,scruffy,rubyonrails'
SELECT #InputString = #InputString + ','
;WITH RecursiveCSV(x,y)
AS
(
SELECT
x = SUBSTRING(#InputString,0,CHARINDEX(',',#InputString,0)),
y = SUBSTRING(#InputString,CHARINDEX(',',#InputString,0)+1,LEN(#InputString))
UNION ALL
SELECT
x = SUBSTRING(y,0,CHARINDEX(',',y,0)),
y = SUBSTRING(y,CHARINDEX(',',y,0)+1,LEN(y))
FROM
RecursiveCSV
WHERE
SUBSTRING(y,CHARINDEX(',',y,0)+1,LEN(y)) <> '' OR
SUBSTRING(y,0,CHARINDEX(',',y,0)) <> ''
)
SELECT
*
FROM
Tags
WHERE
Name IN (select x FROM RecursiveCSV)
OPTION (MAXRECURSION 32767);
I use a more concise version of the top voted answer:
List<SqlParameter> parameters = tags.Select((s, i) => new SqlParameter("#tag" + i.ToString(), SqlDbType.NVarChar(50)) { Value = s}).ToList();
var whereCondition = string.Format("tags in ({0})", String.Join(",",parameters.Select(s => s.ParameterName)));
It does loop through the tag parameters twice; but that doesn't matter most of the time (it won't be your bottleneck; if it is, unroll the loop).
If you're really interested in performance and don't want to iterate through the loop twice, here's a less beautiful version:
var parameters = new List<SqlParameter>();
var paramNames = new List<string>();
for (var i = 0; i < tags.Length; i++)
{
var paramName = "#tag" + i;
//Include size and set value explicitly (not AddWithValue)
//Because SQL Server may use an implicit conversion if it doesn't know
//the actual size.
var p = new SqlParameter(paramName, SqlDbType.NVarChar(50) { Value = tags[i]; }
paramNames.Add(paramName);
parameters.Add(p);
}
var inClause = string.Join(",", paramNames);
Here is another answer to this problem.
(new version posted on 6/4/13).
private static DataSet GetDataSet(SqlConnectionStringBuilder scsb, string strSql, params object[] pars)
{
var ds = new DataSet();
using (var sqlConn = new SqlConnection(scsb.ConnectionString))
{
var sqlParameters = new List<SqlParameter>();
var replacementStrings = new Dictionary<string, string>();
if (pars != null)
{
for (int i = 0; i < pars.Length; i++)
{
if (pars[i] is IEnumerable<object>)
{
List<object> enumerable = (pars[i] as IEnumerable<object>).ToList();
replacementStrings.Add("#" + i, String.Join(",", enumerable.Select((value, pos) => String.Format("#_{0}_{1}", i, pos))));
sqlParameters.AddRange(enumerable.Select((value, pos) => new SqlParameter(String.Format("#_{0}_{1}", i, pos), value ?? DBNull.Value)).ToArray());
}
else
{
sqlParameters.Add(new SqlParameter(String.Format("#{0}", i), pars[i] ?? DBNull.Value));
}
}
}
strSql = replacementStrings.Aggregate(strSql, (current, replacementString) => current.Replace(replacementString.Key, replacementString.Value));
using (var sqlCommand = new SqlCommand(strSql, sqlConn))
{
if (pars != null)
{
sqlCommand.Parameters.AddRange(sqlParameters.ToArray());
}
else
{
//Fail-safe, just in case a user intends to pass a single null parameter
sqlCommand.Parameters.Add(new SqlParameter("#0", DBNull.Value));
}
using (var sqlDataAdapter = new SqlDataAdapter(sqlCommand))
{
sqlDataAdapter.Fill(ds);
}
}
}
return ds;
}
Cheers.
The only winning move is not to play.
No infinite variability for you. Only finite variability.
In the SQL you have a clause like this:
and ( {1}==0 or b.CompanyId in ({2},{3},{4},{5},{6}) )
In the C# code you do something like this:
int origCount = idList.Count;
if (origCount > 5) {
throw new Exception("You may only specify up to five originators to filter on.");
}
while (idList.Count < 5) { idList.Add(-1); } // -1 is an impossible value
return ExecuteQuery<PublishDate>(getValuesInListSQL,
origCount,
idList[0], idList[1], idList[2], idList[3], idList[4]);
So basically if the count is 0 then there is no filter and everything goes through. If the count is higher than 0 the then the value must be in the list, but the list has been padded out to five with impossible values (so that the SQL still makes sense)
Sometimes the lame solution is the only one that actually works.

How do I turn an array into a record / tuple / row type?

I need to execute a dynamic insert into a table with a variable number of columns.
Right now I'm quoting both the column names, with quote_ident, and the actual values, with quote_nullable, and then joining them with array_to_string:
for ... loop
...
cols := array_append(cols, quote_ident(column_name));
vals := array_append(vals, quote_nullable(column_value));
end loop;
execute format('insert into %s (%s) values (%s)',
target_table,
array_to_string(cols, ', ')
array_to_string(vals, ', ')
);
It's a pattern found all over the place, including on the official documentation. But it feels a bit unclean. I'd rather pass the array of values a parameter in the using clause:
execute format('insert into %s (%s) ... $1 ...',
target_table,
array_to_string(cols, ', ')
)
using vals;
Notice the using vals, which is what I'd like to achieve. But I cannot seem to be able to fill in the dots in the insert statement. Maybe some kind of select ... from ...?
More generally, how do I turn an array into a record / tuple / row type?
I'm using code very like this in production:
EXECUTE ( SELECT 'insert into sometable '
'('||string_agg(quote_ident(p.key),',') ||
') values ('|| string_agg(quote_nullable(p.value),',') || ');'
FROM each(payload) as p
);
but here payload is a hstore, not a pair of arrays.

Remove duplicates from associative array with multiple columns in PL/SQL

I have seen several great examples of removing duplicates in an Oracle associative array using MULTISET UNION DISTINCT and SET. It works great when there is only a single column.
I have an associative array based on a RECORD type which contains 3 columns.
Is it possible to use the methods mentioned above?
DECLARE
TYPE rec_type IS RECORD(
column1 VARCHAR2(5)
,column2 VARCHAR2(5));
TYPE my_aa IS TABLE OF rec_type;
p_tbl my_aa := my_aa();
q_tbl my_aa := my_aa();
BEGIN
p_tbl.extend(4);
p_tbl(1).column1 := 'A1';
p_tbl(1).column2 := 'a';
--
p_tbl(2).column1 := 'A1';
p_tbl(2).column2 := 'b';
--
p_tbl(3).column1 := 'A1'; -- Dup
p_tbl(3).column2 := 'a'; -- Dup
--
p_tbl(4).column1 := 'A1';
p_tbl(4).column2 := 'c';
--
dbms_output.put_line('-- First output contains duplicated');
--
FOR a IN p_tbl.first .. p_tbl.last LOOP
dbms_output.put_line(a || ' = ' || p_tbl(a).column1 || ' / ' || p_tbl(a).column2);
END LOOP;
--
--
q_tbl := p_tbl MULTISET UNION DISTINCT p_tbl;
--
--
dbms_output.new_line;
dbms_output.put_line('-- Duplicates have been removad');
FOR a IN q_tbl.first .. q_tbl.last LOOP
dbms_output.put_line(a || ' = ' || q_tbl(a).column1|| ' / '||q_tbl(a).column2);
END LOOP;
END;
It is impossible to do distinction by MULTISET UNION DISTINCT with underlying RECORD type due to doc
The element types of the nested tables must be comparable. Please
refer to "Comparison Conditions " for information on the comparability
of nonscalar types.
https://docs.oracle.com/cd/B12037_01/server.101/b10759/operators006.htm
Records are not comparable. You can use user defined objects with MAP method for this
Two objects of nonscalar type are comparable if they are of the same
named type and there is a one-to-one correspondence between their
elements. In addition, nested tables of user-defined object types,
even if their elements are comparable, must have MAP methods defined
on them to be used in equality or IN conditions.
https://docs.oracle.com/cd/B12037_01/server.101/b10759/conditions002.htm#i1033286
If you still whant do this with records then try SELECT DISTINCT ... FROM TABLE() approach, but your types must be defined on schema level for server older than oracle 12c.

How to replace a value in array

I have a data like below.
id col1[]
--- ------
1 {1,2,3}
2 {3,4,5}
My question is how to use replace function in arrays.
select array_replace(col1, 1, 100) where id = 1;
but it gives an error like:
function array_replace(integer[], integer, integer) does not exist
can anyone suggest how to use it?
Your statement (augmented with the missing FROM clause):
SELECT array_replace(col1, 1, 100) FROM tbl WHERE id = 1;
As commented by #mu, array_replace() was introduced with pg 9.3. I see 3 options for older versions:
1. intarray
As long as ...
we are dealing with integer arrays.
elements are unique.
and the order of elements is irrelevant.
A simple and fast option would be to install the additional module intarray, which (among other things) provides operators to subtract and add elements (or whole arrays) from/to integer arrays:
SELECT CASE col1 && '{1}'::int[] THEN (col1 - 1) + 100 ELSE col1 END AS col1
FROM tbl WHERE id = 1;
2. Emulate with SQL functions
A (slower) drop-in replacement for array_replace() using polymorphic types, so it works for any base type:
CREATE OR REPLACE FUNCTION f_array_replace(anyarray, anyelement, anyelement)
RETURNS anyarray LANGUAGE SQL IMMUTABLE AS
'SELECT ARRAY (SELECT CASE WHEN x = $2 THEN $3 ELSE x END FROM unnest($1) x)';
Does not replace NULL values. Related:
Replace NULL values in an array in PostgreSQL
If you need to guarantee order of elements:
PostgreSQL unnest() with element number
3. Apply patch to source and recompile
Get the patch "Add array_remove() and array_replace() functions" from the git repo, apply it to the source of your version and recompile. May or may not apply cleanly. The older your version the worse are your chances. I have not tried that, I would rather upgrade to the current version.
You can create your own
based on this source :
CREATE TABLE arr(id int, col1 int[]);
INSERT INTO arr VALUES (1, '{1,2,3}');
INSERT INTO arr VALUES (2, '{3,4,5}');
SELECT array(
SELECT CASE WHEN q = 1 THEN 100 ELSE q END
FROM UNNEST(col1::int[]) q)
FROM arr;
array
-----------
{100,2,3}
{3,4,5}
You can create your own function and put it in your public schema if you still want to call by function though it will be slightly different than the original.
UPDATE tbl SET col1 = array_replace(col1, 1, 100) WHERE id = 1;
Here is the sample query for a test-array:
SELECT test_id,
test_array,
(array (
-- Replace existing value 'int' of an array with given value 'Text'
SELECT CASE WHEN a = '0' THEN 'MyEntry'
WHEN a = '1' THEN 'Apple'
WHEN a = '2' THEN 'Banana'
WHEN a = '3' THEN 'ChErRiEs'
WHEN a = '4' THEN 'Dragon Fruit'
WHEN a = '5' THEN 'Eat a Fruit in a Day'
ELSE 'NONE' END
FROM UNNEST(test_array::TEXT[]) a) ::TEXT
-- UNNEST : Lists out values of my_test_array
) test_result
FROM (
--my_test_array
SELECT 1 test_id, '{0,1,2,3,4,5,6,7,8,9}'::TEXT[][] test_array
) test;

Resources