I'd like to get value by the following SQL using Eloquent ORM.
- SQL
SELECT COUNT(*) FROM
(SELECT * FROM abc GROUP BY col1) AS a;
Then I considered the following.
- Code
$sql = Abc::from('abc AS a')->groupBy('col1')->toSql();
$num = Abc::from(\DB::raw($sql))->count();
print $num;
I'm looking for a better solution.
Please tell me simplest solution.
In addition to #delmadord's answer and your comments:
Currently there is no method to create subquery in FROM clause, so you need to manually use raw statement, then, if necessary, you will merge all the bindings:
$sub = Abc::where(..)->groupBy(..); // Eloquent Builder instance
$count = DB::table( DB::raw("({$sub->toSql()}) as sub") )
->mergeBindings($sub->getQuery()) // you need to get underlying Query Builder
->count();
Mind that you need to merge bindings in correct order. If you have other bound clauses, you must put them after mergeBindings:
$count = DB::table( DB::raw("({$sub->toSql()}) as sub") )
// ->where(..) wrong
->mergeBindings($sub->getQuery()) // you need to get underlying Query Builder
// ->where(..) correct
->count();
Laravel v5.6.12 (2018-03-14) added fromSub() and fromRaw() methods to query builder (#23476).
The accepted answer is correct but can be simplified into:
DB::query()->fromSub(function ($query) {
$query->from('abc')->groupBy('col1');
}, 'a')->count();
The above snippet produces the following SQL:
select count(*) as aggregate from (select * from `abc` group by `col1`) as `a`
The solution of #JarekTkaczyk it is exactly what I was looking for. The only thing I miss is how to do it when you are using
DB::table() queries. In this case, this is how I do it:
$other = DB::table( DB::raw("({$sub->toSql()}) as sub") )->select(
'something',
DB::raw('sum( qty ) as qty'),
'foo',
'bar'
);
$other->mergeBindings( $sub );
$other->groupBy('something');
$other->groupBy('foo');
$other->groupBy('bar');
print $other->toSql();
$other->get();
Special atention how to make the mergeBindings without using the getQuery() method
From laravel 5.5 there is a dedicated method for subqueries and you can use it like this:
Abc::selectSub(function($q) {
$q->select('*')->groupBy('col1');
}, 'a')->count('a.*');
or
Abc::selectSub(Abc::select('*')->groupBy('col1'), 'a')->count('a.*');
There are many readable ways to do these kinds of queries at the moment (Laravel 8).
// option 1: DB::table(Closure, alias) for subquery
$count = DB::table(function ($sub) {
$sub->from('abc')
->groupBy('col1');
}, 'a')
->count();
// option 2: DB::table(Builder, alias) for subquery
$sub = DB::table('abc')->groupBy('col1');
$count = DB::table($sub, 'a')->count();
// option 3: DB::query()->from(Closure, alias)
$count = DB::query()
->from(function ($sub) {
$sub->from('abc')
->groupBy('col1')
}, 'a')
->count();
// option 4: DB::query()->from(Builder, alias)
$sub = DB::table('abc')->groupBy('col1');
$count = DB::query()->from($sub, 'a')->count();
For such small subqueries, you could even try fitting them in a single line with PHP 7.4's short closures but this approach can be harder to mantain.
$count = DB::table(fn($sub) => $sub->from('abc')->groupBy('col1'), 'a')->count();
Note that I'm using count() instead of explicitly writing the count(*) statement and using get() or first() for the results (which you can easily do by replacing count() with selectRaw(count(*))->first()).
The reason for this is simple: It returns the number instead of an object with an awkwardly named property (count(*) unless you used an alias in the query)
Which looks better?
// using count() in the builder
echo $count;
// using selectRaw('count(*)')->first() in the builder
echo $count->{'count(*)'};
Correct way described in this answer: https://stackoverflow.com/a/52772444/2519714
Most popular answer at current moment is not totally correct.
This way https://stackoverflow.com/a/24838367/2519714 is not correct in some cases like: sub select has where bindings, then joining table to sub select, then other wheres added to all query. For example query:
select * from (select * from t1 where col1 = ?) join t2 on col1 = col2 and col3 = ? where t2.col4 = ?
To make this query you will write code like:
$subQuery = DB::query()->from('t1')->where('t1.col1', 'val1');
$query = DB::query()->from(DB::raw('('. $subQuery->toSql() . ') AS subquery'))
->mergeBindings($subQuery->getBindings());
$query->join('t2', function(JoinClause $join) {
$join->on('subquery.col1', 't2.col2');
$join->where('t2.col3', 'val3');
})->where('t2.col4', 'val4');
During executing this query, his method $query->getBindings() will return bindings in incorrect order like ['val3', 'val1', 'val4'] in this case instead correct ['val1', 'val3', 'val4'] for raw sql described above.
One more time correct way to do this:
$subQuery = DB::query()->from('t1')->where('t1.col1', 'val1');
$query = DB::query()->fromSub($subQuery, 'subquery');
$query->join('t2', function(JoinClause $join) {
$join->on('subquery.col1', 't2.col2');
$join->where('t2.col3', 'val3');
})->where('t2.col4', 'val4');
Also bindings will be automatically and correctly merged to new query.
I like doing something like this:
Message::select('*')
->from(DB::raw("( SELECT * FROM `messages`
WHERE `to_id` = ".Auth::id()." AND `isseen` = 0
GROUP BY `from_id` asc) as `sub`"))
->count();
It's not very elegant, but it's simple.
This works fine
$q1 = DB::table('tableA')->groupBy('col');
$data = DB::table(DB::raw("({$q1->toSql()}) as sub"))->mergeBindings($q1)->get();
I could not made your code to do the desired query, the AS is an alias only for the table abc, not for the derived table.
Laravel Query Builder does not implicitly support derived table aliases, DB::raw is most likely needed for this.
The most straight solution I could came up with is almost identical to yours, however produces the query as you asked for:
$sql = Abc::groupBy('col1')->toSql();
$count = DB::table(DB::raw("($sql) AS a"))->count();
The produced query is
select count(*) as aggregate from (select * from `abc` group by `col1`) AS a;
->selectRaw('your subquery as somefield')
Deriving off mpskovvang's answer, here is what it would look like using eloquent model. (I tried updating mpskovvang answer to include this, but there's too many edit requests for it.)
$qry = Abc::where('col2', 'value')->groupBy('col1')->selectRaw('1');
$num = Abc::from($qry, 'q1')->count();
print $num;
Produces...
SELECT COUNT(*) as aggregate FROM (SELECT 1 FROM Abc WHERE col2='value' GROUP BY col1) as q1
I have used too much time (days) on this and I really hope someone can help me out.
I found a good article on describing my problem in a generic way so let's stick to it.
I am trying to build this query but NHibernate fails to build the correct sql and returns a sql query exception.
Column vSagsAendring.Id is invalid in the select list because it is not contained in either an aggregate function or the GROUP BY clause. It could not execute the following query:
select
viewsagsae0_.Id as Id155_,
viewsagsae0_.SagId as SagId155_,
viewsagsae0_.JournalNr as JournalNr155_,
viewsagsae0_.LbfNr as LbfNr155_,
viewsagsae0_.OrgNr as OrgNr155_,
viewsagsae0_.OrgNavn as OrgNavn155_,
viewsagsae0_.AfdNavn as AfdNavn155_,
viewsagsae0_.SagsType as SagsType155_,
viewsagsae0_.Status as Status155_,
viewsagsae0_.SagsbehandlerInit as Sagsbeh10_155_,
viewsagsae0_.Dato as Dato155_,
viewsagsae0_.JournalAktionType as Journal12_155_,
viewsagsae0_.Beskrivelse as Beskriv13_155_,
viewsagsae0_.Ekstern as Ekstern155_
from vSagsAendring viewsagsae0_
group by viewsagsae0_.SagId
var query = from p in _session.Query<ViewSagsAendring>()
group p by p.SagId
into grp
select grp.OrderByDescending(g => g.Dato).First();
This is another version also took from the article:
var query = from p in _session.Query<ViewSagsAendring>()
group p by p.SagId
into grp
let maxDato = grp.Max(g => g.Dato)
from p in grp
where p.Dato == maxDato
select p;
It's have been a long journey, but now it's over. I hope that I can help someone else in the same situation by answering my own question.
var aendring = from sagsAendring in _session.Query<ViewSagsAendring>()
where sagsAendring.Dato ==
(
from innersagsAendring in _session.Query<ViewSagsAendring>()
where innersagsAendring.SagId == sagsAendring.SagId
select innersagsAendring.Dato
).Max()
select sagsAendring;
var result = aendring.ToList();
And because you can chain linq statements you can build a linq filter like this
if(Filters.VisInterneAendringer == false)
query = query.Where(x => x.Ekstern == true);
if (Filters.VisKunNyesteAendringer)
{
query = query.Where(sagsAendring => sagsAendring.Dato ==
(
from innerSagsAendring in Session.Query<ViewSagsAendring>() where innerSagsAendring.SagId == sagsAendring.SagId
select innerSagsAendring.Dato
).Max());
}
return query;
Your queries seem legit for LINQ in EntityFramework.
I'm not sure about hibernate, you might try to use QueryOver API instead of Query
http://nhibernate.info/blog/2009/12/17/queryover-in-nh-3-0.html
I am trying to work out how to covert the script below from SQL in to LINQ. Any help would be welcome.
SELECT *
FROM
[tableName]
WHERE
[MyDate] IN
(SELECT
MAX([MyDate])
FROM
[tableName]
GROUP BY
[MyID])
I can't find an equivalent for the "IN" clause section. There are existing questions on this forum but none that cover selecting a DateTime.
Thanks in advance.
You can use the ".Contains(..)" function:
e.g.
var itemQuery = from cartItems in db.SalesOrderDetails
where cartItems.SalesOrderID == 75144
select cartItems.ProductID;
var myProducts = from p in db.Products
where itemQuery.Contains(p.ProductID)
select p;
Although it looks like 2 round trips, as the LINQ only constructs the query when the IEnumerable is tripped, you should get reasonable performance.
I think Any() is what you are looking for:
var result = tableName.Where(x =>
(from t in tableName
group t by t.MyID into g
where g.Max(y => y.MyDate) == x.MyDate
select 1).Any())
I'm currently struggling on how to execute my query on a Table object in Zend and get a Rowset in return. Reason I need particularly THIS is because I'm modifying a code for existing project and I don't have much flexibility.
Query:
SELECT *
FROM `tblname` ud
WHERE ud.user_id = some_id
AND
(
(ud.reputation_level > 1)
OR
(
(SELECT COUNT( * )
FROM `tblname` t
WHERE t.user_id = ud.user_id
AND t.category_id <=> ud.category_id
AND t.city_id <=> ud.city_id
) = 1
)
)
Is there a way to describe this query using Select object?
Previous SQL solution was very simple and consisted of one WHERE clause:
$where = $this->getAdapter()->quoteInto("user_id = ?",$user_id);
return $this->fetchAll($where);
I need to produce same type of the result (so that it could be processed by existing code) but for more complicated query.
Things I've tried
$db = Zend_Db_Table::getDefaultAdapter();
return $db->query($sql)->fetchAll();
---------------- OR ----------------------
return $this->fetchAll($select);
---------------- OR ----------------------
return $this->_db->query($sql)->fetchAll();
But they either produce arrays instead of objects or fail with Cardinality violation message.
I would appreciate any help on how to handle SQL text queries in Zend.
$dbAdapter = Zend_Db_Table::getDefaultAdapter();
//change the fetch mode becouse you don't like the array
$dbAdapter->setFetchMode(Zend_Db::FETCH_OBJ);
$sql = "you're long sql here";
$result = $dbAdapter->fetchAll($sql);
Zend_Debug::dump($result);
exit;
For a list of all fetch modes go to Zend_Db_Adapter
To write you're query using Zend_Db_Select instead of manual string , look at Zend_Db_Slect
I've got myself in a bit of a pickle!
I've done a snazzy LINQ statement that does the job in my web app, but now I'd like to use this in a stored procedure:
var r = (from p in getautocompleteweightsproducts.tblWeights
where p.MemberId == memberid &&
p.LocationId == locationid
select p);
if (level != "0")
r = r.Where(p => p.MaterialLevel == level);
if (column == "UnitUserField1")
r = r.Where(p => p.UnitUserField1 == acitem);
if (column == "UnitUserField2")
r = r.Where(p => p.UnitUserField2 == acitem);
return r.OrderBy(p => p.LevelNo).ToList();
However, I can't for the life of me get the conditional where clause to work!!
If someone can point me in the right direction, I'd be most grateful.
Kind regards
Maybe something like this?
SELECT *
FROM dbo.weights
WHERE member_id = #memberid
AND location_id = #locationid
AND material_level = CASE WHEN #level = '0' THEN material_level
ELSE #level END
AND #acitem = CASE #column WHEN 'UnitUserField1' THEN unit_user_field_1
WHEN 'UnitUserField2' THEN unit_user_field_2
ELSE #acitem END
ORDER BY level_no
Have you tried LinqPAD, I'm pretty sure last time I played with that you could enter "LINQ to SQL" code and see the resulting SQL that produced. Failing that, place a SQL trace/profiler on your code running the LinqTOSQL and find the query being executed in the trace.
LukeH's answer will give you the correct rows, but there is something lost when you try to replace a query-generating-machine with a single query. There are parts of that query that are opaque to the optimizer.
If you need the original queries as-would-have-been-generated-by-linq, there are two options.
Generate every possible query and control which one runs by IF ELSE.
Use Dynamic sql to construct each query (although this trades away many of the benefits of using a stored procedure).
If you do decide to use dynamic sql, you should be aware of the curse and blessings of it.