I have table (about 80'000 rows), looks like
id, parentId, col1, col2, col3...
1, null, 'A', 'B', 'C'
2, 1, ...
3, 1, ...
4, null, ...
5, 4, ...
(one level parent - child only)
and I need get all dependent rows -
SELECT ...
FROM table
WHERE id = :id OR parentId = :id OR id IN (
SELECT parentId
FROM table
WHERE id = :id
)
but why this request working slowly instead 2 request - if I get parentId on php first?
$t = executeQuery('SELECT parentId FROM table WHERE id = :Id;', $id);
if ($t) {
$id = $t;
}
$t = executeQuery('SELECT * FROM table WHERE id = :id OR parentId = :id ORDER BY id;', $id);
PS: max depends rows < 70
PPS:
id select_type table type possible_keys key key_len ref rows Extra
1 PRIMARY product ALL PRIMARY,parentId NULL NULL NULL 73415 Using where
2 DEPENDENT SUBQUERY product const PRIMARY,parentId PRIMARY 4 const 1
Change the IN for an equal =
SELECT ...
FROM table
WHERE id = :id OR parentId = :id OR id = (
SELECT parentId
FROM table
WHERE id = :id
)
or change it to a join:
SELECT ...
FROM table
inner join (
SELECT parentId
FROM table
WHERE id = :id
) s on s.parentID = table.id or s.parentID = table.parentID
Well, in the first case, MySQL need to create an intermediate result, store it in memory and then iterate over it to find all the relevant id in the table. In the second way, assuming you correctly created an index on id and parent id, it simply go straigth to the index, find the relevant rows, and send you back the result immediately.
UNION works faster for this case
this allows first query to user UNION INDEX and second just uses inner join, then merges results.
SELECT *
FROM `table`
WHERE id = :id OR parentId = :id
UNION
SELECT t1.*
FROM `table` t1 JOIN `table` t2 ON t2.parentId = t1.id AND t2.id = :id
An EXPLAIN might shed some more light on the problem for you.
Look into EXISTS, or rewriting your query as a JOIN.
It's a long shot but in first case you have "IN" statement of the WHERE part of the query. Maybe MySQL tries to optimize the query as if there would be multiple options and in the second case there is no IN part, so the compiled query is more straight forward for the database - thus utilizing the indexes in better manner.
Basically for 2 queries on the same connection the overhead of performing the queries should be minimal and irelevant in this case. Also subqueries in general are not very optimizable by the query parser. Try using JOIN instead (if possible).
Related
I want to execute a query where I can find one ID in a list of ID.
table user
id_user | name | id_site
-------------------------
1 | james | 1, 2, 3
1 | brad | 1, 3
1 | suko | 4, 5
and my query (doesn't work)
SELECT * FROM `user` WHERE 3 IN (`id_site`)
This query work (but doesn't do the job)
SELECT * FROM `user` WHERE 3 IN (1, 2, 3, 4, 6)
That's not how IN works. I can't be bothered to explain why, just read the docs
Try this:
SELECT * FROM `user` WHERE FIND_IN_SET(3,`id_site`)
Note that this requires your data to be 1,2,3, 1,3 and 4,5 (ie no spaces). If this is not an option, try:
SELECT * FROM `user` WHERE FIND_IN_SET(3,REPLACE(`id_site`,' ',''))
Alternatively, consider restructuring your database. Namely:
CREATE TABLE `user_site_links` (
`id_user` INT UNSIGNED NOT NULL,
`id_site` INT UNSIGNED NOT NULL,
PRIMARY KEY (`user_id`,`site_id`)
);
INSERT INTO `user_site_links` VALUES
(1,1), (1,2), (1,3),
(2,1), (2,3),
(3,4), (3,5);
SELECT * FROM `user` JOIN `user_site_links` USING (`id_user`) WHERE `id_site` = 3;
Try this: FIND_IN_SET(str,strlist)
NO! For relation databases
Your table doesn't comfort first normal form ("each attribute contains only atomic values, and the value of each attribute contains only a single value from that domain") of a database and you:
use string field to contain numbers
store multiple values in one field
To work with field like this you would have to use FIND_IN_SET() or store data like ,1,2,3, (note colons or semicolons or other separator in the beginning and in the end) and use LIKE "%,7,%" to work in every case. This way it's not possible to use indexes[1][2].
Use relation table to do this:
CREATE TABLE user_on_sites(
user_id INT,
site_id INT,
PRIMARY KEY (user_id, site_id),
INDEX (user_id),
INDEX (site_id)
);
And join tables:
SELECT u.id, u.name, uos.site_id
FROM user_on_sites AS uos
INNER JOIN user AS u ON uos.user_id = user.id
WHERE uos.site_id = 3;
This way you can search efficiently using indexes.
The problem is that you are searching within several lists.
You need something more like:
SELECT * FROM `user` WHERE id_site LIKE '%3%';
However, that will also select 33, 333 and 345 so you want some more advanced text parsing.
The WHERE IN clause is useful to replace many OR conditions.
For exemple
SELECT * FROM `user` WHERE id IN (1,2,3,4)
is cleaner than
SELECT * FROM `user` WHERE id=1 OR id=2 OR id=3 OR id=4
You're just trying to use it in a wrong way.
Correct way :
WHERE `field` IN (list_item1, list_item2 [, list_itemX])
This question may have been asked before but I don't really know what verbiage to search with.
I have a mysql DB that has a table with 3 columns [ID, fieldName and fieldValue] that is used to describe attributes of objects in another table. The ID field stores the foreign key of object in the other table and the fieldName and fieldValue store things like title, description, file size and summary.
I am trying to write a query that returns rows where a fieldName and fieldValue pair match known values and the returned row ID has a another distinct fieldValue in another row. Right now I am accomplishing it with two queries and an if statement. Here is the sudo code:
$result = SELECT * FROM table_a WHERE fieldName = 'title' and fieldValue = 'someTitle'
$test = SELECT * FROM table_a WHERE fieldValue = 'someValue' and id = '{$result['id']}'
if ($test) {
/* Result Found */
}
You can self-join the table:
SELECT * FROM table_a AS s1
JOIN table_a AS s2 USING (id)
WHERE
s1.fieldName = 'Title' AND s1.fieldValue = 'someTitle'
AND s2.fieldValue = 'someValue'
What you said translated in sql would be:
SELECT b.*
FROM table_a a
INNER JOIN table_a b ON a.id = b.id
WHERE a.fieldName = 'title'
AND a.fieldValue = 'someTitle'
AND a.fieldValue <> b.fieldValue
This gets you the rows in table_a that have the same id as the row with you predefined values, but with a different fieldValue. This assumes that id is not the primary key, otherwise there will not be another row with the same id, but it looks in your question that this isn't the case. (If you want to check for a specific value you can do: AND b.fieldValue = 'someValue' in the last line)
I want to join these two table but the join key of the second table is in a query string,
page table,
page_id url
1 a
2 c
3 d
system table,
system_id query
1 page_id=1&content=on&image=on
2 type=post&page_id=2&content=on
as you can see that page_id is part of the query string in system table.
so how can I join them like the standard joining table method below?
SELECT*
FROM page AS p
LEFT JOIN system AS s
ON p.page_id = s.page_id
EDIT:
I def can change the system table into something like this,
system_id page_id query
1 1 page_id=1&content=on&image=on
2 2 type=post&page_id=2&content=on
3 NULL type=page
But the reason why I don't want to do this is that the page_id is no need for many certain records. I don't want make a column with too many null.
I would definitely create columns for page_id,content, image and type (and get them indexed). Then the database would be much lighter and would work much faster.
Joining two tables without the common field and data type, is fundamentally wrong IMO.
I will suggest that you extract the page_id and insert it in the database and use a normal join to accomplish what you are searching for.
SO making the columns like
+------------+-----------+---------+
| system_id | page_id | query |
------------------------------------
Here is a snippet with which you are extract the page_id.
$query = 'page_id=1&content=on&image=on';
$queryParts = explode('&', $query);
$params = array();
foreach ($queryParts as $param) {
$item = explode('=', $param);
$params[$item[0]] = $item[1];
}
$page_id = $parems['page_id'];
Then you can go on with the insert and use simple join statement to solve your problem in a proper way.
Update:
Since you are able to change the schema to a feasible one. You dont need to worry about some rows having empty rows on this.
I guess you wanted something like this (MSSQL!):
DECLARE #query VARCHAR(50)
DECLARE #Lenght INT
DECLARE #PageID INT
SET #query = '4kkhknmnkpage_id=231&content=on&image=on'
SET #Lenght = PATINDEX('%&%', substring(#query,PATINDEX('%page_id=%', #query),50)) - 9
SET #PageID = CAST(SUBSTRING(#query,PATINDEX('%page_id=%', #query) + 8,#Lenght) AS INT)
SELECT #PageID -- you can do as you please now :)
OR:
SELECT*
FROM page AS p
LEFT JOIN (SELECT CAST(SUBSTRING(query,PATINDEX('%page_id=%', query) + 8,(PATINDEX('%&%', substring(query,PATINDEX('%page_id=%', query),50)) - 9)) AS INT) AS page_id
FROM system) AS s
ON p.page_id = s.page_id
-- Do as you please again :)
I guess what you really wanted was something like this (MYSQL!):
SET #query := '4kkhknmnkpage_id=231&content=on&image=on';
SET #Lenght := POSITION('&' IN (SUBSTR(#query,POSITION('page_id=' IN #query),50))) - 9;
SET #PageID := CAST(SUBSTR(#query,POSITION('page_id=' IN #query) + 8,#Lenght) AS SIGNED );
SELECT #PageID
OR
SELECT*
FROM page AS p
LEFT JOIN (SELECT CAST(SUBSTR(query,POSITION('page_id=' IN query) + 8,(POSITION('&' IN (SUBSTR(query,POSITION('page_id=' IN query),50))) - 9)) AS SIGNED) AS pageID
FROM system) AS s
ON p.page_id = s.pageID
There might be an easy solution here, but it seems to have me tripped up. I'm trying to query a table based on an array of values in two columns. Here is the pertinent table structure and sample data
comment table
id, userId, articleId .... etc etc
article table
id, userId .... etc etc
Data: UserIds = 3, 10. ArticleIds = 1, 2
Let's say I'm trying to find all the comments for a particular set of article IDs: 1,2
I can easily use this query
select * from comments WHERE articleId IN(1,2)
However, here is where it gets complex. I have a query that executes prior to the comments query that determines the appropriate article IDs. Those IDs are in an array. Also in an array are the corresponding user IDs for each article.
What I want to do now is query the comments table for only the articles in the array (1,2) AND only for those comments made by the original author (3, 10).
The simple query above will bring back all the comments for articleId 1 and 20. So for example I can't figure out where to add another conditional that says onyl comments for articleId 1, 20 AND corresponding userId, 3, 10.
Any help or suggestions would be much appreciated!
I think the simplest way is to write:
SELECT comments.*
FROM articles
JOIN comments
ON articles.id = comments.articleId
AND articles.userId = comments.userId
WHERE articles.id IN (1, 2)
;
The AND articles.userId = comments.userId clause is what enforces your "only for those comments made by the original author" requirement.
Alternatively, you can use an EXISTS clause:
SELECT *
FROM comments
WHERE articleId IN (1, 2)
AND EXISTS
( SELECT 1
FROM articles
WHERE id = comments.articleId
AND userId = comments.userId
)
;
or a single-row subquery:
SELECT *
FROM comments
WHERE articleId IN (1, 2)
AND userId =
( SELECT userId
FROM articles
WHERE id = comments.articleId
)
;
Have you tried just
select * from comments WHERE articleId IN(1,2) and authorId in (3,10)
If not, please update your question why it's so.
You can add as many IN statements as you want:
select * from comments WHERE articleId IN(1,2) AND userId IN (3,10)
Can't you just make the query that runs first a subquery within your stated query:
SELECT * FROM `comments` WHERE `articleId` IN(1,2) AND `userId` IN (__subquery__)
-- you can use subquery
select * from comments WHERE articleId IN(1,2)
and userId in ( select userId from article where id in (1,2) )
Hello i have a question on picking random entries from a database. I have 4 tables, products, bids and autobids, and users.
Products
-------
id 20,21,22,23,24(prime_key)
price...........
etc...........
users
-------
id(prim_key)
name user1,user2,user3
etc
bids
-------
product_id
user_id
created
autobids
--------
user_id
product_id
Now a multiple users can have an autobid on an product. So for the next bidder I want to select a random user from the autobid table
example of the query in language:
for each product in the autobid table I want a random user, which is not the last bidder.
On product 20 has user1,user2,user3 an autobidding.
On product 21 has user1,user2,user3 an autobidding
Then I want a resultset that looks for example like this
20 – user2
21 – user3
Just a random user. I tried miximg the GOUP BY (product_id) and making it RAND(), but I just can't get the right values from it. Now I am getting a random user, but all the values that go with it don't match.
Can someone please help me construct this query, I am using php and mysql
The first part of the solution is concerned with identifying the latest bid for each product: these eventually wind up in temporary table "latest_bid".
Then, we assign randon rank values to each autobid for each product - excluding the latest bid for each product. We then choose the highest rank value for each product, and then output the user_id and product_id of the autobids with those highest rank values.
create temporary table lastbids (product_id int not null,
created datetime not null,
primary key( product_id, created ) );
insert into lastbids
select product_id, max(created)
from bids
group by product_id;
create temporary table latest_bid ( user_id int not null,
product_id int not null,
primary key( user_id, product_id) );
insert into latest_bid
select product_id, user_id
from bids b
join lastbids lb on lb.product_id = b.product_id and lb.created = b.created;
create temporary table rank ( user_id int not null,
product_id int not null,
rank float not null,
primary key( product_id, rank ));
# "ignore" duplicates - it should not matter
# left join on latest_bid to exclude latest_bid for each product
insert ignore into rank
select user_id, product_id, rand()
from autobids a
left join latest_bid lb on a.user_id = lb.user_id and a.product_id = lb.product_id
where lb.user_id is null;
create temporary table choice
as select product_id,max(rank) choice
from rank group by product_id;
select user_id, res.product_id from rank res
join choice on res.product_id = choice.product_id and res.rank = choice.choice;
You can use the LIMIT statement in conjunction with server-side PREPARE.
Here is an example that selects a random row from the table mysql.help_category:
select #choice:= (rand() * count(*)) from mysql.help_category;
prepare rand_msg from 'select * from mysql.help_category limit ?,1';
execute rand_msg using #choice;
deallocate prepare rand_msg;
This will need refining to prevent #choice becoming zero, but the general idea works.
Alternatively, your application can construct the count itself by running the first select, and constructing the second select with a hard-coded limit value:
select count(*) from mysql.help_category;
# application then calculates limit value and constructs the select statement:
select * from mysql.help_category limit 5,1;