I have 2 tables, student and grades
student table contains id, name and date_of_birth
grades table contains id, student_id, grade and course
Actual table contain more data.
I have a query like
SELECT s.*, AVG(g.grade) as average_grade
FROM student s LEFT JOIN grade g ON s,id = g.student_id
WHERE g.course = 'mathematics' and s.id = 1
With this I could get the data i needed which are student details and the average grade, then come the problem where when the course = "mathematics" is not found in the grades table, the query will return NULL. My question is, is there a way for me to get the s.id = 1 details together with NULL average instead of all NULL value?
I would prefer if it is able to do it with 1 query, as because in my current I am using subquery and it takes very long to get the data. My main objective is to get more faster speed if you have better way instead of using 1 query feel free to comment your idea. In addition I have tried multiple query and sub query to get all the data but it all take too long.
Move your filter criteria for g.course = 'mathematics' in joining part
SELECT s.*, AVG(g.grade) as average_grade
FROM student s
LEFT JOIN grade g ON s.id = g.student_id AND g.course = 'mathematics'
WHERE s.id = 1
Your query produces result as inner join not left because putting g.course = 'mathematics in where clause turns your left join to inner join, Moving this part in on clause will still return data from student table if there were no rows found from grade table with course = 'mathematics'
If the course is not 'mathematics' you would still get the student data if you put it like this.
SELECT s.*, AVG(g.grade) as average_grade
FROM student s LEFT JOIN grade g ON s,id = g.student_id
WHERE (g.course = 'mathematics' AND s.id = 1) OR s.id = 1
Related
SELECT *
FROM `all_salary_slip_details` `assd`
INNER JOIN `employee_master` `em` ON `assd`.`EMPLOYEE_ID` = `em`.`EMPLOYEE_ID`
INNER JOIN `employee_details` `ed` ON `assd`.`EMPLOYEE_ID` = `ed`.`EMPLOYEE_ID`
INNER JOIN `organizations` `o` ON `o`.`ORG_ID` = `assd`.`ORG_ID`
INNER JOIN `months` `mo` ON `mo`.`id` = `assd`.`PAY_MONTH`
This is my query to fetch data from db. I have only 4 rows but in my view I get 48 rows. I dont understand why this happens.
You need to use a GROUP BY clause on a column, then your query will only return a single row for each unique entry in that column.
In your case I'm guessing it would make sense to use the employee name.
What result does this query output?
SELECT assd.PAY_YEAR, assd.PDF, em.EMPLOYEE_NAME, o.ORG_NAME, mo.month_name
FROM all_salary_slip_details assd
INNER JOIN employee_master em
ON assd.EMPLOYEE_ID = em.EMPLOYEE_ID
INNER JOIN organizations o
ON o.ORG_ID = assd.ORG_ID
INNER JOIN months mo
ON mo.id = assd.PAY_MONTH
GROUP BY em.EMPLOYEE_NAME
You will get 1 row returned in your query for each record in a joined table that matches the join condition.
So, if you have multiple records in any one of the employee_master, employee_details, organisations or months tables that match those in all_salary_slip_details you will get additional rows returned. Look at the rows returned, you will see differences in each and where they differ will tell you where the join condition for the table isn't specific enough to return a single row.
I'm really struggling to get my head around this. I am trying to run a SELECT query from multiple tables.
This is what I have so far that doesn't work;
SELECT jira_issues.*, session_set.* FROM jira_issues, session_set
INNER JOIN reports on jira_issues.report_id = reports.id
WHERE jira_issues.report_id = 648
I have other tables (session_set, report_device) which has a ReportID and report_id column respectively.
I have a report table which has a Primary Key id. In the other tables the report.id key is linked with foreign keys.
Ultimately what I am trying to achieve is this:
I have an entry in the reports table with an id of 648. In the other tables (jira_issues, report_device, session_set), I also have entries which has a foreign key linked to the report id in the report table.
I want to run one SELECT Query to query the tables (jira_issues, report_device and session_set) and get all the data from them based on the report.id.
Thanks!
What about this:
SELECT * FROM jira_issues ji
LEFT JOIN session_set ss ON ji.report_id = ss.ReportID
LEFT JOIN report_device rd ON rd.report_id = ji.report_id
WHERE ji.report_id = 648;
Just say "no" to commas in the from clause. Always use explicit join syntax:
SELECT ji.*, session_set.*
FROM jira_issues ji inner join
reports r
on ji.report_id = r.id inner join
session_set ss
on ss.ReportId = r.report_id
WHERE ji.report_id = 648;
If some of the tables might have no corresponding rows, you might want left outer join instead of inner join.
Kindly try this out. You may get syntax error.
SELECT a., b. FROM jira_issues a, session_set b, reports c
Where a.report_id = c.id and b.report_id = c.id AND a.report_id = 648
Help me out with this query:
I have 3 tables with this structure.
items_to_groups (item_id | group_id)
item_to_regions (item_id | region_id)
items [a bunch of columns]
I need to select every row on the item table that has an item_id match on item_to_groups table WHERE group = x AND has an item_id match on item_to_regions table WHERE region = y
Currently the code I have is a horrible subquery with loops and all.
What would be a better way of doing this?
I've thought about JOIN and such, but can't really get my head around on how to do it.
SELECT bunch_of_columns
FROM items i
INNER JOIN items_to_groups ig ON i.id=ig.item_id
INNER JOIN items_to_regions ir on i.id=ir.item_d
WHERE ir.region_id=y
AND ig.group_id=x
Have a look at the JOIN documentation on MySQL. Joins are important for relational databases.
As you said you have a hard time grasping joins, have a look at A Visual Explanation of SQL Joins by Jeff Atwood. Maybe it helps.
SELECT colums
FROM items
INNER JOIN items_to_groups ON items.item_id = items_to_groups.item_id AND group_id = x
INNER JOIN items_to_regions ON items.item_id = items_to_regions.item_id AND region_id = y
SELECT * FROM items
JOIN items_to_groups ON (items.item_id = items_to_groups.item_id AND group_id = ?)
JOIN items_to_regions ON (items.item_id = items_to_regions.item_id AND region_id = ?)
GROUP BY items.item_id
Im trying to construct a query that goes over 3 tables and im COMPLETELY stumped ... my knowledge limit is basic 1 table query and i need some help before i stick my head in a blender.
I have the following query
SELECT * FROM internalrole WHERE introle = $imarole
Im fine with that part .. its the next thats getting me all stressed.
That query returns the following columns ( id, user_id, introle, proven, used )
What i then need to do is take the user_id from the results returned and use it to get the following
SELECT * FROM users WHERE id = user_id(from previous query) AND archive = 0 and status = 8
I need to put that into 1 query, but wait, theres more .... from the results there, i need to check if that user's 'id' is in the availability table, if it is, check the date ( column name is date ) and if it matches todays date, dont return that one user.
I need to put all that in one query :S ... i have NO IDEA how to do it, thinking about it makes my head shake ... If someone could help me out, i would be eternaly grateful.
Cheers,
Use INNER JOIN, which links tables to each other based on a common attribute (typically a primary - foreign key relationship)
say an attribute, 'id', links table1 and table2
SELECT t1.att1, t2.att2
FROM table1 t1
INNER JOIN table2 t2
ON t1.id = t2.id --essentially, this links ids that are equal with each other together to make one large table row
To add more tables, just add more join clauses.
SELECT u.*
FROM internalrole ir
INNER JOIN users u
ON ir.user_id = u.id
AND u.archive = 0
AND u.status = 8
LEFT JOIN availability a
ON ir.user_id = a.user_id
AND a.date = CURDATE()
WHERE ir.introle = $imarole
AND a.user_id IS NULL /* User does NOT exist in availability table w/ today's date */
EDIT: This second query is based on the comments below, asking to show only users who do exist in the availability table.
SELECT u.*
FROM internalrole ir
INNER JOIN users u
ON ir.user_id = u.id
AND u.archive = 0
AND u.status = 8
INNER JOIN availability a
ON ir.user_id = a.user_id
WHERE ir.introle = $imarole
Hmm, maybe something like this
SELECT * FROM users WHERE id IN (SELECT user_id FROM internalrole WHERE introle = $imarole) AND archive = 0 and status = 8;
A handy thing for me to remember is that tables are essentially arrays in SQL.
HTH!
Nested queries are your friend.
SELECT * FROM users WHERE id in (SELECT user_id FROM internalrole WHERE introle = $imarole) AND archive = 0 and status = 8
Alternatively joins:
SELECT * FROM users INNER JOIN internalrole ON users.id = internalrole.user_id WHERE internalrole.user_id = $imarole AND users.archive = 0 and users.status = 8
Hey guys, I have this query.
"SELECT COUNT(purchase_log.id), purchase_log.date_purchased, purchase_log.total_cost, purchase_log.payment_status, cart_contents.product_name, members.first_name, members.last_name, members.email FROM `purchase_log` LEFT JOIN `cart_contents` ON purchase_log.id = cart_contents.purchase_id LEFT JOIN `members` ON purchase_log.member_id = members.id";
As you can see I have 3 different tables I am trying to draw data from. After running this query, it returns with the count of 9. That is actually the total number of rows in the table "cart_contents" but what I want is for it to return 5 because the table "purchase_log" contains the total transaction per row and each transaction can have 1 or more rows in table "cart_contents".
So how can I form the query to count the right amount?
You can try,
SELECT COUNT(DISTINCT
purchase_log.id),
purchase_log.date_purchased,
purchase_log.total_cost,
purchase_log.payment_status,
cart_contents.product_name,
members.first_name, members.last_name,
members.email FROM purchase_log LEFT
JOIN cart_contents ON
purchase_log.id =
cart_contents.purchase_id LEFT JOIN
members ON purchase_log.member_id =
members.id