I'm using Access for a cookbook database, an excercise for Uni.
I'm trying to make a query for a many-to-many relationship.
I have ricette and ingredienti tables, and a junction table named ricetta_ingrediente. Now I should make a query that retrieves each ricette associated with ingredienti used.
EDIT: Part of the question got deleted, I need to retrieve the recipe's ingredients of the one with the most ingredients used, it's the result I have to obtain.
Every single try got me either a syntax error or an empty result - how can I achieve this query?
MORE INFOS
The relationship schema
[
I've tried to implement this suggestion, failing at it, how should it be?
Also here's my try and Access error :
[
Use the query builder DesignView to assist in building SQL statement. Result should look like:
SELECT ricette.nome, ingredienti.nome
FROM ingredienti
RIGHT JOIN (ricette RIGHT JOIN ricetta_ingrediente
ON ricette.ID = ricetta_ingrediente.id_ricetta)
ON ingredienti.ID = ricetta_ingrediente.id_ingrediente;
To retrieve recipe with the most ingredients as well as the ingredients, like:
SELECT TOP 1 ricette.nome, ingredienti.nome
FROM (SELECT id_ricetta, Count([id_ingrediente]) AS CountIng
FROM ricetta_ingrediente GROUP BY id_ricetta) AS Q1
RIGHT JOIN (ricette RIGHT JOIN (ingredienti RIGHT JOIN ricetta_ingrediente
ON ingredienti.ID = ricetta_ingrediente.id_ingrediente)
ON ricette.ID = ricetta_ingrediente.id_ricetta)
ON Q1.id_ricetta = ricetta_ingrediente.id_ricetta
ORDER BY Q1.CountIng DESC;
This will not resolve ties. All recipes with the number of ingredients matching the TOP 1 count will return. How should the query know you want only 1 and which one?
Your query is fine. You just need parentheses, because this is MS Access.
I would also use table aliases:
SELECT r.nome, i.nome
FROM (ricette as r INNER JOIN
ricetta_ingrediente as ri
ON r.ID = ri.id_ricetta
) INNER JOIN
ingredienti as i
ON i.ID = ri.id_ingrediente;
EDIT:
For the revised question:
SELECT TOP (1) r.nome
FROM (ricette as r INNER JOIN
ricetta_ingrediente as ri
ON r.ID = ri.id_ricetta
) INNER JOIN
ingredienti as i
ON i.ID = ri.id_ingrediente
GROUP BY r.nome
ORDER BY COUNT(*) DESC;
Related
I have two scripts that work in isolation, but I don't know how to stitch them together to do this all at once.
The first script:
SELECT *
FROM PROD_ANALYTIC.SRC_MVC_AU_DBO.BCAPPLICATIONQUOTES bq
JOIN PROD_ANALYTIC.SRC_MVC_AU_DBO.PRODUCT p ON bq.ProductId = p.Id
JOIN PROD_ANALYTIC.SRC_MVC_AU_DBO.BCAPPLICATIONQUOTEFEES F ON F.ApplicationQuoteId = bq.Id
LEFT JOIN PROD_ANALYTIC.SRC_MVC_AU_DBO.COVERAGESECTIONTYPE CST ON F.CoverageSectionTypeID = CST.SECTIONID
This produces a table (is that the right word?). Suppose I could save that table as 'appQuote'. I then want to run this script on it:
SELECT
ApplicationId,
STRING_AGG(YEARLYPAYAMOUNT) allPremium,
STRING_AGG(SECTIONID) allSection,
STRING_AGG(SHORTNAME) allSectionName,
FROM
appQuote
GROUP BY
ApplicationId
For storage reasons I don't want to actually save the table from the first script, I just want to get the result from the first script and immediately apply the second script to it.
This is very basic, so any guidance would help.
The relevant term is known as a derived table.
In SQL you can use the result of a query as the input to another. The "nested query" produces the derived table which is then consumed by the outer query:
Select <columns>
from (
select <columns> from table join table etc
)as TableAliasName
To get the nomenclature right, what you have are not scripts or tables, but rather two queries. When a query runs it produces a result set.
To accomplish your goal, either of these queries will work:
SELECT
ApplicationId,
STRING_AGG(YEARLYPAYAMOUNT) allPremium,
STRING_AGG(SECTIONID) allSection,
STRING_AGG(SHORTNAME) allSectionName,
FROM (
SELECT *
FROM PROD_ANALYTIC.SRC_MVC_AU_DBO.BCAPPLICATIONQUOTES bq
JOIN PROD_ANALYTIC.SRC_MVC_AU_DBO.PRODUCT p ON bq.ProductId = p.Id
JOIN PROD_ANALYTIC.SRC_MVC_AU_DBO.BCAPPLICATIONQUOTEFEES F ON F.ApplicationQuoteId = bq.Id
LEFT JOIN PROD_ANALYTIC.SRC_MVC_AU_DBO.COVERAGESECTIONTYPE CST ON F.CoverageSectionTypeID = CST.SECTIONID
) appQuote
GROUP BY
ApplicationId
Here, the inner nested query is formally called a derived table, but people will often use the phrase "subquery".
The other option is a Common Table Expression (CTE):
With appQuote As (
SELECT *
FROM PROD_ANALYTIC.SRC_MVC_AU_DBO.BCAPPLICATIONQUOTES bq
JOIN PROD_ANALYTIC.SRC_MVC_AU_DBO.PRODUCT p ON bq.ProductId = p.Id
JOIN PROD_ANALYTIC.SRC_MVC_AU_DBO.BCAPPLICATIONQUOTEFEES F ON F.ApplicationQuoteId = bq.Id
LEFT JOIN PROD_ANALYTIC.SRC_MVC_AU_DBO.COVERAGESECTIONTYPE CST ON F.CoverageSectionTypeID = CST.SECTIONID
)
SELECT
ApplicationId,
STRING_AGG(YEARLYPAYAMOUNT) allPremium,
STRING_AGG(SECTIONID) allSection,
STRING_AGG(SHORTNAME) allSectionName,
FROM
appQuote
GROUP BY
ApplicationId
Finally, you could also create a View from the first query:
CREATE View AppQuote As
SELECT *
FROM PROD_ANALYTIC.SRC_MVC_AU_DBO.BCAPPLICATIONQUOTES bq
JOIN PROD_ANALYTIC.SRC_MVC_AU_DBO.PRODUCT p ON bq.ProductId = p.Id
JOIN PROD_ANALYTIC.SRC_MVC_AU_DBO.BCAPPLICATIONQUOTEFEES F ON F.ApplicationQuoteId = bq.Id
LEFT JOIN PROD_ANALYTIC.SRC_MVC_AU_DBO.COVERAGESECTIONTYPE CST ON F.CoverageSectionTypeID = CST.SECTIONID
A view is not just a query; it will actually become part of the database, such that you can use it in many of the same ways you would use a table saved on disk, and then the second query in the original question will run unmodified.
This is the schema for my questions
Hi, I don't have experience in SQL Developer and I'm trying to build a query for the following question:
I need that for each DVD in the catalog, display the title, length, release_date, and how many times it has been checked out by all customers across all libraries.
Also I want to include those that have not been checked out yet displaying 0, and sort results by title.
So far I have this in the query but I'm stock here:
--Question C. ************* VERIFY
Select
Catalog_Item.Title,
DVD.Length,
Catalog_Item.Release_Date,
(
Select
Count(Transaction.Transaction_ID)
From Transaction
Where
DVD.Catalog_Item_ID = Physical_Item.Catalog_Item_ID
And Physical_Item.Physical_Item_ID = Transaction.Physical_Item_ID
) as "Total_DVD"
From
Catalog_Item,DVD,
Physical_Item
Group by
Catalog_Item.Title,
DVD.Length,
Catalog_Item.Release_Date
If I run this exact query I get error
Not a Group By Expression
And if I exclude the GROUP BY, I get results by doesn't look like the correct outputs.
Any suggestions on what syntax I can use to achieve the desired output? Thanks!
You put three tables to the query but you missed to link them. If you don't link them, you will see too much-duplicated rows.
Also, your sub-query links were wrong, I assume you tried to put the links here that you missed in the main query.
I believe you need something like that:
Select
CI.Title
,DVD.Length
,CI.Release_Date
,NVL(TR.TotalTransactions,0) TotalTransactions
From Catalog_Item CI
INNER JOIN DVD ON DVD.Catalog_Item_ID = CI.Catalog_Item_ID
LEFT JOIN Physical_Item PHI ON CI.Catalog_Item_ID = PHI.Catalog_Item_ID
LEFT JOIN (SELECT Physical_Item_ID
, Count(Transaction_ID) TotalTransactions
FROM Transaction
GROUP BY Physical_Item_ID
) TR ON PHI.Physical_Item_ID = TR.Physical_Item_ID
For a start, join Catalog_Item, Physical_Item and DVD together. Without appropriate join conditions, these three tables will join using a cartesian product join - which is probably one of the reasons why you are seeing unexpected results.
I have 7 Tables as per attached following Image.
I will either enter Engine Number or Chassis Number and it should show the respective tables information (these tables have only mentioned fields) so all fields can be shown as result.
I can use hard coded Engine Number or Chassis Number. Every time of execution of this Query, I will hard code the required Engine/Chassis Number and will get the result.
Can anybody please help me to write this query for me?
Click Here to See the Tables
This might be a starting point for your solution.
SELECT prod.EngineNo AS engNo, prod.ChassisNo, doral.doralNo [, table.column [AS name]]
FROM DOProductSpecsDetais AS prod
INNER JOIN DORAL AS doral
ON prod.DOProductSpecsDetailID = doral.DOProductSpecsID
INNER JOIN DOProductDetail AS prodDetail
ON prod.DOProductDetailID = prodDetail.DOProductDetailID
WHERE prod.ChassisNo = '<input>' OR prod.EngineNo='<input>'
Between the SELECT and the FROM Statement, you can select any column out of your JOIN.
You can cascade as many JOINs as you like...
Which DBMS are you going to use?
One suggestion: Try to simplify the names of your columns, if possible.
One more: If you just started to do Database things, it is always helpful to start a test environment and use a client tool.
You can write query something like this:
select * from
DoProductSpecsDetail tbl1 inner join Doral tbl2
on tbl1.DoProductSpecsDetailId = tbl2.DoProductSpecsId
inner join DoproductDetail tbl3
on tbl1.DoProductDetailId = tbl3.DoProductDetailId
inner join ProductColor tbl4
on tbl1.ProductColorId = tbl4.ProductColorId
inner join DoDetail tbl5
on tbl3.DeliveryOrderDetailId = tbl5.DeliveryOrderId
inner join ProductMain tbl6
on tbl3.ProductId = tbl6.ProductId
inner join BPMain tbl7
on tbl5.BusinessPartnerId = tbl7.BusinessPartnerId
I'm trying to run a query on a table (the one Wordpress uses) where I want to select the ID and post_type columns from one table, then do a Left Join to another table, two separate times (getting separate data).
This is what I have so far, but it's not cutting the mustard:
SELECT derby_posts.id AS pID,
derby_posts.post_type AS tier
FROM derby_posts
LEFT JOIN (SELECT derby_postmeta.post_id AS dbID1,
derby_postmeta.meta_key AS dbMeta1)
ON pid = dbid1
AND dbmeta1 = 'twitter'
LEFT JOIN (SELECT derby_postmeta.post_id AS dbID2,
derby_postmeta.meta_key AS dbMeta2)
ON pid = dbid2
AND dbmeta2 = 'website'
WHERE tier IN ('local', 'regional', 'national')
I'm sure I'm missing something super simple...
Edit: here's the solution that worked for me. Table alias helped, putting all my SELECT statements together cleaned things up. Also, I realized I could remove items from the SELECT, even though I'm using them in the Join, which cleans up the results a lot.
SELECT
db.ID as id,
db.post_type as tier,
dpm1.meta_value as twitter,
dpm2.meta_value as website
FROM derby_posts db
LEFT JOIN derby_postmeta dpm1 ON (db.ID = dpm1.post_id AND dpm1.meta_key = 'twitter' )
LEFT JOIN derby_postmeta dpm2 ON (db.ID = dpm2.post_id AND dpm2.meta_key = 'website' )
WHERE db.post_type IN ('local','regional','national')
I 'm sure I'm missing something super simple...
You are right!
You need to give your selects an alias, and use that alias in the ON clause. You are also missing a FROM <table> - a required part of a SELECT statement that reads from a table:
LEFT JOIN (
SELECT derby_postmeta.post_id AS dbID1,
derby_postmeta.meta_key AS dbMeta1
FROM someTable
) dpm ON pid = dpm.dbid1 AND dpm.dbmeta1 = 'twitter'
I gave the results of your SELECT an alias dpm, and used it to "link up" the rows from the inner select to the rows of your outer select.
SELECT
db.derby_posts.ID as pID,
db.derby_posts.post_type as tier,
dpm1.post_id as dbID1,
dpm1.meta_key as dbMeta1,
dpm2.post_id as dbID2,
dpm2.meta_key as dbMeta2
FROM derby_posts db
LEFT JOIN derby_postmeta dpm1 ON (db.pID = dpm1.post_id AND dpm1.meta_key= 'twitter')
LEFT JOIN derby_postmeta dpm2 ON (db.pID = dbm2.post_id AND dbm2.meta_key = 'website')
WHERE tier IN ('local','regional','national')
I have this query and I need to include another join on a table called "likes" where updates.id = likes.update_id. There will be 0 or more matches on this join.
"SELECT * FROM users
INNER JOIN updates ON users.remote_id=updates.owner_id
ORDER BY updates.status_time DESC LIMIT 50"
This is probably fairly simple, but I haven't been able to find any examples for this kind of query.
The situation is basically that I'm displaying a list of items. I do a join on the users table to grab the user who created each item. I also need to do a join on the "likes" tables to display the 0+ people who liked each item.
EDIT: My Solution
Ok, here's the successful join and combining of duplicate results (due to the fact that there are multiple "likes" for each update) using GROUP_CONCAT.
"SELECT users.*,updates.update_id,updates.content,updates.status_time,
GROUP_CONCAT(likes.liker SEPARATOR ',') AS liked_by
FROM updates
LEFT OUTER JOIN likes ON updates.update_id = likes.update_id
JOIN users ON users.remote_id = updates.owner_id
GROUP BY updates.update_id
ORDER BY updates.status_time DESC
LIMIT 200"
SELECT * FROM users
LEFT OUTER JOIN updates ON users.remote_id=updates.owner_id
INNER JOIN likes ON <join condition here>
ORDER BY updates.time DESC LIMIT 50
Will that work?
If not, can you clarify what exactly your question is? Does the query you have not work?
if you have a 1 to n relation the you should use a LEFT JOIN , i mean if there's only 1 user and many likes.user_id you should doit with the following way
"SELECT * FROM users LEFT JOIN updates ON (users.remote_id=updates.owner_id) ORDER BY updates.time DESC LIMIT 50"
this way you'd get all the updates from a certain user :)
cheers