I have a first table called emails with a list of all the emails of my colleagues
| email |
| ----------------------- |
| saramaia#email.com |
| miguelferreira#email.com |
| joaosilva#email.com |
| joanamaia#email.com |
I have a second table called aliases, with a list of all the secondary emails/aliases my colleagues are using
| alias1 | alias2 |
| ------------------------ | ------------------- |
| joanamaia#email.com | maiajoana#email.com |
| maiajoana#email.com | maia#email.com |
| miguelferreira#email.com | miguel#email.com |
| maia#email.com | joana#email.com |
| joanamaia#email.com | jomaia#email.com |
| joana#email.com | jmaia#email.com |
I can see that the users joanamaia#email.com and miguelferreira#email.com are using aliases. But let's focus on the user joanamaia#email.com.
I need to get a list of all the email addresses the user joanamaia#email.com is using. The difficult part is that I need to get a list with the main email address plus all the intersections where the first email and consecutive ones are being used by this user. The end result should look like this
| emails |
| ------------------- |
| joanamaia#email.com |
| jomaia#email.com |
| maiajoana#email.com |
| maia#email.com |
| joana#email.com |
| jmaia#email.com |
If I do WHERE email='joanamaia#email.com' it should look like this, but I also need the same result if I do
WHERE email='jmaia#email.com'
I've been through some days of testing queries and I don't seem to have a solution for this (I've been using right joins, full outer joins and unions, but no luck so far). Is there a good way to do this?
You can use a recursive CTE to walk the graph and get the full list of interconnected aliases. Care needs to be taken to handle cycles; that requires the query to use UNION instead of the traditional UNION ALL to separate the anchor and recursive member of the CTE.
The query can take the form:
with recursive
n as (
select 'joanamaia#email.com' as email
union
select case when a.alias1 = n.email then a.alias2 else a.alias1 end
from n
join aliases a on (a.alias1 = n.email or a.alias2 = n.email)
and a.alias1 <> a.alias2
)
select * from n;
Result:
email
-------------------
joanamaia#email.com
maiajoana#email.com
jomaia#email.com
maia#email.com
joana#email.com
jmaia#email.com
See running example at DB Fiddle.
Related
i was wondering what would be the best schema to show two different type of contents in a feed. I came up with 2 ways, one used by WP and the other one is what i think should be best but i'm still not sure.
First schema and relations.
feed
| id | type | type_id |
shots
| id | user_id | title | desc | image | date creation | date update |
snippet
| id | user_id | title |desc | url | date creation | date update |
In this case, i would do something like
SELECT *
FROM feed
INNER JOIN shots
ON feed.type_id = shots.id
INNER JOIN snippets
ON feed.type_id = snippets.id
And the other schema would be WP type schema
feed
| id | user_id | title | desc | image | url | date creation | date update
And only one query without join would be done, BUT all records would be in the same table and potentially if i had other fields i would add columns here, which is not scalable at all.
I could change to the following
extra
| id | owner_id | name | value |
| 1 | 123 | url | ..... |
| 2 | 456 | image| ..... |
feed
| id | user_id | title | desc | date creation | date update
And eventually i would have the following
SELECT *
FROM feed
LEFT JOIN extra
ON feed.id = extra.owner_id
Which one is better ? Keep in mind that in future i would have more and more different kind of content (questions, videos, comments, tutorials, posts...)
I have these tables in my MySQL database:
General table:
+----generalTable-----+
+---------------------+
| id | scenario | ... |
+----+----------+-----+
| 1 | facebook | ... |
| 2 | chief | ... |
| 3 | facebook | ... |
| 4 | chief | ... |
Facebook Table:
+----facebookTable-----+
+----------------------+
| id | expiresAt | ... |
+----+-----------+-----+
| 1 | 12345678 | ... |
| 3 | 45832458 | ... |
Chief Table:
+------chiefTable------+
+----------------------+
| id | expiresAt | ... |
+----+-----------+-----+
| 2 | 43547343 | ... |
| 4 | 23443355 | ... |
Basically, the general table holds some (obviously) general data. Based on the generalTable.scenario you can look up more details in the other two tables, which are in some columns familiar (expiresAt for example) but in others not.
My question is, how to get the joined data of generalTable and the right detailed table in just one query.
So, I would like a query like this:
SELECT id, scenario, expiresAt
FROM generalTable
JOIN facebookTable
ON generalTable.id = facebookTable.id
JOIN chiefTable
ON generalTable.id = chiefTable.id
And an output like this:
| id | scenario | expiresAt |
+----+----------+-----------+
| 1 | facebook | 12345678 |
| 2 | chief | 43547343 |
| 3 | facebook | 45832458 |
| 4 | chief | 23443355 |
However, this doesn't work, because both facebookTable and chiefTable have ambiguous column name "expiresAt". For the ease of use I want to keep it that way. The result table should also only have one column "expiresAt" that is automatically filled with the right values from either facebookTable or chiefTable.
You might want to consider adding expiredAt to your general table, and removing it from the others, to remove duplication in the schema, and to make this particular query simpler.
If you need to stick with your current schema, you can use table aliases to resolve the name ambiguity, and use two joins and a union to create the result you are looking for:
SELECT g.id, g.scenario, f.expiresAt
FROM generalTable g
JOIN facebookTable f
ON g.id = f.id
UNION ALL
SELECT g.id, g.scenario, c.expiresAt
FROM generalTable g
JOIN chiefTable c
ON g.id = c.id;
The outer join approach mentioned in another answer would also solve the problem.
One way you could accomplish it is with LEFT JOIN. In the result fields you can do something like this for common fields IF(fTbl.id IS NULL, cTbl.expiresAt, fTbl.expiresAt) AS expiresAt.
I've a query that I need some help with -
As part of a form I've got a serial number field that is populated if there is a serial number, blank if it's not, or no result if it's an invalid serial number.
select *
from cust_site_contract as cs
where cs.serial_no = 'C20050' or (cs.serial_no <> 'C20050' and if(cs.serial_no = 'C20050',1,0)=0)
limit 10;
Here's a sample of the regular data:
+----------------------+-----------+-----------+-----------
| idcust_site_contract | system_id | serial_no | end_date
+----------------------+-----------+-----------+-----------
| 561315 | SH001626 | C19244 | 2009-12-21
| 561316 | SH001626 | C19244 | 2010-06-30
| 561317 | SH002125 | C19671 | 2010-05-31
| 561318 | SH001766 | C14781 | 2010-09-25
| 561319 | SH001766 | C14781 | 2011-02-15
| 561320 | SH002059 | C19020 | 2008-07-09
| 561321 | SH002639 | C18889 | 2008-03-31
| 561322 | SH002639 | C18889 | 2008-06-30
| 561323 | SH002715 | C20051 | 2010-04-30
| 561324 | SH002719 | C20057 | 2010-04-30
And an exact result would look something like this:
| 561487 | SH002837 | C20050 | 2012-07-04
I was writing this as a subquery so I could match the system_ids to customer and contract names, but realised I was getting garbage pretty early on.
I'm tempted to try and simplify it by saying the third case might not hold true (i.e. if it's an invalid serial number, allow the choice of any customer name and simply flag it in the data)
Has anyone got any ideas of where I'm going wrong? The combination of conditions is clearly wrong, and I can't work out how to make each side of the or statement mutually exclusive
Even if I try to evaluate only the if(sn = 'blah') I get the wrong result for obvious reasons, but can't think of a sane way to express it.
Many thanks
Scott
If there is is no contract with a serial number of C20050, this query will return all rows, otherwise, it will return only one row where serial_no is C20050:
SELECT a.*
FROM cust_site_contract a
INNER JOIN
(
SELECT COUNT(*) AS rowexists
FROM cust_site_contract
WHERE serial_no = 'C20050'
) b ON b.rowexists = 0
UNION ALL
(
SELECT *
FROM cust_site_contract
WHERE serial_no = 'C20050'
LIMIT 1
)
If you just write the query as below you will get blank if doesn't exists or it's an invalid serial number.
select cs.serial_no from cust_site_contract as cs where cs.serial_no = 'C20050'
Some background: an 'image' is part of one 'photoshoot', and may be a part of zero or many 'galleries'. My tables:
'shoots' table:
+----+--------------+
| id | name |
+----+--------------+
| 1 | Test shoot |
| 2 | Another test |
| 3 | Final test |
+----+--------------+
'images' table:
+----+-------------------+------------------+
| id | original_filename | storage_location |
+----+-------------------+------------------+
| 1 | test.jpg | store/test.jpg |
| 2 | test.jpg | store/test.jpg |
| 3 | test.jpg | store/test.jpg |
+----+-------------------+------------------+
'shoot_images' table:
+----------+----------+
| shoot_id | image_id |
+----------+----------+
| 1 | 1 |
| 1 | 2 |
| 3 | 3 |
+----------+----------+
'gallery_images' table:
+------------+----------+
| gallery_id | image_id |
+------------+----------+
| 1 | 1 |
| 1 | 2 |
| 2 | 3 |
| 3 | 1 |
| 4 | 1 |
+------------+----------+
What I'd like to get back, so I can say 'For this photoshoot, there are X images in total, and these images are featured in Y galleries:
+----+--------------+-------------+---------------+
| id | name | image_count | gallery_count |
+----+--------------+-------------+---------------+
| 3 | Final test | 1 | 1 |
| 2 | Another test | 0 | 0 |
| 1 | Test shoot | 2 | 4 |
+----+--------------+-------------+---------------+
I'm currently trying the SQL below, which appears to work correctly but only ever returns one row. I can't work out why this is happening. Curiously, the below also returns a row even when 'shoots' is empty.
SELECT shoots.id,
shoots.name,
COUNT(DISTINCT shoot_images.image_id) AS image_count,
COUNT(DISTINCT gallery_images.gallery_id) AS gallery_count
FROM shoots
LEFT JOIN shoot_images ON shoots.id=shoot_images.shoot_id
LEFT JOIN gallery_images ON shoot_images.image_id=gallery_images.image_id
ORDER BY shoots.id DESC
Thanks for taking the time to look at this :)
You are missing the GROUP BY clause:
SELECT
shoots.id,
shoots.name,
COUNT(DISTINCT shoot_images.image_id) AS image_count,
COUNT(DISTINCT gallery_images.gallery_id) AS gallery_count
FROM shoots
LEFT JOIN shoot_images ON shoots.id=shoot_images.shoot_id
LEFT JOIN gallery_images ON shoot_images.image_id=gallery_images.image_id
GROUP BY 1, 2 -- Added this line
ORDER BY shoots.id DESC
Note: The SQL standard allows GROUP BY to be given either column names or column numbers, so GROUP BY 1, 2 is equivalent to GROUP BY shoots.id, shoots.name in this case. There are many who consider this "bad coding practice" and advocate always using the column names, but I find it makes the code a lot more readable and maintainable and I've been writing SQL since before many users on this site were born, and it's never cause me a problem using this syntax.
FYI, the reason you were getting one row before, and not getting and error, is that in mysql, unlike any other database I know, you are allowed to omit the group by clause when using aggregating functions. In such cases, instead of throwing a syntax exception, mysql returns the first row for each unique combination of non-aggregate columns.
Although at first this may seem abhorrent to SQL purists, it can be incredibly handy!
You should look into the MySQL function group by.
I am trying to do multiple joins on the same MySQL table, but am not getting the results that I expect to get. Hopefully someone can point out my mistake(s).
Table 1 - cpe Table
|id | name
|----------
| 1 | cat
| 2 | dog
| 3 | mouse
| 4 | snake
-----------
Table 2 - AutoSelect
|id | name | cpe1_id | cpe2_id | cpe3_id |
|-----------------------------------------------
| 1 | user1 | 1 | 3 | 4 |
| 2 | user2 | 3 | 1 | 2 |
| 3 | user3 | 3 | 3 | 2 |
| 4 | user4 | 4 | 2 | 1 |
------------------------------------------------
I would like to see an output of
user1 | cat | mouse | snake |
user2 | mouse | snake | dog |
..etc
Here is what I have tried
SELECT * FROM AutoSelect
LEFT JOIN cpe ON
( cpe.id = AutoSelect.cpe1_id ) AND
( cpe.id = AutoSelect.cpe2_id ) AND
( cpe.id = AutoSelect.cpe3_id )
I get blank results. I thought i knew how to do these joins, but apparently when I'm trying to match cpe?_id with the name of the cpe table.
Thanks in advance for any assistance.
You need left join 3 times as well. Currently your query only joins 1 time with 3 critieria as to the join. This should do:
SELECT a.name, cpe1.name, cpe2.name, cpe3.name FROM AutoSelect as a
LEFT JOIN cpe as cpe1 ON ( cpe1.id = a.cpe1_id )
LEFT JOIN cpe as cpe2 ON ( cpe2.id = a.cpe2_id )
LEFT JOIN cpe as cpe3 ON ( cpe3.id = a.cpe3_id )
And you probably mean to INNER JOIN rather than LEFT JOIN unless NULL values are allowed in your AutoSelect table.
I think your design is wrong.
With tables like that, you get it the way it's meant to be in relational databases :
table 1 : animal
id name
1 cat
2 dog
3 mouse
4 snake
table 2 : user
|id | name |
|--------------
| 1 | user1 |
| 2 | user2 |
| 3 | user3 |
| 4 | user4 |
table 3 : association
|id_user | id_animal|
|--------------------
| 1 | 1 |
| 1 | 3 |
| 1 | 4 |
| 2 | 3 |
| 2 | 1 |
| 2 | 2 |
| 3 | 3 |
| 3 | 2
| 4 | 4 |
| 4 | 2 |
| 4 | 1 |
---------------------
Then :
select u.name, a.name from user u, animal a, association ass where ass.user_id = u.id and ass.animal_id = a.id;
In this case, your solution won't produce a good dynamic database. There are other ways to make combinations of multiple tables. I can show you by my own database what you should use and when you should use this solution. The scheme is in dutch, but you'll probably understand the keywords.
Like you, I had to combine my windmills with a kWh-meter, which has to measure the energyproduction of my windmills. What you should do, is this case, is making another table(in my case molenkWhlink). Make sure your tables are INNODB-types(for making Foreign keys). What I've done is combining my meters and mills by putting a pointer(a foreign key) of their ID(in Dutch Volgnummer) in the new table. An advantage you may not need, but I certainly did, is the fact I was able to extend the extra table with connection and disconnection info like Timestamps and metervalues when linking or unlinking. This makes your database way more dynamic.
In my case, I Also had a table for meassurements(metingoverzicht). As you can see in the scheme, I've got 2 lines going from Metingoverzicht to molenkwhlink. The reason for this is quite simple. All meassurements I take, will be saved in table Metingoverzicht. Daily meassurements(which are scheduled) will have a special boolean put on, but unscheduled meassurements, will also me saved here, with the bollean turned off. When switching meters, I need the endvalue from the leaving meter and the startvalue from the new meter, to calculate the value of todays eneryproduction. This is where your solution comes in and an extra table won't work. Usually, when you need just one value from another table a JOIN will be used. The problem in this case is, I've got 2 meassurementIDs in 1 link(1 for connecting and 1 for disconnecting). They both point to the same tablecolumn, because they both need to hold the same type of information. That is when you can use a double JOIN from one table towards the other. Because, both values will only be used once, and just needed to be saved in a different place to avoid having 1 action stored on different locations, which should always be avoided.
http://s1101.photobucket.com/user/Manuel_Barcelona/media/schemedatabase.jpg.html