Simple query?

[Lars Brownies]

I'm testing with filter queries.

I have a table with fields ID_person and Category. Both fields belong to the
primary key. Pulling out the unique ID numbers which have Category A is
simple:

SELECT ID_person
FROM tblCategory
WHERE (((Category)="A"));

But I would like to pull out every unique ID numbers that has for instance
both "A" and "C" as its category. How can I do that?

Thanks,

Lars

 

[Jeff Boyce]

Lars

I may be reading too much into your description...

It sounds like your table for Categories include Persons (or at least
PersonIDs). In a well-normalized relational database table of categories,
there'd be no person-related information.

If you are noting persons by category, that would seem to imply three
tables, one for persons, one for categories, and one for the junction
between them (i.e., person X category).

If you want to see a query that pulls all [ID_Person] where Category = A or
Category = B, consider creating a new query in design view, setting these
selection criteria, then switching the view to the SQL view.

Good luck!

Regards

Jeff Boyce
Microsoft Access MVP

--
Disclaimer: This author may have received products and services mentioned
in this post. Mention and/or description of a product or service herein
does not constitute endorsement thereof.

Any code or pseudocode included in this post is offered "as is", with no
guarantee as to suitability.

You can thank the FTC of the USA for making this disclaimer
possible/necessary.

 

[Lars Brownies]

I have 2 tables:

tblPerson
ID_person*
Lastname
Firstname
etc.

tblPersonCategory
ID_person*
Category* (which can have value A, B, or C)

Many people can have zero, one or more categories.

Since the values are A, B, and C, I felt no need to put them in an extra
table.

If you want to see a query that pulls all [ID_Person] where Category = A
or Category = B, consider creating a new query in design view, setting
these selection criteria, then switching the view to the SQL view.

I have no problem with that *OR* query. What I need is all the unique
ID_numbers that have *both* A and C as their category. Or as another example
A and B and C. An AND statement doesn't work in this case.

Lars

Lars

I may be reading too much into your description...

It sounds like your table for Categories include Persons (or at least
PersonIDs). In a well-normalized relational database table of categories,
there'd be no person-related information.

If you are noting persons by category, that would seem to imply three
tables, one for persons, one for categories, and one for the junction
between them (i.e., person X category).

If you want to see a query that pulls all [ID_Person] where Category = A
or Category = B, consider creating a new query in design view, setting
these selection criteria, then switching the view to the SQL view.

Good luck!

Regards

Jeff Boyce
Microsoft Access MVP

--
Disclaimer: This author may have received products and services mentioned
in this post. Mention and/or description of a product or service herein
does not constitute endorsement thereof.

Any code or pseudocode included in this post is offered "as is", with no
guarantee as to suitability.

You can thank the FTC of the USA for making this disclaimer
possible/necessary.

I'm testing with filter queries.

I have a table with fields ID_person and Category. Both fields belong to
the primary key. Pulling out the unique ID numbers which have Category A
is simple:

SELECT ID_person
FROM tblCategory
WHERE (((Category)="A"));

But I would like to pull out every unique ID numbers that has for
instance both "A" and "C" as its category. How can I do that?

Thanks,

Lars

 

[Dirk Goldgar]

I'm testing with filter queries.

I have a table with fields ID_person and Category. Both fields belong to
the primary key. Pulling out the unique ID numbers which have Category A
is simple:

SELECT ID_person
FROM tblCategory
WHERE (((Category)="A"));

But I would like to pull out every unique ID numbers that has for instance
both "A" and "C" as its category. How can I do that?

If I understand your question correctly, you may have multiple records in
tblCategory for the same ID_Person, each with a different value for
Category, and you want to extract those values of ID_Person for which there
is a record for Category "A" and another record for Category "C". Is that
correct?

If so, then a query like this would probably do it:

SELECT DISTINCT ID_person FROM tblCategory C
WHERE
Exists(SELECT T.ID_person FROM tblCategory T
WHERE T.ID_person = C.ID_Person And T.Category = "A")
AND
Exists(SELECT T.ID_person FROM tblCategory T
WHERE T.ID_person = C.ID_Person And T.Category = "C")

There are a couple of other ways to do it, I think, but this one follows the
logic (as I understand it) closest.

 

[Steve]

In your form to enter PersonCategory, do you enter Category through a
combobox with a rowsource of a value list? If not and you type in A, B, C,
then you rin a large risk of a typo in a record making the data in
TblPersonCategory unreliable.

Steve
(e-mail address removed)

I have 2 tables:

tblPerson
ID_person*
Lastname
Firstname
etc.

tblPersonCategory
ID_person*
Category* (which can have value A, B, or C)

Many people can have zero, one or more categories.

Since the values are A, B, and C, I felt no need to put them in an extra
table.

If you want to see a query that pulls all [ID_Person] where Category = A
or Category = B, consider creating a new query in design view, setting
these selection criteria, then switching the view to the SQL view.

I have no problem with that *OR* query. What I need is all the unique
ID_numbers that have *both* A and C as their category. Or as another
example A and B and C. An AND statement doesn't work in this case.

Lars

Lars

I may be reading too much into your description...

It sounds like your table for Categories include Persons (or at least
PersonIDs). In a well-normalized relational database table of
categories, there'd be no person-related information.

If you are noting persons by category, that would seem to imply three
tables, one for persons, one for categories, and one for the junction
between them (i.e., person X category).

If you want to see a query that pulls all [ID_Person] where Category = A
or Category = B, consider creating a new query in design view, setting
these selection criteria, then switching the view to the SQL view.

Good luck!

Regards

Jeff Boyce
Microsoft Access MVP

--
Disclaimer: This author may have received products and services mentioned
in this post. Mention and/or description of a product or service herein
does not constitute endorsement thereof.

Any code or pseudocode included in this post is offered "as is", with no
guarantee as to suitability.

You can thank the FTC of the USA for making this disclaimer
possible/necessary.

I'm testing with filter queries.

I have a table with fields ID_person and Category. Both fields belong to
the primary key. Pulling out the unique ID numbers which have Category A
is simple:

SELECT ID_person
FROM tblCategory
WHERE (((Category)="A"));

But I would like to pull out every unique ID numbers that has for
instance both "A" and "C" as its category. How can I do that?

Thanks,

Lars

 

[Hans Up]

But I would like to pull out every unique ID numbers that has for
instance both "A" and "C" as its category. How can I do that?

Does your data look anything like this?

ID_person Category
1 a
1 b
1 c
2 c
2 d
3 b
4 a

If so, and 1 is the only ID_person value which satisfies your criteria,
try this query:

SELECT DISTINCT a.ID_person
FROM
(
SELECT ID_person
FROM tblCategory
WHERE Category="A"
) AS a
INNER JOIN
(
SELECT ID_person
FROM tblCategory
WHERE Category="C"
) AS c
ON a.ID_person = c.ID_person;

 

[Lars Brownies]

Thanks! Dirk, others,
Indeed that's what I meant.

In relation to my other post 'Filter suggestions in form' I don't see that I
can add the main table (Person) to this 3-in-1 query. The query wouldn't be
editable. So this seems to be no option.

I'm trying to normalize a table, and even in 1st normal form, it seems to
give me more hassle than benefits. Do you agree?

Lars

 

[Paul Shapiro]

Which table do you want to be able to edit? If you're form's record source
is based on Person, then maybe a filter clause like this one based on Dirk's
response would work? I think this should work with a form record source like
Select * From Person.

Exists (
SELECT * FROM tblCategory T
WHERE T.ID_person = Person.ID_Person And T.Category = "A"
)
And
Exists (
SELECT * FROM tblCategory T
WHERE T.ID_person = Person.ID_Person And T.Category = "B"
)

If not, you see if a form record source like this would leave Person data
editable:

Select Person.*, Not IsNull(C1.personID) as HasCategoryA, Not
IsNull(C2.personID) as HasCategoryB
From Person
Left Outer Join tblCategory C1 On C1.personID=Person.personID And
C1.Category = "A"
Left Outer Join tblCategory C2 On C2.personID=Person.personID And
C2.Category = "B"

 

[Dirk Goldgar]

In relation to my other post 'Filter suggestions in form' I don't see that
I can add the main table (Person) to this 3-in-1 query. The query wouldn't
be editable. So this seems to be no option.

I think Paul Shapiro addressed this in his reply.

I'm trying to normalize a table, and even in 1st normal form, it seems to
give me more hassle than benefits. Do you agree?

No. Properly normalized tables increase the power and flexibility of your
database. Though they may sometimes require more complex SQL to answer
certain kinds of questions, they make it possible to answer all sorts of
questions that are otherwise difficult or impossible even to frame.

 

[John Spencer]

One more possibility

SELECT tblPerson.*
FROM tblPerson
WHERE tblPerson.ID_Person in
(SELECT ID_Person
FROM tblPersonCategory
WHERE Category in ('A','B')
GROUP BY Id_Person
HAVING Count(Id_Person) = 2)

This should work since tblPersonCategory has a primary key based on Id_Person
plus Category. At least that is my assumption based on the asterisks in your
table descriptions.


John Spencer
Access MVP 2002-2005, 2007-2010
The Hilltop Institute
University of Maryland Baltimore County

 

[Lars Brownies]

Yes, the 'exists' filter works. I'm still able to edit. Didn't know that was
even possible! Thanks.

Lars

 

[Lars Brownies]

Point taken.
Thanks,

Lars

I think Paul Shapiro addressed this in his reply.


No. Properly normalized tables increase the power and flexibility of your
database. Though they may sometimes require more complex SQL to answer
certain kinds of questions, they make it possible to answer all sorts of
questions that are otherwise difficult or impossible even to frame.

--
Dirk Goldgar, MS Access MVP
Access tips: www.datagnostics.com/tips.html

(please reply to the newsgroup)

 

[Lars Brownies]

Yes, I use a composite key. Your solution works as well and I'm also able to
edit! Thanks.

One additional questions regarding performance:
When working with great amout of records, what option will be faster/better?
The filter option or the 'record source' option? Any other benefits on using
one over the other?

Lars

One more possibility

SELECT tblPerson.*
FROM tblPerson
WHERE tblPerson.ID_Person in
(SELECT ID_Person
FROM tblPersonCategory
WHERE Category in ('A','B')
GROUP BY Id_Person
HAVING Count(Id_Person) = 2)

This should work since tblPersonCategory has a primary key based on
Id_Person plus Category. At least that is my assumption based on the
asterisks in your table descriptions.


John Spencer
Access MVP 2002-2005, 2007-2010
The Hilltop Institute
University of Maryland Baltimore County

 

[John Spencer]

I don't know which would be faster. I would have to build the tables and try
the different options.

If I was guessing I would guess that the option I proposed would have a good
chance of winning a speed race. However; if you can make the second option
proposed by Paul Shapiro work in Access, it may be as fast or faster.
Although I would have used INNER JOINS since you only want to return records
that match.

The exists option would be the slowest since it is using two correlated
sub-queries - which means that two separate queries would run for every record
in the Person table


John Spencer
Access MVP 2002-2005, 2007-2010
The Hilltop Institute
University of Maryland Baltimore County