Thread: Telling how many records are joined

Greetings-

I'm sure there's a (moderately) simple answer to this, but I'd love some
help with it.

I have a database with four tables:

papers  -  information about newspapers, including a unique paperid
letters -  information and text of letters to the editors, including    a unique paperid and the paperid in which it
appeared
patterns-  word patterns occurring in one or more letters, including    a unique patternid
pattern_occurrences - a linking table containing a letterid, a patternid,                and a count (the number of
timesthe pattern occurs                      in the letter).
 

The patterns and pattern_occurrences tables are quite large (around 3
million records each).

I'd like to know, for example, how many letters from each paper have one
or more patterns already coded. The best I can do is:

SELECT papercode, count(papercode) FROM papers, letters
WHERE papers.paperid=letters.paperid AND
letters.letterid IN (SELECT DISTINCT o_letterid FROM pattern_occurrences) 
GROUP BY papercode;

Thanks for any advice.

----------------------------------------------------------------------
Andrew J Perrin - andrew_perrin@unc.edu - http://www.unc.edu/~aperrinAssistant Professor of Sociology, U of North
Carolina,Chapel Hill     269 Hamilton Hall, CB#3210, Chapel Hill, NC 27599-3210 USA
 

It sounds like you want unique papercode items right?

Thats what distinct keyword is for:

SELECT distinct papercode, count(papercode) FROM papers, letters
WHERE papers.paperid=letters.paperid AND
letters.letterid IN (SELECT DISTINCT o_letterid FROM pattern_occurrences) 
GROUP BY papercode;

Also a another trick you can include just those having a count of more 
than one:

SELECT distinct papercode, count(papercode) FROM papers, letters
WHERE papers.paperid=letters.paperid AND
letters.letterid IN (SELECT DISTINCT o_letterid FROM pattern_occurrences) 
GROUP BY papercode HAVING count(papercode) > 0;

- Bill



Andrew Perrin wrote:

>Greetings-
>
>I'm sure there's a (moderately) simple answer to this, but I'd love some
>help with it.
>
>I have a database with four tables:
>
>papers  -  information about newspapers, including a unique paperid
>letters -  information and text of letters to the editors, including
>        a unique paperid and the paperid in which it appeared
>patterns-  word patterns occurring in one or more letters, including
>        a unique patternid
>pattern_occurrences - a linking table containing a letterid, a patternid, 
>                   and a count (the number of times the pattern occurs
>                       in the letter).
>
>The patterns and pattern_occurrences tables are quite large (around 3
>million records each).
>
>I'd like to know, for example, how many letters from each paper have one
>or more patterns already coded. The best I can do is:
>
>
>
>Thanks for any advice.
>
>----------------------------------------------------------------------
>Andrew J Perrin - andrew_perrin@unc.edu - http://www.unc.edu/~aperrin
> Assistant Professor of Sociology, U of North Carolina, Chapel Hill
>      269 Hamilton Hall, CB#3210, Chapel Hill, NC 27599-3210 USA
>
>
>
>---------------------------(end of broadcast)---------------------------
>TIP 3: if posting/reading through Usenet, please send an appropriate
>subscribe-nomail command to majordomo@postgresql.org so that your
>message can get through to the mailing list cleanly
>