I am wondering how to use dataset operations to obtain an effect equivalent 
to MySQL's GROUP_CONCAT statement.

Right now I have a table with a list of scientific sources, a table with a 
list of authors, and a junction table for the many-many relation between 
authors and sources.  The primary keys in the 'source' and 'author' table 
are both 'id'; the junction table contains columns named 'id_author' and 
'id_source'.  When I simply join these three tables together, I get a table 
where there is a row for each author-source combination.  

*table = DB[:author_to_source].join(src, :id=>:id_source).join(DB[:author], 
:id=>:author_to_source__id_author)
*
*
*
*"SELECT * FROM `author_to_source` INNER JOIN (SELECT * FROM `source`) AS 
`t1` ON (`t1`.`id` = `author_to_source`.`id_source`) INNER JOIN (SELECT * 
FROM `author`) AS `t2` ON (`t2`.`id` = `t1`.`id_author`)"*

I need to collapse all rows having the same source into a single row, by 
combining all the author names into a single comma-separated list that can 
fit in one field.  In MySQL I can do this with GROUP_CONCAT.  I can't 
figure out how to do it Sequel.  I have tried:

*table.group(:id_source)*
*
*
*"SELECT * FROM `author_to_source` INNER JOIN (SELECT * FROM `source`) AS 
`t1` ON (`t1`.`id` = `author_to_source`.`id_source`) INNER JOIN (SELECT * 
FROM `author`) AS `t2` ON (`t2`.`id` = `author_to_source`.`id_author`) 
GROUP BY `id_source`"*

But as you can see, this just adds a GROUP BY clause, which is a necessary 
part of GROUP_CONCAT but does not do the same thing.  Is there a way to get 
GROUP_CONCAT functionality?  Thanks.

-- 
You received this message because you are subscribed to the Google Groups 
"sequel-talk" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/sequel-talk/-/ksXbH9_SrOMJ.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/sequel-talk?hl=en.

Reply via email to