Question

So I have 2 tables I need to join. Table1 includes a column called JDAY which hold the values 1.5, 2.5, 3.5…365.5. It looks like:

JDAY
1.5
2.5
3.5
4.5
5.5
etc.

I would like to join it with Table2, which looks like:

JDAY  WSC
1     1
5     .9
8     .7
366   .5

The final result should have a JDAY column with all the values from Table1, and the WSC value from Table2 corresponding to the closest JDAY value less than or equal to that in Table2. For example, JDAY=5.5 in Table1 corresponds to WSC=.9, because 5.5 is between 5 and 8. It would look like this:

JDAY   WSC
1.5    1
2.5    1
3.5    1
4.5    1
5.5    .9
6.5    .9
7.5    .9
8.5    .7
9.5    .7
etc.

Answer 1

One way to get the result is to use a correlated subquery:

UPDATED

SELECT t1.JDAY
     , (SELECT t2.WSC FROM Table2 t2 
         WHERE t2.JDAY < t1.JDAY
         ORDER BY t2.JDAY DESC LIMIT 1
       ) AS WSC
  FROM Table1 t1
 ORDER BY t1.JDAY 

You noted you wanted the largest value of WSC less than JDAY; I’m taking that to mean that you want the maximum value of WSC from the rows in Table2 that have a JDAY value LESS THAN the JDAY value from the Table1 row. That seems to fit with the result set you are showing. (If I got that wrong (which I may have), it was my misunderstanding, reading something you didn’t intend, if so, please feel free to correct me.)

UPDATE

I have a better understanding, you want the value of WSC from the row in Table2 that has the largest JDAY value that is earlier than the JDAY value in Table1. You want the WSC from the row with with the latest earlier date. The queries in my answer are modified to satisfy this requirement.

---

Another alternative to get basically the same result set is to use a theta-join (that is, a join on an inequality predicate)

UPDATED

SELECT t.JDAY
     , s.WSC
  FROM ( SELECT t1.JDAY
              , MAX(t2.JDAY) AS max_t2_jday
           FROM Table1 t1
           LEFT
           JOIN Table2 t2
             ON t2.JDAY < t1.JDAY
          GROUP BY t1.JDAY
          ORDER BY t1.JDAY
       ) t
  LEFT
  JOIN Table2 s
    ON s.JDAY = t.max_t2_jday

That will work as long as there aren’t any “duplicate” values of JDAY in Table2. If there are duplicate values, you’ll need to have a way to eliminate any duplicates that get returned. Adding a GROUP BY t.JDAY would be sufficient, but not deterministic. (That is, you wouldn’t be guaranteed which row you were getting the WSC value from.)

Here, we are asking for every row from Table1, AND any “matching” rows from Table2, matching on the JDAY from Table2 being LESS THAN the JDAY from Table1. For each row in Table1, we will rummage through all the “matching” rows from Table2 (if any) and we will pick out the largest WSC value matching JDAY from Table2 and return it in the subquery (inline view) aliased as t. We can then join that result set back to Table2, to get the matching row, along with the WSC value on that row.

One difference between these queries comes into play when there are “duplicate” values of JDAY in Table1, they will return slightly different result sets. The first query will return all the rows from Table1, including duplicate values of JDAY. The second query will remove duplicate values of JDAY from Table 1, and return a distinct set of values for JDAY.

---

UPDATE:

I’ve updated the answer above, not getting the MAX(WSC) value, but instead getting the WSC value from the row in Table2 with the closest JDAY that is less than the Table1 JDAY.