add a doc example of SQL's ON statement

[cpcloud]

see this SO question

[jreback]

why wouldn't you just pre filter the frames before the merge?

[cpcloud]

hmm that's probably how you should do it...ok maybe then a cookbook/sql -> pandas example in docs?

[jreback]

sure

[guyrt]

That link is dead. Does anyone have a synopsis?

[cpcloud]

IIRC the question was about joining two frames with an arbitrary expression in the on kwarg. The solution is to execute the expression and then perform the merge.

[jreback]

@cpcloud maybe make a separate docs tag for things involveing sql/r doc translations?

[cpcloud]

good idea

[cpcloud]

closing in favor of #4611

[hayd]

Here's the question in full:

Is it possible to pass a generic lambda condition to the merge method, to do something to the effect of:

frame1.merge(frame2, how='left', cond=lambda row1, row2: row2['minA'] <= row1['colA'] <= row2['maxA'])

effectively emulating a generic sql join:

SELECT A.*, B.* FROM A LEFT JOIN B ON A.colA BETWEEN B.minA AND B.maxA..

The only way I can think of right now is to do a cartesian product and then do a filter, which doesn't seem right..

OP deleted it because it was suggested it was a dupe of another question :(.