Same join query gives difeerent results on changing alias

Question

I executed the following query,
select      distinct(gr1.i_supergroups_names)
from      dm_group_r gr1,
     dm_group_r gr2
where      gr1.r_object_id = gr2.r_object_id and
     gr2.users_names = 'grass_mig3'
322 rows returned.
Immediately, I executed the same query again with a minor change in the join condition. Instead of GR2.USER_NAMES, i used GR1.USER_NAMES.
i.e.
select      distinct(gr1.i_supergroups_names)
from      dm_group_r gr1,
     dm_group_r gr2
where      gr1.r_object_id = gr2.r_object_id and
     gr1.users_names = 'grass_mig3'
284 rows returned.
Would anyone be able to help me as to why the query is functioning like this?
This is the first time I am facing this problem Until now, I had never even thought that using different aliases could change the end result. Whenever I wrote a query, I would blindly use any one of the aliases?

Answer 1

I'm just wondering if your queries are equivalent
select distinct(gr1.i_supergroups_names)
from dm_group_r gr1,
dm_group_r gr2
where gr1.r_object_id = gr2.r_object_id and
gr2.users_names = 'grass_mig3'
select distinct(gr1.i_supergroups_names)
from dm_group_r gr1,
dm_group_r gr2
where gr1.r_object_id = gr2.r_object_id and
gr1.users_names = 'grass_mig3' Well, I think they are.
Henry's question is valid why don't you just rewrite your query to
select distinct(gr1.i_supergroups_names)
from dm_group_r gr1
where gr1.users_names = 'grass_mig3'
and   gr1.r_object_id is not null -- NULLs here got eliminated by the join aboveMessage was edited by:
Jens Petersen