[0.9.5] arithmetic on one field when selecting multiple fields returns nothing

[beckettsean]

Note that this is slightly different behavior from #4046.

Single point in the measurement:

> select * from fieldmath
name: fieldmath
---------------
time                foo value
2015-12-02T21:10:05.868748621Z  bar 12

Division works fine when selecting only that field

> select value/12 from fieldmath
name: fieldmath
---------------
time                
2015-12-02T21:10:05.868748621Z  1

> select value/12 as value from fieldmath
name: fieldmath
---------------
time                value
2015-12-02T21:10:05.868748621Z  1

However, all returns are null for that field + arithmetic if another field is also selected in the query.

> select value/12, foo from fieldmath
name: fieldmath
---------------
time                    foo
2015-12-02T21:10:05.868748621Z      bar

> select value/12 as value, foo from fieldmath
name: fieldmath
---------------
time                value   foo
2015-12-02T21:10:05.868748621Z      bar

> select foo, value/12 as value from fieldmath
name: fieldmath
---------------
time                foo value
2015-12-02T21:10:05.868748621Z  bar 

> select foo, value/12 from fieldmath
name: fieldmath
---------------
time                foo 
2015-12-02T21:10:05.868748621Z  bar 

The behavior is the same even if the additional field is just a repeat of the math field:

> select * from graph
name: graph
-----------
time                value
2015-11-04T21:29:44.423498682Z  12

> select value/12, value from graph
name: graph
-----------
time                    value
2015-11-04T21:29:44.423498682Z      12

> select value/12 as foo, value from graph
name: graph
-----------
time                foo value
2015-11-04T21:29:44.423498682Z      12

> select value, value/12 as foo from graph
name: graph
-----------
time                value   foo
2015-11-04T21:29:44.423498682Z  12  

> select value, value/12 from graph
name: graph
-----------
time                value   
2015-11-04T21:29:44.423498682Z  12  

[beckettsean]

@benbjohnson another one...

[mail2fish]

the same question +1

[desa]

fixed by #4984

[beckettsean]

Will be in the 0.9.6-nightlies starting tonight.