[Sqlalchemy-tickets] Issue #3987: Select duplicates itself when using column() on its own column (zzzeek/sqlalchemy)

[dataradiant <iss...@bi...>]

New issue 3987: Select duplicates itself when using column() on its own column
https://bitbucket.org/zzzeek/sqlalchemy/issues/3987/select-duplicates-itself-when-using-column

dataradiant:

I suspect `Select.column()` is intended to be passed columns from a `table` rather than itself, but if that's the case a note in the documentation would helpful.  My actual usage was a lot more complicated than the example and this took a while to figure out.


```
#!python

from sqlalchemy import select, table, column, func

t1 = table("t1", column("c1"))
s1 = select([t1])

s2 = s1.column(func.min(s1.c.c1))

#  >>> str(s2)
#  'SELECT t1.c1, min(c1) AS min_1 \nFROM t1, (SELECT t1.c1 AS c1 \nFROM t1)'

s3 = s1.column(func.min(t1.c.c1))

#  Note that using the column from the table rather than the Select works fine
#  >>> str(s3)
#  'SELECT t1.c1, min(t1.c1) AS min_1 \nFROM t1'
```