dupe col logic when case sensitive is false

is worse on 1.0, but fails on 1.1 thus:

from sqlalchemy import create_engine, select, literal_column

e = create_engine("sqlite://", case_sensitive=False)

row = e.execute(select([
    literal_column('1').label('SOMECOL'),
    literal_column('1').label('SOMECOL'),
])).first()

row['somecol']

the lookup should raise ambiguous column. on 1.1 it doesn't. On 1.0 we get an error when it's trying to set up the dupe lookup.

fix is likely:

1.0:

diff --git a/lib/sqlalchemy/engine/result.py b/lib/sqlalchemy/engine/result.py
index 850bc62..a4bfa02 100644
--- a/lib/sqlalchemy/engine/result.py
+++ b/lib/sqlalchemy/engine/result.py
@@ -304,6 +304,7 @@ class ResultMetaData(object):
                 for rec in raw:
                     key = rec[1]
                     if key in seen:
+                        key = key.lower() if not self.case_sensitive else key
                         by_key[key] = (None, by_key[key][1], None)
                     seen.add(key)

1.1:

diff --git a/lib/sqlalchemy/engine/result.py b/lib/sqlalchemy/engine/result.py
index 0333d9e..773022e 100644
--- a/lib/sqlalchemy/engine/result.py
+++ b/lib/sqlalchemy/engine/result.py
@@ -257,6 +257,7 @@ class ResultMetaData(object):
                     if key in seen:
                         # this is an "ambiguous" element, replacing
                         # the full record in the map
+                        key = key.lower() if not self.case_sensitive else key
                         by_key[key] = (None, key, None)
                     seen.add(key)

Comments (3)