How to get column names from SQLAlchemy result (declarative syntax)

Question

I am working in a pyramid project and I've the table in SQLAlchemy in declarative syntax

"""models.py"""
class Projects(Base):
    __tablename__ = 'projects'
    __table_args__ = {'autoload': True}

I get the results by using

""""views.py"""
session = DBSession()
row_data = session.query(Projects).filter_by(id=1).one()

How can I get the column names from this result.

PS: I am unable to use this method since I am using the declarative syntax.

Solution 1

The difference is between ORM and non-ORM, not declarative, which is just a helper for the ORM.

Query has a method column_descriptions() that was added for this purpose::

http://www.sqlalchemy.org/docs/orm/query.html#sqlalchemy.orm.query.Query.column_descriptions

the example there seems like it has a typo, says q.columns but it should be q.column_descriptions (edit: just fixed it).

Solution 2

You can do something similar to Foo Stack's answer without resorting to private fields by doing:

conn.execute(query).keys()

Solution 3

from sqlalchemy.ext.declarative import declarative_base
from sqlalchemy import (Column, Index, Date, DateTime, Numeric, BigInteger, String, ForeignKey, Boolean)

Base = declarative_base()

class Project(Base):
    """sqlalchemy ORM for my table."""
    __tablename__ = "table1"
    id = Column("id", BigIntegerID, primary_key=True, autoincrement=True)
    date = Column("date", Date, nullable=False)
    value = Column("value", Numeric(20, 8))
    ...
    ...

Then this will return the columns names ['id', 'date', 'value', ...]:

Project.__table__.columns.keys()

Or this

Project.metadata.tables['table1'].columns.keys()

Solution 4

Just playing around, this syntax will give you all the columns (so to solve your problem, set query to look at one table/object only):

conn.execute(query)._metadata.keys

Solution 5

This link shows how to get all the metadata you could ever need about a table, column and more.

SQLAlchemy Metadata

Many of the answers above are based on the info on this page. Suppose we have declared a table.

employees = Table('employees', metadata,
    Column('employee_id', Integer, primary_key=True),
    Column('employee_name', String(60), nullable=False),
    Column('employee_dept', Integer, ForeignKey("departments.department_id"))
)

Here are some examples of getting metadata about the table.

# access the column "EMPLOYEE_ID":
employees.columns.employee_id

# or just
employees.c.employee_id

# via string
employees.c['employee_id']

# iterate through all columns
for c in employees.c:
    print(c)

# get the table's primary key columns
for primary_key in employees.primary_key:
    print(primary_key)

# get the table's foreign key objects:
for fkey in employees.foreign_keys:
    print(fkey)

# access the table's MetaData:
employees.metadata

# access the table's bound Engine or Connection, if its MetaData is bound:
employees.bind

# access a column's name, type, nullable, primary key, foreign key
employees.c.employee_id.name
employees.c.employee_id.type
employees.c.employee_id.nullable
employees.c.employee_id.primary_key
employees.c.employee_dept.foreign_keys

# get the "key" of a column, which defaults to its name, but can
# be any user-defined string:
employees.c.employee_name.key

# access a column's table:
employees.c.employee_id.table is employees

# get the table related by a foreign key
list(employees.c.employee_dept.foreign_keys)[0].column.table

View more solutions

Comments

[Sukumar]

Thanks for the answer! Unfortunately, this returns [{'aliased': False, 'expr': <class 'metrics.models.Projects'>, 'type': <class 'metrics.models.Projects'>, 'name': 'Projects'}] I am trying to use reflection so that I don't have to specify the column name and with this method I'll still have to specify the column names. Or am I missing something?

[Sukumar]

I think I found the solution. I was able to get the column list by using Projects.__table__.columns.keys()

[CubeBot88]

Note that 'columns' is interchangeable with 'c', so for above Project.metadata.tables[Project.__tablename__].c.keys() would also work. As would Project.__table__.c.keys()

[Grail Finder]

its still there 5 years later. Only way you can make something reliable is to rely on that.

[Jcc.Sanabria]

Nice! It'll be useful in a property (setter) method

[medley56]

I think this only works if you are specifying which columns you want back in the query (which means you already know the column names). What @Sukumar and I need is the columns for a query constructed like q=sess.query(MyTableObject).all()

[wolfmanx]

column_descriptions is an attribute, not a method.

[Tom]

I know this is an old answer, but hoping someone knows: if the query is for more than one entity, how do you get unambiguous column names? For a query like Query(DataClass1).join(DataClass2).add_entity(DataClass2), if DataClass1 and DataClass2 have fields with the same names, then keys() returns a list with duplicate entries.

[Kots]

bravo! i learned something and didn't just copy