Iterate through properties

[matt connolly]

Is there a way to iterate through the properties (attributes) of a
pyodbc.Row ?

If I run:

cursor.execute("SELECT * FROM MyTable")
row = cursor.fetchone()

then I can access the fields like `row.field1` which is great.

But when I do:

for field in row:

[matt connolly]

sorry, here is rest:

when I do:

for field in Row:
print field

field contains the values.

Is there a way to iterate over the row and get the field names, like a
dict?

[mkleehammer]

Absolutely. The way specified in the DB API (and therefore portable
to other db libraries) is to use cursor.description, which is a tuple
of tuples, documented here: http://code.google.com/p/pyodbc/wiki/Cursor

The first item in the tuples is the column names, so you can extract
them like so:

cols = [ t[0] for t in cursor.description ]

You could then do something like:

for name, value in zip(cols, row):
print '%s: %s' % (name, value)

I created the Row class to be useful even after the cursor and
connection are closed, so I added the same tuple to all rows as
Row.cursor_description. Apparently I forgot to make the Row wiki
page, but you can see the API documentation by executing help
(pyodbc.Row).

BTW, the Row internals are specially designed to use very little
memory. I use Rows where I would have to create tiny classes or
structs in languages like Java and C/C++. The row names and the
cursor_description are shared by all rows from the same execute call,
so they use less memory than if you created separate objects, one for
each row.

[matt connolly]

Thanks!

Row.cursor_description is perfect for my needs.

Don't know how I missed it in the help()....

Great to know about the Row memory sharing.