UnicodeDecodeError when altering table

Question

I'm trying to do automatic alter table - adding column when it is necessary. The problem is that I've started to getting UnicodeDecodeError.

I don't understand why this error is raising. Why it wants to use 'ascii' charset.

I've tried to print attr.__class__ which prints unicode. I've already tried to attr.decode('utf-8') but still the same problem is raising.

LINE:

self.cur.execute("""ALTER TABLE data ADD COLUMN {} TEXT""".format(attr))

RETURNS:

UnicodeEncodeError: 'ascii' codec can't encode character u'\u010d' in position 5: ordinal not in range(128)

I have # -*- coding: utf-8 -*- on the top of the file.

What should I do in this case? Is it possible to find out which encoding is used?

EDIT: If I rewrite the alter not to use variables, it works.

Like a lot of Unicode-related issues, this would probably either go away or at least give you a better error message if you were using Python 3. — lvc
– lvc, Commented Sep 10, 2015 at 10:48

Daniel Roseman · Accepted Answer · 2015-09-10 10:35:25Z

1

To convert from unicode to bytes you use encode, not decode.

Alternatively, make the SQL string unicode:

self.cur.execute(u"""ALTER TABLE data ADD COLUMN {} TEXT""".format(attr))

answered Sep 10, 2015 at 10:35

Daniel Roseman

602k68 gold badges910 silver badges923 bronze badges

Sign up to request clarification or add additional context in comments.

1 Answer 1