when use read_sql_query to get data if it exist special character "£"

[oceanchow]

Hi guys,
Thanks for your hard work to let this lib easily to use.
here is my question
when I use a 3rd package using pyathena, it works and return right data, while it can't work when I use below codes

wr.athena.read_sql_query(sql, database=database, use_threads=True, ctas_approach=False)

after search settings in AWSDATAWrangler about encoding, unfortunately, nothing can solve my problem, could you please help me with it? Thanks.

[jaidisido]

When setting ctas_approach=False, a regular query against athena is started and the results are stored as CSV in S3 before being read.

Special characters have caused issues in the past in a CSV, but a fix was pushed in #1257 waiting for release.

I suggest you either:

• Test with ctas_approach=True or unload_approach=True to read the results as parquet instead
• Test the fix from the PR:

pip uninstall awswrangler -y
pip install git+https://github.com/awslabs/aws-data-wrangler.git@main

[oceanchow]

Thanks Jaidisido, I will make some tests accordingly.

[oceanchow]

seems option1 don't work for me:(, but option2 works.
Hope you guys can publish a new version soon. Thanks @jaidisido

[cotrariello84]

has the bug been fixed?

[malachi-constant]

@cotrariello84 Yes this was published in release 2.16.0 and should not be an issue in awswrangler>=2.16.0

[cotrariello84]

Hi @malachi-constant I still have the bug in version 2.17.0. I had to use pyathena to solve it. but I 'd like to use wrangler.

[malachi-constant]

Hi @malachi-constant I still have the bug in version 2.17.0. I had to use pyathena to solve it. but I 'd like to use wrangler.

See comment