Skip to content

Add support for versionId for s3 path #97

@Rajashow

Description

@Rajashow

When trying to use DuckDB with httpfs to query version file from s3 using versionId in s3 url.
The following exception is thrown

s3_version_path = 's3a://s3-bucket-of-mine/testFile.parquet?VersionId=7J.noTMeAWVAK07BKaQEpJ7NMjF8HeSZ'
df = con.execute(f"""SELECT * FROM read_parquet('{s3_version_path}')""").df()
IOException: IO Error: Invalid query parameters found. Supported parameters are:
's3_region', 's3_access_key_id', 's3_secret_access_key', 's3_session_token',
's3_endpoint', 's3_url_style', 's3_use_ssl'

Note s3_version_path = 's3a://s3-bucket-of-mine/testFile.parquet does work however this only fetch the latest version.

It seem that error is being thrown here:

throw IOException("Invalid query parameters found. Supported parameters are:\n's3_region', 's3_access_key_id', "

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions