windmill-labs/windmill · error · Exception

Type of file_content not supported

Error message

Type of file_content not supported

What it means

write_s3_file only accepts file_content that is raw bytes (bytes) or a BufferedReader (a file opened in binary mode), because httpx request content must be bytes or a bytes generator. Any other type (str, io.StringIO, io.BytesIO, text-mode file objects, etc.) is rejected with this exception before any network call is made.

Source

Thrown at python-client/wmill/wmill/client.py:987

        s3_obj = S3Object(s3="/path/to/my_file.txt")

        # for an in memory bytes array:
        file_content = b'Hello Windmill!'
        client.write_s3_file(s3_obj, file_content)

        # for a file:
        with open("my_file.txt", "rb") as my_file:
            client.write_s3_file(s3_obj, my_file)
        '''
        """
        s3object = parse_s3_object(s3object)
        # httpx accepts either bytes or "a bytes generator" as content. If it's a BufferedReader, we need to convert it to a generator
        if isinstance(file_content, BufferedReader):
            content_payload = bytes_generator(file_content)
        elif isinstance(file_content, bytes):
            content_payload = file_content
        else:
            raise Exception("Type of file_content not supported")

        query_params = {}
        if s3object is not None and s3object["s3"] != "":
            query_params["file_key"] = s3object["s3"]
        if s3_resource_path is not None and s3_resource_path != "":
            query_params["s3_resource_path"] = s3_resource_path
        if (
            s3object is not None
            and "storage" in s3object
            and s3object["storage"] is not None
        ):
            query_params["storage"] = s3object["storage"]
        if content_type is not None:
            query_params["content_type"] = content_type
        if content_disposition is not None:
            query_params["content_disposition"] = content_disposition

        try:

View on GitHub (pinned to e474e8803c)

Solutions

1. Encode strings to bytes: file_content.encode('utf-8')
2. Open files in binary mode: open(path, 'rb')
3. For in-memory buffers pass buf.getvalue() bytes or io.BytesIO(...).read()
4. Check isinstance(file_content, (bytes, io.BufferedReader)) before calling

Example fix

// before
wm.write_s3_file(s3obj, "my file contents")
// after
wm.write_s3_file(s3obj, "my file contents".encode("utf-8"))

import io
if not isinstance(file_content, (bytes, io.BufferedReader)):
    raise TypeError(f"file_content must be bytes or BufferedReader, got {type(file_content).__name__}")

def is_valid_s3_content(c: object) -> bool:
    import io
    return isinstance(c, (bytes, io.BufferedReader))

try:
    wm.write_s3_file(s3obj, file_content)
except Exception as e:
    if "Type of file_content not supported" in str(e):
        fixed = file_content.encode("utf-8") if isinstance(file_content, str) else bytes(file_content)
        wm.write_s3_file(s3obj, fixed)
    else:
        raise

When it happens

Trigger: Calling wm.write_s3_file with file_content as a str, an io.StringIO, an io.BytesIO, a text-mode file object, or json.dumps() output without .encode(), e.g. wm.write_s3_file(s3obj, 'hello') where 'hello' is a str.

Common situations: Forgetting 'rb' mode when opening files (open(path) instead of open(path, 'rb')), passing string data without encoding, or passing pandas/numpy/buffer objects directly.

Related errors

• Invalid s3 object {s3_object!r}: expected an s3://<storage>/
• Could not generate DuckDB S3 connection settings from the pr
• Could not generate Polars S3 connection settings from the pr
• Could not generate Boto3 S3 connection settings from the pro
• Invalid s3 object URI {s3_object!r}: expected s3://<storage>

AI-assisted analysis of windmill-labs/windmill@e474e8803c (2026-09-03). Data as JSON: /api/errors/cfc087bb24c73fe7. Report an issue: GitHub.