stream assumes all data as UTF-8. That's not always the case.

resp = kubernetes.stream.stream(api.connect_get_namespaced_pod_exec, 'sleeping-bash', namespace='default', command=['bash', '-c', 'echo -n -e \\\\xD4'], stderr=True, stdin=False, stdout=True, tty=False)

site-packages\kubernetes\stream\ws_client.py in update(self, timeout)
    176            elif op_code == ABNF.OPCODE_BINARY or op_code == ABNF.OPCODE_TEXT:
    177                data = frame.data
    178                 if six.PY3:
--> 179                     data = data.decode("utf-8")
    180                 if len(data) > 1:

UnicodeDecodeError: 'utf-8' codec can't decode byte 0xd4 in position 1: unexpected end of data

the same error here. I solve it by change to data = data.decode("utf-8", "replace")

but. when the file is encode by other charset, this code is not used. any other solutions?

/kind bug

/sig cluster-lifecycle

when the file is encode by other charset, this code is not used. any other solutions?

from the rfc we should decode uft8 for OPCODE_TEXT but leave OPCODE_BINARY as raw. this client should not assume encoding for arbitrary binary data

the same error here. I solve it by change to data = data.decode("utf-8", "replace")

but. when the file is encode by other charset, this code is not used. any other solutions?

I get the same error when use vim command in web terminal.
And I use this code replace.
I haven't found any other coding errors yet

import chardet
data = data.decode(chardet.detect(data).get('encoding') or "utf-8")

15474ef helped to fix the error above, but it seems like this issue should be re-opened with the intent of being able to stream bytes without trying to decode them

/reopen

@pigletfly

@pigletfly: You can't reopen an issue/PR unless you authored it or you are a collaborator.

Details

In response to this:

/reopen

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

Navigation Menu

Repository navigation

stream assumes all data as UTF-8. That's not always the case. #88

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

stream assumes all data as UTF-8. That's not always the case. #88

Description

Activity

whybeyoung commented on Nov 26, 2018

whybeyoung commented on Nov 26, 2018

whybeyoung commented on Nov 26, 2018

roycaihw commented on Feb 4, 2019

leavest commented on Apr 17, 2019

tbarrella commented on Jul 3, 2019

pigletfly commented on Jun 11, 2020

k8s-ci-robot commented on Jun 11, 2020

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions