Requirements are the same, so it's still possible to use `urllib3 < 2`
or `requests == 2.28.2` for example.
Signed-off-by: Felix Fontein <felix@fontein.de>
Only remove `:ro` or `:rw` suffixes in their entirety; do not
strip arbitrary `r` / `o` / `w` / `:` characters individually.
Signed-off-by: Loïc Leyendecker <loic.leyendecker@gmail.com>
This is related to https://github.com/moby/moby/pull/44216
Prunes will, by default, no longer prune named volumes, only anonymous ones.
Signed-off-by: Brian Goff <cpuguy83@gmail.com>
Add support for floats to docker logs params `since` / `until` since the
Docker Engine APIs support it.
This allows using fractional seconds for greater precision.
Signed-off-by: Archi Moebius <poerhiz@gmail.com>
I noticed one Dockerfile was pinned to 1.4; given that there's a
backward compatibility guarantee on the stable syntax, the general
recommendation is to use `dockerfile:1`, which makes sure that the
latest stable release of the Dockerfile syntax is pulled before
building.
While changing, I also made some minor changes to some Dockerfiles
to reduce some unneeded layers.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
While removing any usage of the deprecated `distutils` package,
("The distutils package is deprecated and slated for removal in Python 3.12.")
this internal utility can be removed straightaway because the
`shutil.which` replacement for `distutils.spawn.find_executable`
already honors the `PATHEXT` environment variable in windows systems.
See https://docs.python.org/3/library/shutil.html#shutil.which
Signed-off-by: Daniel Möller <n1ngu@riseup.net>
GitHub Actions workflow to create a release: will upload to PyPI
and create a GitHub release with the `sdist` and `bdist_wheel`
as well.
The version code is switched to `setuptools_scm` to work well
with this flow (e.g. avoid needing to write a script that does
a `sed` on the version file and commits as part of release).
Signed-off-by: Milas Bowman <milas.bowman@docker.com>
Allow an Image to be deleted by calling the remove() method on it,
just like a Volume.
Signed-off-by: Ahmon Dancy <dancy@dancysoft.com>
Signed-off-by: Milas Bowman <milas.bowman@docker.com>
Co-authored-by: Ahmon Dancy <dancy@dancysoft.com>
`rollback_config` was not in the list of `CREATE_SERVICE_KWARGS`
which prevented it from being an argument when creating services.
It has now been added and the problem fixed, allowing services to
have a rollback_config during creation and updating.
Fixes#2832.
Signed-off-by: Fraser Patten <pattenf00@gmail.com>
Signed-off-by: Milas Bowman <milas.bowman@docker.com>
Co-authored-by: Milas Bowman <milas.bowman@docker.com>
* Upgrade websocket-client to latest
* Add basic integration test for streaming logs via websocket
Signed-off-by: Milas Bowman <milas.bowman@docker.com>
Fix & enable SSH integration test suite.
This also adds a new test for connecting to unknown hosts when
using the Python SSH implementation (Paramiko). See #2932 for
more info.
Because of the above, some of the config/static key files have been
moved around and adjusted.
Signed-off-by: Milas Bowman <milas.bowman@docker.com>
Set `daemon` attribute instead of using `setDaemon` method that
was deprecated in Python 3.10.
Signed-off-by: Karthikeyan Singaravelan <tir.karthi@gmail.com>
This is no longer needed as it exists in every supported (non-EOL)
version of Python that we target.
Signed-off-by: Milas Bowman <milas.bowman@docker.com>
* Upgrade to latest Sphinx / recommonmark
* Small CSS fix for issue in new version of Alabaster theme
* Fix `Makefile` target for macOS
Signed-off-by: Milas Bowman <milas.bowman@docker.com>
This test looks for some behavior on non-chunked HTTP requests.
It now fails because it looks like recent versions of Docker Engine
ALWAYS return chunked responses (or perhaps this specific response
changed somehow to now trigger chunking whereas it did not previously).
The actual logic it's trying to test is also unusual because it's
trying to hackily propagate errors under the assumption that it'd get
a non-chunked response on failure, which is...not reliable. Arguably,
the chunked reader should be refactored somehow but that's a refactor
we can't really commit to (and it's evidently been ok enough as is
up until now).
Signed-off-by: Milas Bowman <milas.bowman@docker.com>
Ensure the container has exited before attempting to grab the logs.
Since we are not streaming them, it's possible to attach + grab logs
before the output is processed, resulting in a test failure. If the
container has exited, it's guaranteed to have logged :)
Signed-off-by: Milas Bowman <milas.bowman@docker.com>
This test was verifying that the container has the right options set (through
`docker inspect`), but also checks if the cgroup-rules are set within the
container by reading `/sys/fs/cgroup/devices/devices.list`
Unlike cgroups v1, on cgroups v2, there is no file interface, and rules are
handled through ebpf, which means that the test will fail because this file
is not present.
From the Linux documentation for cgroups v2:
https://github.com/torvalds/linux/blob/v5.16/Documentation/admin-guide/cgroup-v2.rst#device-controller
> (...)
> Device controller manages access to device files. It includes both creation of
> new device files (using mknod), and access to the existing device files.
>
> Cgroup v2 device controller has no interface files and is implemented on top
> of cgroup BPF. To control access to device files, a user may create bpf programs
> of type BPF_PROG_TYPE_CGROUP_DEVICE and attach them to cgroups with
> BPF_CGROUP_DEVICE flag. (...)
Given that setting the right cgroups is not really a responsibility of this SDK,
it should be sufficient to verify that the right options were set in the container
configuration, so this patch is removing the part that checks the cgroup, to
allow this test to be run on a host with cgroups v2 enabled.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This was using a deprecated function (`urllib.splitnport`),
ostensibly to work around issues with brackets on IPv6 addresses.
Ironically, its usage was broken, and would result in mangled IPv6
addresses if they had a port specified in some instances.
Usage of the deprecated function has been eliminated and extra test
cases added where missing. All existing cases pass as-is. (The only
other change to the test was to improve assertion messages.)
Signed-off-by: Milas Bowman <milas.bowman@docker.com>
ContainerSpec
Docker Engine v1.41 added `CapAdd` and `CapDrop` as part of the
ContainerSpec, and `docker-py` should do the same.
```
GET /services now returns CapAdd and CapDrop as part of the ContainerSpec.
GET /services/{id} now returns CapAdd and CapDrop as part of the ContainerSpec.
POST /services/create now accepts CapAdd and CapDrop as part of the ContainerSpec.
POST /services/{id}/update now accepts CapAdd and CapDrop as part of the ContainerSpec.
GET /tasks now returns CapAdd and CapDrop as part of the ContainerSpec.
GET /tasks/{id} now returns CapAdd and CapDrop as part of the ContainerSpec.
```
I added capabilities on docstrings, `service.create` init method and
create tests for that.
That change was mention in issue #2802.
Signed-off-by: Felipe Ruhland <felipe.ruhland@gmail.com>
How to reproduce the issue:
```py
>>> import docker
>>> cli = docker.from_env()
>>> cli.secrets.create(name="any_name", data="1")
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/home/docker-py/docker/models/secrets.py", line 10, in __repr__
return "<%s: '%s'>" % (self.__class__.__name__, self.name)
File "/home/docker-py/docker/models/secrets.py", line 14, in name
return self.attrs['Spec']['Name']
KeyError: 'Spec'
```
The exception raises because create secrets API `/secrets/create` only
return the `id` attribute:
https://docs.docker.com/engine/api/v1.41/#operation/SecretCreate
The secret model is created using just the `id` attribute and fails
when looking for Spec.Name attribute.
```py
def __repr__(self):
return "<%s: '%s'>" % (self.__class__.__name__, self.name)
```
```py
@property
def name(self):
return self.attrs['Spec']['Name']
```
I came up with a ugly solution but will prevent the problem to happen
again:
```py
def create(self, **kwargs):
obj = self.client.api.create_secret(**kwargs)
+ obj.setdefault("Spec", {})["Name"] = kwargs.get("name")
return self.prepare_model(obj)
```
After the API call, I added the name attribute to the right place to be
used on the property name.
```py
>>> import docker
>>> cli = docker.from_env()
>>> cli.secrets.create(name="any_name", data="1")
<Secret: 'any_name'>
```
It isn't the most elegant solution, but it will do the trick.
I had a previous PR #2517 when I propose using the `id` attribute
instead of `name` on the `__repr__` method, but I think this one will be better.
That fixes#2025
Signed-off-by: Felipe Ruhland <felipe.ruhland@gmail.com>
Mariano Scazzariello
Felix Fontein
Lorin Bucher
Andy Roxby
Milas Bowman
Yanlong Wang
loicleyendecker
Brian Goff
Rhiza
Sebastiaan van Stijn
Chris Hand
Quentin Mathorel
Leonard Kinday
Ningú
Clément Loiselet
ercildoune
Till!
David
Ben Fasoli
Karthikeyan Singaravelan
Hugo van Kemenade
Anca Iordache
Anthony Sottile
Sebastiano Mariani
Felipe Ruhland
Ulysses Souza
Nicolas De Loof