This fixes a case where a non-tty will have control characters + the log
line for every single read operation.
Signed-off-by: Brian Goff <cpuguy83@gmail.com>
This code depended on the registry Service interface, which has been removed,
so needed to be refactored. Digging further into the reason this code existed,
it looked like the Class=plugin was previously required on Docker Hub to handle
plugins, but this requirement is no longer there, so we can remove this special
handling.
This patch removes the special handling to both remove the use of the registry.Service
interface, as well as removing complexity that is no longer needed.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This function was deprecated in b4ca1c7368,
which is part of the v23.0 release, and is no longer used, so we can remove it.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The comment was not formatted correctly, and because of that not picked up as
being deprecated.
updates b4ca1c7368
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
These tests were deliberately producing errors as part of the test, but
printing those errors could be confusing / make it more difficult to find
actual test-failures.
Before this patch:
=== RUN TestVolumeCreateErrors
Error: conflicting options: either specify --name or provide positional arg, not both
Error: "create" requires at most 1 argument.
See 'create --help'.
Usage: create [OPTIONS] [VOLUME] [flags]
Create a volume
Error: error creating volume
--- PASS: TestVolumeCreateErrors (0.00s)
PASS
With this patch applied:
=== RUN TestVolumeCreateErrors
--- PASS: TestVolumeCreateErrors (0.00s)
PASS
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Make the error more specific by stating that it's caused by a specific
environment variable and not an environment as a whole.
Also don't escape the variable to make it more readable.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
On Windows, ignore all variables that start with "=" when building an
environment variables map for stack.
For MS-DOS compatibility cmd.exe can set some special environment
variables that start with a "=" characters, which breaks the general
assumption that the first encountered "=" separates a variable name from
variable value and causes trouble when parsing.
These variables don't seem to be documented anywhere, but they are
described by some third-party sources and confirmed empirically on my
Windows installation.
Useful sources:
https://devblogs.microsoft.com/oldnewthing/20100506-00/?p=14133https://ss64.com/nt/syntax-variables.html
Known variables:
- `=ExitCode` stores the exit code returned by external command (in hex
format)
- `=ExitCodeAscii` - same as above, except the value is the ASCII
representation of the code (so exit code 65 (0x41) becomes 'A').
- `=::=::\` and friends - store drive specific working directory.
There is one env variable for each separate drive letter that was
accessed in the shell session and stores the working directory for that
specific drive.
The general format for these is:
`=<DRIVE_LETTER>:=<CWD>` (key=`=<DRIVE_LETTER>:`, value=`<CWD>`)
where <CWD> is a working directory for the drive that is assigned to
the letter <DRIVE_LETTER>
A couple of examples:
`=C:=C:\some\dir` (key: `=C:`, value: `C:\some\dir`)
`=D:=D:\some\other\dir` (key: `=C:`, value: `C:\some\dir`)
`=Z:=Z:\` (key: `=Z:`, value: `Z:\`)
`=::=::\` is the one that seems to be always set and I'm not exactly
sure what this one is for (what's drive `::`?). Others are set as
soon as you CD to a path on some drive. Considering that you start a
cmd.exe also has some working directory, there are 2 of these on start.
All these variables can be safely ignored because they can't be
deliberately set by the user, their meaning is only relevant to the
cmd.exe session and they're all are related to the MS-DOS/Batch feature
that are irrelevant for us.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
Tests mocking the output of GET images/json with fakeClient used an
array with one empty element as an empty response.
Change it to just an empty array.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
The error returned from "os/exec".Command when attempting to execute a
directory has been changed from syscall.EACCESS to syscall.EISDIR on
Go 1.20. 2b8f214094
Consequently, any runc runtime built against Go 1.20 will return an
error containing 'is a directory' and not 'permission denied'. Update
the string matching so the CLI exits with status code 126 on 'is a
directory' errors (EISDIR) in addition to 'permission denied' (EACCESS).
Signed-off-by: Cory Snider <csnider@mirantis.com>
This prevents us needing to attempt to reconstruct the exact indentation
registry side, which is not canonical - so may differ.
Signed-off-by: Justin Chadwell <me@jedevc.com>
This behavior should not break any more use cases than before.
Previously, if the mismatch occured, we would actually push a manifest
that we then never referred to in the manifest list! If this was done in
a new repository, the command would fail with an obscure error from the
registry - the content wouldn't exist with the descriptor we expect it
to.
Signed-off-by: Justin Chadwell <me@jedevc.com>
The DockerCLI interface was repeating the Streams interface. Embed
the interface to make it more transparent that they're the same.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Synchronize append on the `removed` slice with mutex because
containerRemoveFunc is called in parallel for each removed container by
`container rm` cli command.
Also reduced the shared access area by separating the scopes of test
cases.
Signed-off-by: Paweł Gronowski <pawel.gronowski@docker.com>
This comment was added in 7929888214
when this code was still in the Moby repository. That comment doesn't appear
to apply to the CLI's usage of this struct though, as nothing in the CLI
sets this field (or uses it), so this should be safe to remove.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This allows the cli to be initialized with a (custom) API client.
Currently to be used for unit tests, but could be used for other
scenarios.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Make sure that the container has multiple port-mappings to illustrate
that only the given port is matched.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
- use strings.Cut
- don't use nat.NewPort as we don't accept port ranges
- use an early return if there's no results
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
if a context is set (e.g. through DOCKER_CONTEXT or the CLI config file), but
wasn't found, then a "stub" context is added, including an error message that
the context doesn't exist.
DOCKER_CONTEXT=nosuchcontext docker context ls
NAME DESCRIPTION DOCKER ENDPOINT ERROR
default Current DOCKER_HOST based configuration unix:///var/run/docker.sock
nosuchcontext * context "nosuchcontext": context not found: …
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This updates `docker context ls` to:
- not abort listing contexts when failing one (or more) contexts
- instead, adding an ERROR column to inform the user there was
an issue loading the context.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This allows commands that don't require a client connection (such as `context use`)
to be functional, but still produces an error when trying to run a command that
needs to connect with the API;
mkdir -p ~/.docker/ && echo '{"currentContext":"nosuchcontext"}' > ~/.docker/config.json
docker version
Failed to initialize: unable to resolve docker endpoint: load context "nosuchcontext": context does not exist: open /root/.docker/contexts/meta/8bfef2a74c7d06add4bf4c73b0af97d9f79c76fe151ae0e18b9d7e57104c149b/meta.json: no such file or directory
docker context use default
default
Current context is now "default"
docker version
Client:
Version: 22.06.0-dev
API version: 1.42
...
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The "docker context show" command is intended to show the currently configured
context. While the context that's configured may not be valid (e.g., in case
an environment variable was set to configure the context, or if the context
was removed from the filesystem), we should still be able to _show_ the
context.
This patch removes the context validation, and instead only shows the context.
This can help in cases where the context is used to (e.g.) set the command-
prompt, but the user removed the context. With this change, the context name
can still be shown, but commands that _require_ the context will still fail.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This internalizes constructing the Client(), which allows us to provide
fallbacks when trying to determin the current API version.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Also move the resolveContextName() function together with the
method for easier cross-referencing.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
There's no strict need to perform this validation inside this function;
validating flags should happen earlier, to allow faster detecting of
configuration issues (we may want to have a central config "validate"
function though).
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
resolveContextName() is used to find which context to use, based on the
available configuration options. Once resolved, the context name is
used to load the actual context, which will fail if the context doesn't
exist, so there's no need to produce an error at this stage; only
check priority of the configuration options to pick the context
with the highest priority.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
CommonOptions was inherited from when the cli and daemon were in the same
repository, and some options would be shared between them. That's no longer
the case, and some options are even "incorrect" (for example, while the
daemon can be configured to run on multiple hosts, the CLI can only connect
with a single host / connection). This patch does not (yet) address that,
but merges the CommonOptions into the ClientOptions.
An alias is created for the old type, although it doesn't appear there's
any external consumers using the CommonOptions type (or its constructor).
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
- Make the package-level configMergeTests local to the test itself.
- Rename fields to better describe intent
- Remove some redundant variables
- Reverse "expected" and "actual" fields for consistency
- Use assert.Check() to not fail early
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Looks like the linter uses an explicit -lang, which (for go1.19)
results in some additional formatting for octal values.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The package defined various special errors; these errors existed for two reasons;
- being able to distinguish "not found" errors from other errors (as "not found"
errors can be ignored in various cases).
- to be able to update the context _name_ in the error message after the error
was created. This was needed in cases where the name was not available at the
location where the error was produced (e.g. only the "id" was present), and
the helpers to detect "not found" errors did not support wrapped errors (so
wrapping the error with a "name" could break logic); a `setContextName` interface
and corresponding `patchErrContextName()` utility was created for this (which
was a "creative", but not very standard approach).
This patch:
- Removes the special error-types, replacing them with errdefs definitions (which
is a more common approach in our code-base to detect error types / classes).
- Removes the internal utilities for error-handling, and deprecates the exported
utilities (to allow external consumers to adjust their code).
- Some errors have been enriched with detailed information (which may be useful
for debugging / problem solving).
- Note that in some cases, `patchErrContextName()` was called, but the code
producing the error would never return a `setContextName` error, so would
never update the error message.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Before this change, running `docker context rm --force` would fail if the context
did not exist. This behavior was different from other commands, which allowed
ignoring non-existing objects.
For example; when trying to remove a non-existing volume, the command would
fail without "force":
```bash
docker volume rm nosuchvolume
Error: No such volume: nosuchvolume
echo $?
1
```
But using the `-f` / `--force` option would make the command complete successfully
(the error itself is still printed for informational purposes);
```bash
docker volume rm -f nosuchvolume
nosuchvolume
echo $?
0
```
With this patch, `docker context rm` behaves the same:
```bash
docker context rm nosuchcontext
context "nosuchcontext" does not exist
echo $?
1
```
```bash
docker context rm -f nosuchcontext
nosuchcontext
echo $?
0
```
This patch also simplifies how we check if the context exists; previously we
would try to read the context's metadata; this could fail if a context was
corrupted, or if an empty directory was present. This patch now only checks
if the directory exists, without first validating the context's data.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Also removing redundant defer for env.PatchAll(), which is now automatically
handled in t.Cleanup()
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
cli/command/container/opts.go:928:2: assigned to src, but reassigned without using the value (wastedassign)
src := ""
^
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
While fixing, also updated errors without placeholders to `errors.New()`, and
updated some code to use pkg/errors if it was already in use in the file.
cli/command/config/inspect.go:59:10: ST1005: error strings should not be capitalized (stylecheck)
return fmt.Errorf("Cannot supply extra formatting options to the pretty template")
^
cli/command/node/inspect.go:61:10: ST1005: error strings should not be capitalized (stylecheck)
return fmt.Errorf("Cannot supply extra formatting options to the pretty template")
^
cli/command/secret/inspect.go:57:10: ST1005: error strings should not be capitalized (stylecheck)
return fmt.Errorf("Cannot supply extra formatting options to the pretty template")
^
cli/command/trust/common.go:77:74: ST1005: error strings should not be capitalized (stylecheck)
return []trustTagRow{}, []client.RoleWithSignatures{}, []data.Role{}, fmt.Errorf("No signatures or cannot access %s", remote)
^
cli/command/trust/common.go:85:73: ST1005: error strings should not be capitalized (stylecheck)
return []trustTagRow{}, []client.RoleWithSignatures{}, []data.Role{}, fmt.Errorf("No signers for %s", remote)
^
cli/command/trust/sign.go:137:10: ST1005: error strings should not be capitalized (stylecheck)
return fmt.Errorf("No tag specified for %s", imgRefAndAuth.Name())
^
cli/command/trust/sign.go:151:19: ST1005: error strings should not be capitalized (stylecheck)
return *target, fmt.Errorf("No tag specified")
^
cli/command/trust/signer_add.go:77:10: ST1005: error strings should not be capitalized (stylecheck)
return fmt.Errorf("Failed to add signer to: %s", strings.Join(errRepos, ", "))
^
cli/command/trust/signer_remove.go:52:10: ST1005: error strings should not be capitalized (stylecheck)
return fmt.Errorf("Error removing signer from: %s", strings.Join(errRepos, ", "))
^
cli/command/trust/signer_remove.go:67:17: ST1005: error strings should not be capitalized (stylecheck)
return false, fmt.Errorf("All signed tags are currently revoked, use docker trust sign to fix")
^
cli/command/trust/signer_remove.go:108:17: ST1005: error strings should not be capitalized (stylecheck)
return false, fmt.Errorf("No signer %s for repository %s", signerName, repoName)
^
opts/hosts.go:89:14: ST1005: error strings should not be capitalized (stylecheck)
return "", fmt.Errorf("Invalid bind address format: %s", addr)
^
opts/hosts.go💯14: ST1005: error strings should not be capitalized (stylecheck)
return "", fmt.Errorf("Invalid proto, expected %s: %s", proto, addr)
^
opts/hosts.go:119:14: ST1005: error strings should not be capitalized (stylecheck)
return "", fmt.Errorf("Invalid proto, expected tcp: %s", tryAddr)
^
opts/hosts.go:144:14: ST1005: error strings should not be capitalized (stylecheck)
return "", fmt.Errorf("Invalid bind address format: %s", tryAddr)
^
opts/hosts.go:155:14: ST1005: error strings should not be capitalized (stylecheck)
return "", fmt.Errorf("Invalid bind address format: %s", tryAddr)
^
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
We try to keep this package close to upstream golang's code, so suppress the
linter warning.
cli/command/formatter/tabwriter/tabwriter.go:200:1: ST1020: comment on exported method Init should be of the form "Init ..." (stylecheck)
// A Writer must be initialized with a call to Init. The first parameter (output)
^
cli/command/formatter/tabwriter/tabwriter.go:425:1: ST1022: comment on exported const Escape should be of the form "Escape ..." (stylecheck)
// To escape a text segment, bracket it with Escape characters.
^
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
cli/command/cli_options_test.go:29:2: os.Setenv() can be replaced by `t.Setenv()` in TestWithContentTrustFromEnv (tenv)
os.Setenv(envvar, "true")
^
cli/command/cli_options_test.go:31:2: os.Setenv() can be replaced by `t.Setenv()` in TestWithContentTrustFromEnv (tenv)
os.Setenv(envvar, "false")
^
cli/command/cli_options_test.go:33:2: os.Setenv() can be replaced by `t.Setenv()` in TestWithContentTrustFromEnv (tenv)
os.Setenv(envvar, "invalid")
^
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
cli/command/manifest/inspect_test.go:9:2: ST1019: package "github.com/docker/cli/cli/manifest/types" is being imported more than once (stylecheck)
"github.com/docker/cli/cli/manifest/types"
^
cli/command/manifest/inspect_test.go:10:2: ST1019(related information): other import of "github.com/docker/cli/cli/manifest/types" (stylecheck)
manifesttypes "github.com/docker/cli/cli/manifest/types"
^
cli/command/stack/swarm/deploy_composefile.go:14:2: ST1019: package "github.com/docker/docker/client" is being imported more than once (stylecheck)
apiclient "github.com/docker/docker/client"
^
cli/command/stack/swarm/deploy_composefile.go:15:2: ST1019(related information): other import of "github.com/docker/docker/client" (stylecheck)
dockerclient "github.com/docker/docker/client"
^
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
cli/command/image/build/context.go:238:23: "400" can be replaced by http.StatusBadRequest (usestdlibvars)
if resp.StatusCode < 400 {
^
cli/trust/trust.go:139:30: "GET" can be replaced by http.MethodGet (usestdlibvars)
req, err := http.NewRequest("GET", endpointStr, nil)
^
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Having the intermediate variable made it difficult to see if it was
possibly mutated and/or something special done with it, so just use
the cli's accessors to get its Err().
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
As it's just an alias for filepath.IsAbs. Also added a normalize step in
TrimBuildFilesFromExcludes, so that callers are not _required_ to first
normalize the path.
We are considering deprecating and/or removing this function in the archive
package, so removing it in the cli code helps transitioning if we decide to
deprecate and/or remove it.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
`NewDockerCli` was configuring the standard streams using local code; this patch
instead uses the available `WithStandardStreams()` option to do the same.
There is slight difference in the order of events;
Previously, user-provided options would be applied first, after which NewDockerCli
would check if any of "in", "out", or "err" were nil, and if so set them to the
default stream (or writer) for that output.
The new code unconditionally sets the defaults _before_ applying user-provided
options. In practive, howver, this makes no difference; the fields set are not
exported, and the only functions updating them are `WithStandardStreams`,
`WithInputStream`, and `WithCombinedStream`, neither of which checks the old
value (so always overrides).
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Migrating these functions to allow them being shared between moby, docker/cli,
and containerd, and to allow using them without importing all of sys / system,
which (in containerd) also depends on hcsshim and more.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
- moby: a60b458179...d2590dc3cd
- swarmkit: 6068d1894d...48dd89375d
The .Parent field for buildcache entries was deprecated, and replaced with a
.Parents (plural) field. This patch updates the code accordingly. Unlike the
change in buildx
9c3be32bc9
we continue to fall back to the old field (which will be set on older API
versions).
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Note that this does not fully fix the referenced issue, but
at least makes sure that API clients don't hang forever on
the initialization step.
See: https://github.com/docker/cli/issues/3652
Signed-off-by: Nick Santos <nick.santos@docker.com>
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Avoid updating the config-file if nothing changed. This also prevents creating
the file and config-directory if the default is used and no config-file existed
yet.
`config.Save()` performs various steps (creating the directory, updating
or copying permissions, etc etc), which are not needed if the defaults are
used; a445d97c25/cli/config/configfile/file.go (L135-L176)
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
This code was handling validation and parsing, only to discard the
results if it was the default context.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Older versions of Go do not format these comments, so we can already
reformat them ahead of time to prevent gofmt linting failing once
we update to Go 1.19 or up.
Result of:
gofmt -s -w $(find . -type f -name '*.go' | grep -v "/vendor/")
With some manual adjusting.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
Brian Goff
Sebastiaan van Stijn
Sebastiaan van Stijn
Paweł Gronowski
Laura Brehm
Cory Snider
Akihiro Suda
Justin Chadwell
CrazyMax
Jason Hall
Nick Santos
Nick Santos