## Changes
Add JobTaskClusterSpec validate mutator. It catches the case when tasks
don't which cluster to use.
For example, we can get this error with minor modifications to
`default-python` template:
```yaml
tasks:
- task_key: python_file_task
spark_python_task:
python_file: ../src/my_project_10/main.py
```
```
% databricks bundle validate
Error: Missing required cluster or environment settings
at resources.jobs.my_project_10_job.tasks[0]
in resources/my_project_10_job.yml:17:11
Task "print_github_stars" requires a cluster or an environment to run.
Specify one of the following fields: job_cluster_key, environment_key, existing_cluster_id, new_cluster.
```
We implicitly rely on "one of" validation, which does not exist. Many
bundle fields can't co-exist, for instance, specifying:
`JobTask.{existing_cluster_id,job_cluster_key}`, `Library.{whl,pypi}`,
`JobTask.{notebook_task,python_wheel_task}`, etc.
## Tests
Unit tests
---------
Co-authored-by: Pieter Noordhuis <pcnoordhuis@gmail.com>
## Changes
- Extract sync output logic from `cmd/sync` into `lib/sync`
- Add hidden `verbose` flag to the `bundle deploy` command, it's false
by default and hidden from the `--help` output
- Pass output handler to the `deploy/files/upload` mutator if the
verbose option is true
The was an idea to use in-place output overriding each past file sync
event in the output, bit that wont work for the extension, since it
doesn't display deploy logs in the terminal.
Example output:
```
~/tmp/defpy: ~/cli/cli bundle deploy --sync-progress
Building defpy...
Uploading defpy-0.0.1+20240917.112755-py3-none-any.whl...
Uploading bundle files to /Users/ilia.babanov@databricks.com/.bundle/defpy/dev/files...
Action: PUT: requirements-dev.txt, resources/defpy_pipeline.yml, pytest.ini, src/defpy/main.py, src/defpy/__init__.py, src/dlt_pipeline.ipynb, tests/main_test.py, src/notebook.ipynb, setup.py, resources/defpy_job.yml, .vscode/extensions.json, .vscode/settings.json, fixtures/.gitkeep, .vscode/__builtins__.pyi, README.md, .gitignore, databricks.yml
Uploaded tests
Uploaded resources
Uploaded fixtures
Uploaded .vscode
Uploaded src/defpy
Uploaded requirements-dev.txt
Uploaded .gitignore
Uploaded fixtures/.gitkeep
Uploaded src/defpy/__init__.py
Uploaded databricks.yml
Uploaded README.md
Uploaded setup.py
Uploaded .vscode/__builtins__.pyi
Uploaded .vscode/extensions.json
Uploaded src/dlt_pipeline.ipynb
Uploaded .vscode/settings.json
Uploaded resources/defpy_job.yml
Uploaded pytest.ini
Uploaded src/defpy/main.py
Uploaded tests/main_test.py
Uploaded resources/defpy_pipeline.yml
Uploaded src/notebook.ipynb
Initial Sync Complete
Deploying resources...
Updating deployment state...
Deployment complete!
```
Output example in the extension:
<img width="1843" alt="Screenshot 2024-09-19 at 11 07 48"
src="https://github.com/user-attachments/assets/0fafd095-cdc6-44b8-b482-27a38ada0330">
## Tests
Manually for the `sync` and `bundle deploy` commands + vscode extension
sync and deploy flows
## Changes
Upgrade to TF provider 1.52
We also temporarily skip generating plugin framework structs to unblock
upgrade as generation does not work yet and need to be fixed separately
## Summary
Use the friendly name of service principals when shortening their name.
This change is helpful for the prefix in development mode. Instead of
adding a prefix like `[dev 1706906c-c0a2-4c25-9f57-3a7aa3cb8123]`, we'll
prefix like `[dev my_principal]`.
## Changes
This PR aliases and overrides the schema associated with the variables
block in `target` to allow for directly specifying a variable value in
the JSON schema (without an levels of nesting). This is needed because
this direct value is resolved by dynamically parsing the configuration
tree.
ca6332a5a4/bundle/config/root.go (L424)
## Tests
Existing unit tests.
## Changes
This PR makes sweeping changes to the way we generate and test the
bundle JSON schema. The main benefits are:
1. More modular JSON schema. Every definition in the schema now is one
level deep and points to references instead of inlining the entire
schema for a field. This unblocks PyDABs from taking a dependency on the
JSON schema.
2. Generate the JSON schema during CLI code generation. Directly stream
it instead of computing it at runtime whenever a user calls `databricks
bundle schema`. This is nice because we no longer need to embed a
partial OpenAPI spec in the CLI. Down the line, we can add a `Schema()`
method to every struct in the Databricks Go SDK and remove the
dependency on the OpenAPI spec altogether. It'll become more important
once we decouple Go SDK structs and methods from the underlying APIs.
3. Add enum values for Go SDK fields in the JSON schema. Better
autocompletion and validation for these fields. As a follow-up, we can
add enum values for non-Go SDK enums as well (created internal ticket to
track).
4. Use "packageName.structName" as a key to read JSON schemas from the
OpenAPI spec for Go SDK structs. Before, we would use an unrolled
presentation of the JSON schema (stored in `bundle_descriptions.json`),
which was complex to parse and include in the final JSON schema output.
This also means loading values from the OpenAPI spec for `target` schema
works automatically and no longer needs custom code.
5. Support recursive types (eg: `for_each_task`). With us now using
$refs everywhere it's trivial to support.
6. Using complex variables would be invalid according to the schema
generated before this PR. Now that bug is fixed. In the future adding
more custom rules will be easier as well due to the single level nature
of the JSON schema.
Since this is a complete change of approach in how we generate the JSON
schema, there are a few (very minor) regressions worth calling out.
1. We'll lose a few custom descriptions for non Go SDK structs that were
a part of `bundle_descriptions.json`. Support for those can be added in
the future as a followup.
2. Since now the final JSON schema is a static artefact, we lose some
lead time for the signal that JSON schema integration tests are failing.
It's okay though since we have a lot of coverage via the existing unit
tests.
## Tests
Unit tests. End to end tests are being added in this PR:
https://github.com/databricks/cli/pull/1726
Previous unit tests were all deleted because they were bloated. Effort
was made to make the new unit tests provide (almost) equivalent
coverage.
## Changes
Library glob expansion happens during deployment. Before that, all
entries that refer to local paths in resource definitions are made
relative to the _sync root_. Before #1694, they were made relative to
the _bundle root_. This PR didn't update the library glob expansion code
to use the sync root path.
If you were using the sync paths setting with library globs, the CLI
would fail to expand the globs because the code was using the wrong path
to anchor those globs.
This change fixes the issue.
## Tests
Manually confirmed that this fixes the issue reported in #1755.
## Changes
We added a custom resolver for the cluster to add filtering for the
cluster source when we list all clusters.
Without the filtering listing could take a very long time (5-10 mins)
which leads to lookup timeouts.
## Tests
Existing unit tests passing
## Changes
Some call sites hold on to the `dyn.Path` provided to them by the
callback. It must therefore never be mutated after the callback returns,
or these mutations leak out into unknown scope.
This change means it is no longer possible for this failure mode to
happen.
## Tests
Unit test.
## Changes
Fixes an `Error: no value assigned to required variable <variable>.`
when the main complex variable definition is defined in one file but
target override is defined in separate file which is included in the
main one.
## Tests
Added regression test
## Changes
DLT pipeline recreations are destructive. They can lead to lost history
of previous updates, outage of the tables temporarily and are
potentially computationally expensive. Thus we make a breaking change
where a prompt is shown to the user if there configuration changes will
lead to a DLT recreation.
Users can skip the prompt by specifying the `--auto-approve` flag.
This PR also fixes an issue with our test runner where logs from the
cmdio.Logger would not get propagated to the reader returned by our
cobra test runner.
## Tests
Manually, and new unit and integration tests.
```
➜ bundle-playground-3 cli bundle deploy
Uploading bundle files to /Users/63ec021d-b0c6-49c0-93a0-5123953a1cb2/.bundle/test/development/files...
The following DLT pipelines will be recreated. Underlying tables will be unavailable for a transient period until the newly recreated pipelines are run once successfully. History of previous pipeline update runs will be lost because of recreation:
recreate pipeline foo
Would you like to proceed? [y/n]: n
Deployment cancelled!
```
## Changes
We were not using the readers and writers set in the test fixtures in
the progress logger. This PR fixes that. It also modifies
`TestAccAbortBind`, which was implicitly relying on the bug.
I encountered this bug while working on
https://github.com/databricks/cli/pull/1672.
## Tests
Manually.
From non-tty:
```
Error: failed to bind the resource, err: This bind operation requires user confirmation, but the current console does not support prompting. Please specify --auto-approve if you would like to skip prompts and proceed.
```
From tty, bind works as expected.
```
Confirm import changes? Changes will be remotely applied only after running 'bundle deploy'. [y/n]: y
Updating deployment state...
Successfully bound databricks_pipeline with an id '9d2dedbb-f522-4503-96ba-4bc4d5bfa77d'. Run 'bundle deploy' to deploy changes to your workspace
```
## Changes
Explain the error when the `databricks-pydabs` package is not installed
or the Python environment isn't correctly activated.
Example output:
```
Error: python mutator process failed: ".venv/bin/python3 -m databricks.bundles.build --phase load --input .../input.json --output .../output.json --diagnostics .../diagnostics.json: exit status 1", use --debug to enable logging
.../.venv/bin/python3: Error while finding module specification for 'databricks.bundles.build' (ModuleNotFoundError: No module named 'databricks')
Explanation: 'databricks-pydabs' library is not installed in the Python environment.
If using Python wheels, ensure that 'databricks-pydabs' is included in the dependencies,
and that the wheel is installed in the Python environment:
$ .venv/bin/pip install -e .
If using a virtual environment, ensure it is specified as the venv_path property in databricks.yml,
or activate the environment before running CLI commands:
experimental:
pydabs:
venv_path: .venv
```
## Tests
Unit tests
## Changes
Preserve diagnostics if there are any errors or warnings when
PythonMutator normalizes output. If anything goes wrong during
conversion, diagnostics contain the relevant location and path.
## Tests
Unit tests
## Changes
This ensures that the CLI and Terraform can both use an Azure CLI
session configured under a non-standard path. This is the default
behavior on Azure DevOps when using the AzureCLI@2 task.
Fixes#1722.
## Tests
Unit test.
## Changes
Consider serverless clusters as compatible for Python wheel tasks.
Fixes a `Python wheel tasks require compute with DBR 13.3+ to include
local libraries` warning shown for serverless clusters
## Changes
* Provide a more helpful error when using an artifact_path based on
/Volumes
* Allow the use of short_names in /Volumes paths
## Example cases
Example of a valid /Volumes artifact_path:
* `artifact_path:
/Volumes/catalog/schema/${workspace.current_user.short_name}/libs`
Example of an invalid /Volumes path (when using `mode: development`):
* `artifact_path: /Volumes/catalog/schema/libs`
* Resulting error: `artifact_path should contain the current username or
${workspace.current_user.short_name} to ensure uniqueness when using
'mode: development'`
## Changes
Fixes issue introduced here https://github.com/databricks/cli/pull/1699
where PyPi packages were treated as local library.
The reason is that `libraryPath` returns an empty string as a path for
PyPi packages and then `IsLibraryLocal` treated empty string as local
path.
Both of these functions are fixed in this PR.
## Tests
Added regression test
## Changes
This changes makes sure we ignore CLI version check on development
builds of the CLI.
Before:
```
$ cat databricks.yml | grep cli_version
databricks_cli_version: ">= 0.223.1"
$ cli bundle deploy
Error: Databricks CLI version constraint not satisfied. Required: >= 0.223.1, current: 0.0.0-dev+06b169284737
```
after
```
...
$ cli bundle deploy
...
Warning: Ignoring Databricks CLI version constraint for development build. Required: >= 0.223.1, current: 0.0.0-dev+d52d6f08fcd5
```
## Tests
<!-- How is this tested? -->
## Changes
This field allows a user to configure paths to synchronize to the
workspace.
Allowed values are relative paths to files and directories anchored at
the directory where the field is set. If one or more values traverse up
the directory tree (to an ancestor of the bundle root directory), the
CLI will dynamically determine the root path to use to ensure that the
file tree structure remains intact.
For example, given a `databricks.yml` in `my_bundle` that includes:
```yaml
sync:
paths:
- ../common
- .
```
Then upon synchronization, the workspace will look like:
```
.
├── common
│ └── lib.py
└── my_bundle
├── databricks.yml
└── notebook.py
```
If not set behavior remains identical.
## Tests
* Newly added unit tests for the mutators and under `bundle/tests`.
* Manually confirmed a bundle without this configuration works the same.
* Manually confirmed a bundle with this configuration works.
## Changes
In https://github.com/databricks/cli/pull/1490 we regressed and started
using the development mode prefix for UC schemas regardless of the mode
of the bundle target.
This PR fixes the regression and adds a regression test
## Tests
Failing integration tests pass now.
## Changes
While experimenting with DAB I discovered that requirements libraries
are being ignored.
One thing worth mentioning is that `bundle validate` runs successfully,
but `bundle deploy` fails. This PR only covers the second part.
## Tests
<!-- How is this tested? -->
Added a unit test
## Changes
Make `pydabs/venv_path` optional. When not specified, CLI detects the
Python interpreter using `python.DetectExecutable`, the same way as for
`artifacts`. `python.DetectExecutable` works correctly if a virtual
environment is activated or `python3` is available on PATH through other
means.
Extract the venv detection code from PyDABs into `libs/python/detect`.
This code will be used when we implement the `python/venv_path` section
in `databricks.yml`.
## Tests
Unit tests and manually
---------
Co-authored-by: Pieter Noordhuis <pcnoordhuis@gmail.com>
## Changes
These tests inadvertently re-ran mutators, the first time through
`loadTarget` and the second time by running `phases.Initialize()`
themselves. Some of the mutators that are executed in
`phases.Initialize()` are also run as part of `loadTarget`. This is
overdue a refactor to make it unambiguous what runs when. Until then,
this removes the duplicated execution.
## Tests
Unit tests pass.
## Changes
This PR addressed post-merge feedback from
https://github.com/databricks/cli/pull/1673.
## Tests
Unit tests, and manually.
```
Error: experiment undefined-experiment is not defined
at resources.experiments.undefined-experiment
in databricks.yml:11:26
Error: job undefined-job is not defined
at resources.jobs.undefined-job
in databricks.yml:6:19
Error: pipeline undefined-pipeline is not defined
at resources.pipelines.undefined-pipeline
in databricks.yml:14:24
Name: undefined-job
Target: default
Found 3 errors
```
## Changes
This adds configurable transformations based on the transformations
currently seen in `mode: development`.
Example databricks.yml showcasing how some transformations:
```
bundle:
name: my_bundle
targets:
dev:
presets:
prefix: "myprefix_" # prefix all resource names with myprefix_
pipelines_development: true # set development to true by default for pipelines
trigger_pause_status: PAUSED # set pause_status to PAUSED by default for all triggers and schedules
jobs_max_concurrent_runs: 10 # set max_concurrent runs to 10 by default for all jobs
tags:
dev: true
```
## Tests
* Existing process_target_mode tests that were adapted to use this new
code
* Unit tests specific for the new mutator
* Unit tests for config loading and merging
* Manual e2e testing
## Changes
Before this change, the fileset library would take a single root path
and list all files in it. To support an allowlist of paths to list (much
like a Git `pathspec` without patterns; see [pathspec](pathspec)), this
change introduces an optional argument to `fileset.New` where the caller
can specify paths to list. If not specified, this argument defaults to
list `.` (i.e. list all files in the root).
The motivation for this change is that we wish to expose this pattern in
bundles. Users should be able to specify which paths to synchronize
instead of always only synchronizing the bundle root directory.
[pathspec]:
https://git-scm.com/docs/gitglossary#Documentation/gitglossary.txt-aiddefpathspecapathspec
## Tests
New and existing unit tests.
Gleb Kanterov
Andrew Nester
Gleb Kanterov
Ilia Babanov
Lennart Kats (databricks)
shreyas-goenka
dependabot[bot]
Pieter Noordhuis
Witold Czaplewski