snow app bundle

Prepares a local folder with configured app artifacts.

Syntax

snow app bundle
  --package-entity-id <package_entity_id>
  --app-entity-id <app_entity_id>
  --project <project_definition>
  --env <env_overrides>
  --format <format>
  --verbose
  --debug
  --silent
Copy

Arguments

None

Options

--package-entity-id TEXT

The ID of the package entity on which to operate when definition_version is 2 or higher.

--app-entity-id TEXT

The ID of the application entity on which to operate when definition_version is 2 or higher.

-p, --project TEXT

Path where Snowflake project resides. Defaults to current working directory.

--env TEXT

String in format of key=value. Overrides variables from env section used for templates. Default: [].

--format [TABLE|JSON]

Specifies the output format. Default: TABLE.

--verbose, -v

Displays log entries for log levels info and higher. Default: False.

--debug

Displays log entries for log levels debug and higher; debug logs contain additional information. Default: False.

--silent

Turns off intermediate output to console. Default: False.

--help

Displays the help text for this command.

Usage notes

The snow app bundle command creates a temporary local directory that contains all of the Snowflake Native App artifacts. It can also automatically generate SQL scripts from your Snowpark Python code. This command is called automatically by the snow app deploy, snow app run, and snow app version create commands. If, however, you want to see the setup script, artifacts, and generated SQL, before uploading them to a stage, you can run this command manually. For more information about generating SQL code, see Preparing a local folder with configured Snowflake Native App artifacts.

  • The command uses the project definition file to determine the name of the temporary folder to create within your project directory.

    • By default, it will be <project_directory>/output/deploy. This directory, which is also known as the deploy root, mirrors what the structure of the stage will be, once files are uploaded to stage in subsequent commands.

    • If you want Snowflake CLI to create a folder with a custom name instead of output/deploy, you can do so by providing the deploy_root field on the application package entity in the project definition file.

      Note

      You must provide a relative path for the deploy root; absolute paths are rejected. The deploy root path is created inside the project directory.

    • The deploy root is a temporary directory because it gets deleted and recreated every you run snow app bundle or another command invokes the bundle functionality.

  • Because snow app bundle is automatically called as part of he snow app deploy, snow app run, and snow app version create commands, you should make changes to the source files only, outside the deploy root. If you modify files in the deploy root, the files are overwritten by the most recent state of your source files the next time you call one of these commands.

  • If using a version control system such as git, you can choose to not to track the deploy root, as it can change frequently.

  • snow app bundle does not build or compile your artifacts for you, such as creating jar files from your Java files. It only copies the artifacts specified in the project definition file and adds them to the deploy root to mimic the stage’s directory structure.

  • snow app bundle does not need access to your Snowflake account; it only affects your local filesystem.

  • The command has the following copying and symlinking behavior for any artifacts of the application package entity in the project definition file:

    • All directory names in a source path are also created in the deploy root.

    • All files in a source path are symlinked within these directories in the deploy root.

    • Some symlinked files in the deploy root can become hard links if you invoke SQL generation from those files. For more information, see Preparing a local folder with configured Snowflake Native App artifacts.

    Consider the following artifacts list example from a project definition file:

    entities:
  pkg:
    type: application package
    ...
    artifacts:
      - src: dir1/dir2/*
        dest: dest_dir1/dest_dir2/
      - src: dir8/dir9/file.txt
        dest: dest_dir8/dest_file.txt
  ...
    Copy

    where dir1/dir2 in the project root could have other subdirectories, such as dir3 and dir4, and some files, such as file3.txt and file4.txt.

    After running the snow app bundle command, your deploy root should look like the following:

    -- deploy_root
      -- dest_dir1
            -- dest_dir2
                  -- dir3
                      -- ... <entire directory tree of dir3>
                  -- dir4
                      -- ... <entire directory tree of dir4>
                  -- file3.txt
                  -- file4.txt
      -- dest_dir8
            -- dest_file.txt
    Copy

Snowpark annotation processing

Beginning with Snowflake CLI version 2.5.0 and Snowpark Python API version 1.15.0, you can leverage the Snowpark annotation processing feature with the snow app bundle command. This feature lets you annotate your Python code files with Snowpark Python decorators, such as @udf, @sproc, @udaf, and @udtf to let Snowflake CLI automatically the corresponding CREATE FUNCTION or CREATE PROCEDURE SQL statements in setup script files in the project directory. For a better understanding of these decorators, please refer to corresponding Python decorators documentation.

Snowpark annotation processing involves the following:

  • It reads all Python files you marked with a processor field in the project definition file.

  • It creates a separate temporary sandbox Python environment using the environment information provided in the processor’s properties sub-field.

  • It executes those Python files in the sandboxed environment.

  • It collects all decorated functions from those files.

  • With the collected information, Snowflake CLI generates the necessary SQL statements and adds them to the setup script whose location is specified in your manifest.yaml file.

You no longer need to repeat boilerplate SQL code for writing Snowpark extension functions for your Snowflake Native App apps.

For more information about enabling this feature in your project definition files, see Using the Snowpark Python decorators.

For more information about all supported artifact processors, see More information about artifacts processors.

Examples

This example assumes you have made the necessary changes to your code files and added them to your snowflake.yml or snowflake.local.yml files, and also built or compiled any relevant artifacts.

cd my_app_project
snow app bundle
Copy

The command displays information about the various steps that occur while the command runs and creates a new directory a the location specified in your project definition file (default: my_app_project/output/deploy).

To see a simple use case in action, you can leverage the ready-to-use templates using the following commands:

snow init my_app_bundle_project --template app_basic
cd "my_app_bundle_project"
snow app bundle
ls my_app_bundle_project/output/deploy
Copy
Language: English