avro-to-python

avro-to-python is a light tool for compiling avro schema files (.avsc) to python classes making using avro schemata easy.

• Free software: MIT license
• Documentation: https://avro-to-python.readthedocs.io.

Installation

Pip install (recommended)

To install avro-to-python, run this command in your terminal:

$ pip install avro-to-python

Install From Source ()

The sources for avro-to-python can be downloaded source as well.

Clone the public repository:

$ git clone git://github.com/srserves85/avro-to-python

Once you have a copy of the source, you can install it with:

$ python setup.py install

or

$ pip install -e .

Examples

Majority of the use of avro-to-python is assumed to be used as a cli, but you can still import and use the python classes under the hood as well.

CLI (without --pip)

To use the cli, here is the available cli commands:

avro-to-python [source] [target]
    Options:
        --pip TEXT                Make package pip installable using this name.
        --top_level_package TEXT  Make package pip installable using this name.
        --author TEXT             Author name of the pip installable package.
        --package_version TEXT    Version of the pip installable package.  [default:
                        0.1.0]
        --encoding TEXT           Encoding to use for source and target (overrides
                                  system encoding).
        --help                    Show this message and exit.

The above will compile the avsc files and convert the to python classes found in [path_to_target_directory]

An example of doing this is the following:

avro-to-python [path_to_source_avsc_files] [path_to_target_directory]

If you run the above on a valid avro avsc file, you should then be able to import them as you would in the avro idl namespace Here is an example of a single avsc record from the namespace: name.space and name: RecordClass:

from name.space import RecordClass

record = RecordClass({'foo': True, 'bar': 'true', 'baz': 10, 'food': 'CHOCOLATE'})

Tips: To generate classes in a subpackage of your existing application set the "--top_level_package" flags to your subpackage name:

avro-to-python [path_to_source_avsc_files] [path_to_my_subpackage_directory] --top_level_package my.subpackage

from my.subpackage.name.space import RecordClass

record = RecordClass({'foo': True, 'bar': 'true', 'baz': 10, 'food': 'CHOCOLATE'})

CLI (with --pip)

You can also choose to make compiled avro packages *pip installable* by adding the "--pip" flags. An example of this is the following: .. code-block:: bash

avro-to-python [path_to_source_avsc_files] [path_to_target_directory] --pip test_avro

By running this, you should be able to pip install the above package you created from the target directory you specified by running:

pip install -e path_to_target_directory

Now that you have the package installed, you can import it by it's package name and namespace. Here is the same example of the same avsc from above, only with a pip package of test_avro:

from test_avro.name.space import RecordClass

record = RecordClass({'foo': True, 'bar': 'true', 'baz': 10, 'food': 'CHOCOLATE'})

You can customize the top level package name test_avro, modifying the "--top_level_package" flags:

avro-to-python [path_to_source_avsc_files] [path_to_target_directory] --pip test_avro --top_level_package event

from event.name.space import RecordClass

record = RecordClass({'foo': True, 'bar': 'true', 'baz': 10, 'food': 'CHOCOLATE'})

avro-to-python in a Script

You can also use the reader and writer packages in avro-to-python as you would any other python package. Avro to python is split between a reader and writer classes. avro-to-python treats namespaces as acyclic trees and uses depth first search to ensure no duplication or namespace collisions on read and write. An example usage is below:

from avro_to_python.reader import AvscReader
from avro_to_python.writer import AvroWriter

# initialize the reader object
reader = AvscReader(directory='tests/avsc/records/')

# generate the acyclic tree object
reader.read()

# initialize the writer object
writer = AvroWriter(reader.file_tree, pip='test_pip')

# compile python files using 'tests/test_records as the namespace root'
writer.write(root_dir='tests/test_records')

Roadmap

Reader

• [X] Create Namespace Trees on nested namespaces
• [X] Read Record and Enum File
• [X] Primitive types
• [X] Array Types
• [X] Union types
• [X] References to other files
• [X] Map Types
• [ ] Logical Types (Currently just converts to primitive types)

Writer

• [X] Base Schema Writer
• [X] Base Record Schema
• [X] Base Enum Schema
• [X] Primitive Types Schema
• [X] Array Types Schema
• [X] Union Types Schema
• [X] Map Types
• [ ] Logical Types Schema (Currently just converts to primitive types)
• [X] Add configs to pip install package

CLI

• [X] Wrap Writer and Reader into one cli commit
• [X] Add pip install option (would include all files to pip install compiled package)
• [ ] Add better --help documentation

Documentation

• [ ] Document reader class
• [ ] Document writer class
• [ ] Document cli