.. jsonseq documentation master file, created by sphinx-quickstart on Thu Sep 20 08:53:37 2018. You can adapt this file completely to your liking, but it should at least contain the root `toctree` directive. jsonseq ======= `RFC 7464 JSON Text Sequences `__ encoding and decoding for Python. .. image:: https://travis-ci.com/sgillies/jsonseq.svg?branch=master :target: https://travis-ci.com/sgillies/jsonseq :alt: Build Status .. image:: https://coveralls.io/repos/github/sgillies/jsonseq/badge.svg?branch=master :target: https://coveralls.io/github/sgillies/jsonseq?branch=master :alt: Coverage Status .. toctree:: :maxdepth: 2 :caption: Contents: index Introduction ============ The JSONSeqEncoder class takes streams of JSON-serializable Python objects and yields for each object its JSON representation sandwiched between an optional ASCII record separator (RS, ``\x1e``) and a newline (``\n``). .. code-block:: python >>> from jsonseq.encode import JSONSeqEncoder >>> for chunk in JSONSeqEncoder().encode(({"a": i, "b": i} for i in range(3))): ... print(repr(chunk)) ... '{"a": 0, "b": 0}\n' '{"a": 1, "b": 1}\n' '{"a": 2, "b": 2}\n' The RS allows pretty-printed JSON to be streamed out in sequences that can be decoded again. .. code-block:: python >>> for chunk in JSONSeqEncoder(with_rs=True, indent=2).encode(({"a": i, "b": i} for i in range(3))): ... print(repr(chunk)) ... '\x1e{\n "a": 0,\n "b": 0\n}\n' '\x1e{\n "a": 1,\n "b": 1\n}\n' '\x1e{\n "a": 2,\n "b": 2\n}\n' You can also get small chunks of the JSON sequences as they are encoded with JSONSeqEncoder's iterencode method. .. code-block:: python >>> for chunk in JSONSeqEncoder(with_rs=True).iterencode(({"a": i} for i in range(3))): ... print(repr(chunk)) ... '\x1e' '{' '"a"' ': ' '0' '}' '\n' '\x1e' '{' '"a"' ': ' '1' '}' '\n' '\x1e' '{' '"a"' ': ' '2' '}' '\n' You can use either encode or iterencode to copy JSON text sequences to a file. .. code-block:: python with open("/tmp/example.jsons", "w") as f: for chunk in JSONSeqEncoder(with_rs=True, indent=2).iterencode(({"a": i, "b": i} for i in range(3))): f.write(chunk) There is no need to add a newline when calling the file's write method. JSONSeqEncoder ensures that it's already there where it needs to be. The JSONSeqDecoder class takes streams of JSON texts sandwiched between the optional RS and a newline and yields decoded Python objects. .. code-block:: python >>> stream = ['\x1e', '{', '"a"', ': ', '0', '}', '\n', '\x1e', '{', '"a"', ': ', '1', '}', '\n', '\x1e', '{', '"a"', ': ', '2', '}', '\n'] >>> for obj in JSONSeqDecoder().decode(stream): ... print(repr(obj)) ... {'a': 0} {'a': 1} {'a': 2} Objects can be read from a file in the same way. .. code-block:: python >>> with open("/tmp/example.jsons") as f: ... for obj in JSONSeqDecoder().decode(f): ... print(repr(obj)) ... {'a': 0, 'b': 0} {'a': 1, 'b': 1} {'a': 2, 'b': 2} API Documentation ================= jsonseq package --------------- .. automodule:: jsonseq jsonseq.decode module --------------------- .. automodule:: jsonseq.decode :members: :undoc-members: :show-inheritance: :member-order: groupwise jsonseq.encode module --------------------- .. automodule:: jsonseq.encode :members: :undoc-members: :show-inheritance: :member-order: groupwise Project Home ============ `https://github.com/sgillies/jsonseq `__ See Also ======== - `RFC 7464 JSON Text Sequences `__ Indices and tables ================== * :ref:`genindex` * :ref:`modindex` * :ref:`search`