_codecs_json

data.drivers._pelt._codecs_json

module

E2 codec family:json / jsonb.

PostgreSQL's two document types decode to a parsed Python object (whatever loads() yields — dict / list / str / int / float/ …

E2 codec family:json / jsonb.

PostgreSQL's two document types decode to a parsed Python object (whatever loads() yields — dict / list / str / int / float / bool/ None) and encode from one via dumps(). The two differ only in their binary wire framing:

json(OID 114) stores the document verbatim, so its binary format is identical to its text format: UTF-8 JSON bytes, no envelope.
jsonb (OID 3802) prefixes the UTF-8 JSON with a single version byte. Version 1 is the only one PostgreSQL has ever emitted (jsonb_send writes 0x01then the decompressed text;jsonb_recvreads the version then the text). A byte other than 0x01means the stream desynced or the server speaks a future format pelt does not know — that is aProtocolError, never a silent best-effort decode.

These are leaf (non-parametric) codecs — the registry registers them directly from LEAF_CODECS. Composite/array families that contain json land separately in E2. Live-PG parity is deferred to E4/E6 integration; the binary vectors here are pinned to the documentedjsonbwire layout.

data.drivers._pelt._codecs_json

Name	Type	Default	Description
`type`		—
`qualified_name`		—
`element_type`		—
`description`		—
`source_file`		—
`line_number`		—
`is_autodoc`		—
`autodoc_element`		—
`_autodoc_template`		—
`_autodoc_url_path`		—
`_autodoc_page_type`		—
`title`		—
`doc_content_hash`		—

Symbols on this page

function _dumps

Serialize a Python object to compact UTF-8 JSON bytes.

Compact == no whitespace between tokens (separators=(",", ":")). PostgreSQL parses json/jsonb regardless of…

function _loads

Parse UTF-8 JSON bytes to a Python object, surfacing faults asProtocolError.

function _json_codec

json(OID 114): binary == text == UTF-8 JSON, no envelope.

function _jsonb_codec

jsonb (OID 3802): binary is 0x01+ UTF-8 JSON; text is bare UTF-8 JSON.

_dumps

function

def _dumps(value: Any) -> bytes

Serialize a Python object to compact UTF-8 JSON bytes.

Compact == no whitespace between tokens (separators=(",", ":")). PostgreSQL parses json/jsonb regardless of internal whitespace, so dropping the spaces is wire-safe and shrinks the payload.

Parameters

Name	Type	Default	Description
`value`	`Any`	—

_loads

function

def _loads(data: bytes) -> Any

Parse UTF-8 JSON bytes to a Python object, surfacing faults asProtocolError.

Parameters

Name	Type	Default	Description
`data`	`bytes`	—

_json_codec

function

def _json_codec() -> Codec

json(OID 114): binary == text == UTF-8 JSON, no envelope.

No parameters.

_jsonb_codec

function

def _jsonb_codec() -> Codec

jsonb (OID 3802): binary is 0x01+ UTF-8 JSON; text is bare UTF-8 JSON.

No parameters.

View source · /home/runner/work/chirp/chirp/site/../src/chirp/data/drivers/_pelt/_codecs_json.py:1