`flow.record.adapter.avro`¶

Module Contents¶

Classes¶

`AvroWriter`
`AvroReader`

Functions¶

`descriptor_to_schema`
`schema_to_descriptor`
`avro_type_to_flow_type`

Attributes¶

`__usage__`
`AVRO_TYPE_MAP`
`RECORD_TYPE_MAP`
`EPOCH`

flow.record.adapter.avro.__usage__ = Multiline-String¶

Show Value

"""
Apache AVRO adapter
---
Write usage: rdump -w avro://[PATH]
Read usage: rdump avro://[PATH]
[PATH]: path to file. Leave empty or "-" to output to stdout
"""

flow.record.adapter.avro.AVRO_TYPE_MAP¶

flow.record.adapter.avro.RECORD_TYPE_MAP¶

flow.record.adapter.avro.EPOCH¶

class flow.record.adapter.avro.AvroWriter(path: str | pathlib.Path | BinaryIO, **kwargs)¶

Bases: flow.record.adapter.AbstractWriter

fp = None¶

writer = None¶

desc = None¶

schema = None¶

parsed_schema = None¶

codec = 'snappy'¶

write(r: flow.record.Record) → None¶: Write a record.

flush() → None¶: Flush any buffered writes.

close() → None¶: Close the Writer, no more writes will be possible.

class flow.record.adapter.avro.AvroReader(path: str, selector: str | None = None, **kwargs)¶

Bases: flow.record.adapter.AbstractReader

fp = None¶

selector = None¶

reader¶

schema¶

desc¶

datetime_fields¶

__iter__() → collections.abc.Iterator[flow.record.Record]¶: Return a record iterator.

close() → None¶: Close the Reader, can be overriden to properly free resources.

flow.record.adapter.avro.descriptor_to_schema(desc: flow.record.RecordDescriptor) → dict[str, Any]¶

flow.record.adapter.avro.schema_to_descriptor(schema: dict) → flow.record.RecordDescriptor¶

flow.record.adapter.avro.avro_type_to_flow_type(ftype: list) → str¶