MemoryKeyValueStoreClient
Hierarchy
- KeyValueStoreClient
- MemoryKeyValueStoreClient
Index
Methods
__init__
Initialize a new instance.
Preferably use the
MemoryKeyValueStoreClient.openclass method to create a new instance.Parameters
keyword-onlymetadata: KeyValueStoreMetadata
Returns None
delete_value
Delete a value from the key-value store by its key.
The backend method for the
KeyValueStore.delete_valuecall.Parameters
keyword-onlykey: str
Returns None
drop
Drop the whole key-value store and remove all its values.
The backend method for the
KeyValueStore.dropcall.Returns None
get_metadata
Get the metadata of the key-value store.
Returns KeyValueStoreMetadata
get_public_url
Get the public URL for the given key.
The backend method for the
KeyValueStore.get_public_urlcall.Parameters
keyword-onlykey: str
Returns str
get_value
Retrieve the given record from the key-value store.
The backend method for the
KeyValueStore.get_valuecall.Parameters
keyword-onlykey: str
Returns KeyValueStoreRecord | None
iterate_entries
Iterate over all the existing records in the key-value store, including their values.
The backend method for the
KeyValueStore.iterate_entriesandKeyValueStore.iterate_valuescalls.The default implementation lists the keys with
iterate_keysand reads each value separately withget_valueas the iteration advances, so only a single value is held in memory at a time. Whether a record deleted while the iteration is in progress is yielded depends on the implementation. Backends that can read the keys together with their values more efficiently should override this method; such an implementation must still keep the memory usage bounded, e.g. by streaming or by batching on the known record sizes.Parameters
optionalkeyword-onlyexclusive_start_key: str | None = None
Memory implementation of the key-value store client.
This client stores data in memory as Python dictionaries. No data is persisted between process runs, meaning all stored data is lost when the program terminates. This implementation is primarily useful for testing, development, and short-lived crawler operations where persistence is not required.
The memory implementation provides fast access to data but is limited by available memory and does not support data sharing across different processes.