Class PostgresDocumentStore (0.2.1)

PostgresDocumentStore(
    key: object,
    engine: llama_index_cloud_sql_pg.engine.PostgresEngine,
    document_store: llama_index_cloud_sql_pg.async_document_store.AsyncPostgresDocumentStore,
)

Document Store Table stored in an Cloud SQL for PostgreSQL database.

Properties

docs

Get all documents.

Returns
Type	Description
`dict[str, BaseDocument]`	documents

Methods

PostgresDocumentStore

PostgresDocumentStore(
    key: object,
    engine: llama_index_cloud_sql_pg.engine.PostgresEngine,
    document_store: llama_index_cloud_sql_pg.async_document_store.AsyncPostgresDocumentStore,
)

"PostgresDocumentStore constructor.

Parameters
Name	Description
`key`	`object` Key to prevent direct constructor usage.
`engine`	`PostgresEngine` Database connection pool.
`document_store`	`AsyncPostgresDocumentStore` The async only DocumentStore implementation

Exceptions
Type	Description
`Exception`	If constructor is directly called by the user.

add_documents

add_documents(
    docs: typing.Sequence[llama_index.core.schema.BaseNode],
    allow_update: bool = True,
    batch_size: int = 1,
    store_text: bool = True,
) -> None

Adds a document to the store.

Parameters
Name	Description
`docs`	`Sequence[BaseDocument]` documents
`allow_update`	`bool` allow update of docstore from document
`batch_size`	`int` batch_size to insert the rows. Defaults to 1.
`store_text`	`bool` allow the text content of the node to stored. Defaults to "True".

adelete_document

adelete_document(doc_id: str, raise_error: bool = True) -> None

Delete a document from the store.

Parameters
Name	Description
`doc_id`	`str` Id of the document / node to be deleted.
`raise_error`	`bool` to raise error if document is not found.

Exceptions
Type	Description
`ValueError`	If a node is not found and `raise_error` is set to True.

adelete_ref_doc

adelete_ref_doc(ref_doc_id: str, raise_error: bool = True) -> None

Delete a ref_doc and all it's associated nodes.

Parameters
Name	Description
`ref_doc_id`	`str` Ref_doc_id which needs to be deleted.
`raise_error`	`bool` to raise error if ref_doc_info for the ref_doc_id is not found.

Exceptions
Type	Description
`ValueError`	If ref_doc_info for the ref_doc_id doesn't exist and `raise_error` is set to True.

adocument_exists

adocument_exists(doc_id: str) -> bool

Check if document exists.

Parameter
Name	Description
`doc_id`	`str` The document / node id which needs to be found.

Returns
Type	Description
`bool`	True if document exists in the table.

aget_all_document_hashes

aget_all_document_hashes() -> dict[str, str]

Get the stored hash for all documents.

Returns: A dictionary mapping document hashes to document IDs.

aget_all_ref_doc_info

aget_all_ref_doc_info() -> (
    typing.Optional[dict[str, llama_index.core.storage.docstore.types.RefDocInfo]]
)

Get a mapping of ref_doc_id -> RefDocInfo for all ingested documents.

Returns: A dictionary mapping ref_doc_ids to RefDocInfo objects, or None if no documents have been ingested.

aget_document

aget_document(
    doc_id: str, raise_error: bool = True
) -> typing.Optional[llama_index.core.schema.BaseNode]

Retrieves a document from the table by its doc_id.

Parameters
Name	Description
`doc_id`	`str` Id of the document / node to be retrieved.
`raise_error`	`bool` to raise error if document is not found.

Exceptions
Type	Description
`ValueError`	If a node doesn't exist and `raise_error` is set to True.

Returns
Type	Description
`Optional[BaseNode]`	Returns a `BaseNode` object if the document is found

aget_document_hash

aget_document_hash(doc_id: str) -> typing.Optional[str]

Get the stored hash for a document, if it exists.

Returns
Type	Description
`Optional[str]`	The hash for the given doc_id, if available.

aget_ref_doc_info

aget_ref_doc_info(
    ref_doc_id: str,
) -> typing.Optional[llama_index.core.storage.docstore.types.RefDocInfo]

Get the RefDocInfo for a given ref_doc_id.

Parameter
Name	Description
`ref_doc_id`	`str` The ref_doc_id whose ref_doc_info is to be retrieved.

Returns
Type	Description
`Optional[RefDocInfo]`	Returns a `RefDocInfo` object if it exists.

aref_doc_exists

aref_doc_exists(ref_doc_id: str) -> bool

Check if a ref_doc_id has been ingested.

Parameter
Name	Description
`ref_doc_id`	`str` The ref_doc_id whose ref_doc_info is to be found.

Returns
Type	Description
`bool`	True if document exists as a ref doc in the table.

aset_document_hash

aset_document_hash(doc_id: str, doc_hash: str) -> None

Set the hash for a given doc_id.

Parameters
Name	Description
`doc_id`	`str` Id to be updated with the doc_hash.
`doc_hash`	`str` Doc_hash to be updated into the table.

aset_document_hashes

aset_document_hashes(doc_hashes: dict[str, str]) -> None

Set the hash for a given doc_id.

Parameter
Name	Description
`doc_hashes`	`dict[str, str]` Dictionary with doc_id as key and doc_hash as value.

async_add_documents

async_add_documents(
    docs: typing.Sequence[llama_index.core.schema.BaseNode],
    allow_update: bool = True,
    batch_size: int = 1,
    store_text: bool = True,
) -> None

Adds a document to the store.

Parameters
Name	Description
`docs`	`Sequence[BaseDocument]` documents
`allow_update`	`bool` allow update of docstore from document
`batch_size`	`int` batch_size to insert the rows. Defaults to 1.
`store_text`	`bool` allow the text content of the node to stored. Defaults to "True".

create

create(
    engine: llama_index_cloud_sql_pg.engine.PostgresEngine,
    table_name: str,
    schema_name: str = "public",
    batch_size: int = 1,
) -> llama_index_cloud_sql_pg.document_store.PostgresDocumentStore

Create a new PostgresDocumentStore instance.

Parameters
Name	Description
`engine`	`PostgresEngine` Postgres engine to use.
`table_name`	`str` Table name that stores the documents.
`schema_name`	`str` The schema name where the table is located. Defaults to "public"
`batch_size`	`str` The default batch size for bulk inserts. Defaults to 1.

Exceptions
Type	Description
`ValueError`	If the table provided does not contain required schema.

Returns
Type	Description
`PostgresDocumentStore`	A newly created instance of PostgresDocumentStore.

create_sync

create_sync(
    engine: llama_index_cloud_sql_pg.engine.PostgresEngine,
    table_name: str,
    schema_name: str = "public",
    batch_size: int = 1,
) -> llama_index_cloud_sql_pg.document_store.PostgresDocumentStore

Create a new PostgresDocumentStore sync instance.

Parameters
Name	Description
`engine`	`PostgresEngine` Postgres engine to use.
`table_name`	`str` Table name that stores the documents.
`schema_name`	`str` The schema name where the table is located. Defaults to "public"
`batch_size`	`str` The default batch size for bulk inserts. Defaults to 1.

Exceptions
Type	Description
`ValueError`	If the table provided does not contain required schema.

Returns
Type	Description
`PostgresDocumentStore`	A newly created instance of PostgresDocumentStore.

delete_document

delete_document(doc_id: str, raise_error: bool = True) -> None

Delete a document from the store.

Parameters
Name	Description
`doc_id`	`str` Id of the document / node to be deleted.
`raise_error`	`bool` to raise error if document is not found.

Exceptions
Type	Description
`ValueError`	If a node is not found and `raise_error` is set to True.

delete_ref_doc

delete_ref_doc(ref_doc_id: str, raise_error: bool = True) -> None

Delete a ref_doc and all it's associated nodes.

Parameters
Name	Description
`ref_doc_id`	`str` Ref_doc_id which needs to be deleted.
`raise_error`	`bool` to raise error if ref_doc_info for the ref_doc_id is not found.

Exceptions
Type	Description
`ValueError`	If ref_doc_info for the ref_doc_id doesn't exist and `raise_error` is set to True.

document_exists

document_exists(doc_id: str) -> bool

Check if document exists.

Parameter
Name	Description
`doc_id`	`str` The document / node id which needs to be found.

Returns
Type	Description
`bool`	True if document exists in the table.

get_all_document_hashes

get_all_document_hashes() -> dict[str, str]

Get the stored hash for all documents.

Returns: A dictionary mapping document hashes to document IDs.

get_all_ref_doc_info

get_all_ref_doc_info() -> (
    typing.Optional[dict[str, llama_index.core.storage.docstore.types.RefDocInfo]]
)

Get a mapping of ref_doc_id -> RefDocInfo for all ingested documents.

Returns: A dictionary mapping ref_doc_ids to RefDocInfo objects, or None if no documents have been ingested.

get_document

get_document(
    doc_id: str, raise_error: bool = True
) -> typing.Optional[llama_index.core.schema.BaseNode]

Retrieves a document from the table by its doc_id.

Parameters
Name	Description
`doc_id`	`str` Id of the document / node to be retrieved.
`raise_error`	`bool` to raise error if document is not found.

Exceptions
Type	Description
`ValueError`	If a node doesn't exist and `raise_error` is set to True.

Returns
Type	Description
`Optional[BaseNode]`	Returns a `BaseNode` object if the document is found

get_document_hash

get_document_hash(doc_id: str) -> typing.Optional[str]

Get the stored hash for a document, if it exists.

Returns
Type	Description
`Optional[str]`	The hash for the given doc_id, if available.

get_ref_doc_info

get_ref_doc_info(
    ref_doc_id: str,
) -> typing.Optional[llama_index.core.storage.docstore.types.RefDocInfo]

Get the RefDocInfo for a given ref_doc_id.

Parameter
Name	Description
`ref_doc_id`	`str` The ref_doc_id whose ref_doc_info is to be retrieved.

Returns
Type	Description
`Optional[RefDocInfo]`	Returns a `RefDocInfo` object if it exists.

ref_doc_exists

ref_doc_exists(ref_doc_id: str) -> bool

Check if a ref_doc_id has been ingested.

Parameter
Name	Description
`ref_doc_id`	`str` The ref_doc_id whose ref_doc_info is to be found.

Returns
Type	Description
`bool`	True if document exists as a ref doc in the table.

set_document_hash

set_document_hash(doc_id: str, doc_hash: str) -> None

Set the hash for a given doc_id.

Parameters
Name	Description
`doc_id`	`str` Id to be updated with the doc_hash.
`doc_hash`	`str` Doc_hash to be updated into the table.

set_document_hashes

set_document_hashes(doc_hashes: dict[str, str]) -> None

Set the hash for a given doc_id.

Parameter
Name	Description
`doc_hashes`	`dict[str, str]` Dictionary with doc_id as key and doc_hash as value.