Skip to content

Document Stores¤

CouchbaseDocumentStore¤

Base class for Couchbase document stores that provides common functionality for managing connections, scopes, collections, and basic document operations.

Source code in src/couchbase_haystack/document_stores/document_store.py 
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
class CouchbaseDocumentStore:
    """Base class for Couchbase document stores that provides common functionality
    for managing connections, scopes, collections, and basic document operations.
    """

    def __init__(
        self,
        *,
        cluster_connection_string: Secret = Secret.from_env_var("CB_CONNECTION_STRING"),
        authenticator: Union[CouchbasePasswordAuthenticator, CouchbaseCertificateAuthenticator],
        cluster_options: CouchbaseClusterOptions = CouchbaseClusterOptions(),
        bucket: str,
        scope: str,
        collection: str,
        **kwargs: Dict[str, Any],
    ):
        """Creates a new CouchbaseDocumentStore instance.

        Args:
            cluster_connection_string: Connection string for the Couchbase cluster
            authenticator: Authentication method (password or certificate based)
            cluster_options: Options for configuring the cluster connection
            bucket: Name of the Couchbase bucket to use
            scope: Name of the scope within the bucket
            collection: Name of the collection within the scope
            kwargs: Additional keyword arguments passed to the Cluster constructor

        Raises:
            ValueError: If the provided collection name contains invalid characters.
        """
        if collection and not bool(re.match(r"^[a-zA-Z0-9\-_]+$", collection)):
            msg = f'Invalid collection name: "{collection}". It can only contain letters, numbers, -, or _.'
            raise ValueError(msg)

        self.cluster_connection_string = cluster_connection_string
        self.authenticator = authenticator
        self.cluster_options = cluster_options
        self.bucket_name = bucket
        self.scope_name = scope
        self.collection_name = collection
        self._connection: Optional[Cluster] = None
        self._bucket: Optional[Bucket] = None
        self._scope: Optional[Scope] = None
        self._collection: Optional[Collection] = None
        self._kwargs = kwargs

    @property
    def connection(self) -> Cluster:
        """Establishes and returns the Couchbase Cluster connection.

        Initializes the connection if it doesn't exist, applying cluster options and authentication.
        Waits until the cluster is ready before returning.

        Returns:
            The active `couchbase.cluster.Cluster` instance.

        Raises:
            DocumentStoreError: If the connection cannot be established or times out.
        """
        if self._connection is None:
            try:
                cluster_options = self.cluster_options.get_cluster_options(self.authenticator.get_cb_auth())
                if self.cluster_options.get("profile") is not None:
                    cluster_options.apply_profile(self.cluster_options["profile"])
                self._connection = Cluster(
                    self.cluster_connection_string.resolve_value(),
                    cluster_options,
                    **self._kwargs,
                )
                self._connection.wait_until_ready(timeout=timedelta(seconds=60))
            except TimeoutError as e:
                msg = f"Failed to establish connection: {e}"
                raise DocumentStoreError(msg) from e
            except Exception as e:
                msg = f"Failed to establish connection: {e}"
                raise DocumentStoreError(msg) from e
        return self._connection

    @property
    def bucket(self) -> Bucket:
        """Returns the Couchbase `Bucket` object associated with this document store.

        Returns:
            The `couchbase.bucket.Bucket` instance.

        Raises:
            Exceptions from the underlying `connection.bucket()` call if the bucket doesn't exist or is inaccessible.
        """
        if self._bucket is None:
            self._bucket = self.connection.bucket(self.bucket_name)
        return self._bucket

    @property
    def scope(self) -> Scope:
        """Returns the Couchbase `Scope` object associated with this document store.

        Returns:
            The `couchbase.scope.Scope` instance.

        Raises:
            ValueError: If the specified scope or collection does not exist in the bucket.
        """
        if self._scope is None:
            scopes_specs = self.bucket.collections().get_all_scopes()
            scope_found = False
            collection_found = False
            for scope_spec in scopes_specs:
                if scope_spec.name == self.scope_name:
                    scope_found = True
                    for col_spec in scope_spec.collections:
                        if col_spec.name == self.collection_name:
                            collection_found = True
            if not scope_found:
                msg = f"Scope '{self.scope_name}' does not exist in bucket '{self.bucket_name}'."
                raise ValueError(msg)
            if not collection_found:
                msg = f"Collection '{self.collection_name}' does not exist in scope '{self.scope_name}'."
                raise ValueError(msg)
            self._scope = self.bucket.scope(self.scope_name)
        return self._scope

    @property
    def collection(self) -> Collection:
        """Returns the Couchbase `Collection` object associated with this document store.

        Returns:
            The `couchbase.collection.Collection` instance.

        Raises:
            Exceptions from the underlying `scope.collection()` call if the collection is inaccessible.
        """
        if self._collection is None:
            self._collection = self.scope.collection(self.collection_name)
        return self._collection

    def _base_to_dict(self) -> Dict[str, Any]:
        """Creates a base dictionary containing common configuration parameters for serialization.

        This is intended to be used by subclasses in their `to_dict` methods.

        Returns:
            A dictionary with core configuration details.
        """
        return {
            "cluster_connection_string": self.cluster_connection_string.to_dict(),
            "authenticator": self.authenticator.to_dict(),
            "cluster_options": self.cluster_options.to_dict(),
            "bucket": self.bucket_name,
            "scope": self.scope_name,
            "collection": self.collection_name,
            **self._kwargs,
        }

    def write_documents(self, documents: List[Document], policy: DuplicatePolicy = DuplicatePolicy.NONE) -> int:
        """Writes documents into the couchbase collection.

        Args:
            documents: A list of Documents to write to the document store.
            policy: The duplicate policy to use when writing documents.
                    `FAIL`: (Default if `NONE`) Raise an error if a document ID already exists.
                    `OVERWRITE`: Replace existing documents with the same ID.

        Raises:
            DuplicateDocumentError: If `policy` is `FAIL` and a document with the same ID already exists.
            ValueError: If `documents` is not a list of `Document` objects.
            DocumentStoreError: If any other error occurs during the write operation.

        Returns:
            The number of documents successfully written to the document store.
        """
        if len(documents) == 0:
            return 0

        if not isinstance(documents[0], Document):
            msg = "param 'documents' must contain a list of objects of type Document"
            raise ValueError(msg)

        if policy == DuplicatePolicy.NONE:
            policy = DuplicatePolicy.FAIL

        cb_documents = []
        for doc in documents:
            doc_dict = doc.to_dict(flatten=False)
            doc_dict = {k: v for k, v in doc_dict.items() if v is not None}
            if "sparse_embedding" in doc_dict:
                sparse_embedding = doc_dict.pop("sparse_embedding", None)
                if sparse_embedding:
                    logger.warning(
                        "Document %s has the `sparse_embedding` field set,"
                        "but storing sparse embeddings in Couchbase is not currently supported."
                        "The `sparse_embedding` field will be ignored.",
                        doc.id,
                    )
            cb_documents.append(doc_dict)
        written_docs = len(documents)

        operations = {doc["id"]: doc for doc in cb_documents}
        try:
            result: MultiMutationResult
            if policy == DuplicatePolicy.FAIL:
                result = self.collection.insert_multi(operations)
            else:
                result = self.collection.upsert_multi(operations)
        except Exception as e:
            logger.error(f"write error {e}")
            msg = f"Failed to write documents to Couchbase. Error: {e}"
            raise DocumentStoreError(msg) from e
        if not result.all_ok and result.exceptions:
            duplicate_ids = []
            other_errors = []
            for id, ex in result.exceptions.items():
                if isinstance(ex, DocumentExistsException):
                    duplicate_ids.append(id)
                else:
                    other_errors.append({"id": id, "exception": ex})
            if len(duplicate_ids) > 0:
                msg = f"IDs '{', '.join(duplicate_ids)}' already exist in the document store."
                raise DuplicateDocumentError(msg)
            if len(other_errors) > 0:
                msg = f"Failed to write documents to couchbase. Errors:\n{other_errors}"
                raise DocumentStoreError(msg)
        logger.debug("data written")
        return written_docs

    def delete_documents(self, document_ids: List[str]) -> None:
        """Deletes all documents with a matching document_ids from the document store.

        Args:
            document_ids: the document ids to delete
        """
        if not document_ids:
            return
        self.collection.remove_multi(keys=document_ids)

connection property ¤

connection: Cluster

Establishes and returns the Couchbase Cluster connection.

Initializes the connection if it doesn't exist, applying cluster options and authentication. Waits until the cluster is ready before returning.

Returns:

  • Cluster –

    The active couchbase.cluster.Cluster instance.

Raises:

  • DocumentStoreError –

    If the connection cannot be established or times out.

bucket property ¤

bucket: Bucket

Returns the Couchbase Bucket object associated with this document store.

Returns:

  • Bucket –

    The couchbase.bucket.Bucket instance.

scope property ¤

scope: Scope

Returns the Couchbase Scope object associated with this document store.

Returns:

  • Scope –

    The couchbase.scope.Scope instance.

Raises:

  • ValueError –

    If the specified scope or collection does not exist in the bucket.

collection property ¤

collection: Collection

Returns the Couchbase Collection object associated with this document store.

Returns:

  • Collection –

    The couchbase.collection.Collection instance.

__init__ ¤

__init__(
    *,
    cluster_connection_string: Secret = Secret.from_env_var(
        "CB_CONNECTION_STRING"
    ),
    authenticator: Union[
        CouchbasePasswordAuthenticator, CouchbaseCertificateAuthenticator
    ],
    cluster_options: CouchbaseClusterOptions = CouchbaseClusterOptions(),
    bucket: str,
    scope: str,
    collection: str,
    **kwargs: Dict[str, Any]
)

Parameters:

  • cluster_connection_string (Secret, default: from_env_var('CB_CONNECTION_STRING') ) –

    Connection string for the Couchbase cluster

  • authenticator (Union[CouchbasePasswordAuthenticator, CouchbaseCertificateAuthenticator]) –

    Authentication method (password or certificate based)

  • cluster_options (CouchbaseClusterOptions, default: CouchbaseClusterOptions() ) –

    Options for configuring the cluster connection

  • bucket (str) –

    Name of the Couchbase bucket to use

  • scope (str) –

    Name of the scope within the bucket

  • collection (str) –

    Name of the collection within the scope

  • kwargs (Dict[str, Any], default: {} ) –

    Additional keyword arguments passed to the Cluster constructor

Raises:

  • ValueError –

    If the provided collection name contains invalid characters.

Source code in src/couchbase_haystack/document_stores/document_store.py 
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
def __init__(
    self,
    *,
    cluster_connection_string: Secret = Secret.from_env_var("CB_CONNECTION_STRING"),
    authenticator: Union[CouchbasePasswordAuthenticator, CouchbaseCertificateAuthenticator],
    cluster_options: CouchbaseClusterOptions = CouchbaseClusterOptions(),
    bucket: str,
    scope: str,
    collection: str,
    **kwargs: Dict[str, Any],
):
    """Creates a new CouchbaseDocumentStore instance.

    Args:
        cluster_connection_string: Connection string for the Couchbase cluster
        authenticator: Authentication method (password or certificate based)
        cluster_options: Options for configuring the cluster connection
        bucket: Name of the Couchbase bucket to use
        scope: Name of the scope within the bucket
        collection: Name of the collection within the scope
        kwargs: Additional keyword arguments passed to the Cluster constructor

    Raises:
        ValueError: If the provided collection name contains invalid characters.
    """
    if collection and not bool(re.match(r"^[a-zA-Z0-9\-_]+$", collection)):
        msg = f'Invalid collection name: "{collection}". It can only contain letters, numbers, -, or _.'
        raise ValueError(msg)

    self.cluster_connection_string = cluster_connection_string
    self.authenticator = authenticator
    self.cluster_options = cluster_options
    self.bucket_name = bucket
    self.scope_name = scope
    self.collection_name = collection
    self._connection: Optional[Cluster] = None
    self._bucket: Optional[Bucket] = None
    self._scope: Optional[Scope] = None
    self._collection: Optional[Collection] = None
    self._kwargs = kwargs

_base_to_dict ¤

_base_to_dict() -> Dict[str, Any]

Creates a base dictionary containing common configuration parameters for serialization.

This is intended to be used by subclasses in their to_dict methods.

Returns:

  • Dict[str, Any] –

    A dictionary with core configuration details.

Source code in src/couchbase_haystack/document_stores/document_store.py 
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
def _base_to_dict(self) -> Dict[str, Any]:
    """Creates a base dictionary containing common configuration parameters for serialization.

    This is intended to be used by subclasses in their `to_dict` methods.

    Returns:
        A dictionary with core configuration details.
    """
    return {
        "cluster_connection_string": self.cluster_connection_string.to_dict(),
        "authenticator": self.authenticator.to_dict(),
        "cluster_options": self.cluster_options.to_dict(),
        "bucket": self.bucket_name,
        "scope": self.scope_name,
        "collection": self.collection_name,
        **self._kwargs,
    }

write_documents ¤

write_documents(
    documents: List[Document], policy: DuplicatePolicy = DuplicatePolicy.NONE
) -> int

Writes documents into the couchbase collection.

Parameters:

  • documents (List[Document]) –

    A list of Documents to write to the document store.

  • policy (DuplicatePolicy, default: NONE ) –

    The duplicate policy to use when writing documents. FAIL: (Default if NONE) Raise an error if a document ID already exists. OVERWRITE: Replace existing documents with the same ID.

Raises:

  • DuplicateDocumentError –

    If policy is FAIL and a document with the same ID already exists.

  • ValueError –

    If documents is not a list of Document objects.

  • DocumentStoreError –

    If any other error occurs during the write operation.

Returns:

  • int –

    The number of documents successfully written to the document store.

Source code in src/couchbase_haystack/document_stores/document_store.py 
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
def write_documents(self, documents: List[Document], policy: DuplicatePolicy = DuplicatePolicy.NONE) -> int:
    """Writes documents into the couchbase collection.

    Args:
        documents: A list of Documents to write to the document store.
        policy: The duplicate policy to use when writing documents.
                `FAIL`: (Default if `NONE`) Raise an error if a document ID already exists.
                `OVERWRITE`: Replace existing documents with the same ID.

    Raises:
        DuplicateDocumentError: If `policy` is `FAIL` and a document with the same ID already exists.
        ValueError: If `documents` is not a list of `Document` objects.
        DocumentStoreError: If any other error occurs during the write operation.

    Returns:
        The number of documents successfully written to the document store.
    """
    if len(documents) == 0:
        return 0

    if not isinstance(documents[0], Document):
        msg = "param 'documents' must contain a list of objects of type Document"
        raise ValueError(msg)

    if policy == DuplicatePolicy.NONE:
        policy = DuplicatePolicy.FAIL

    cb_documents = []
    for doc in documents:
        doc_dict = doc.to_dict(flatten=False)
        doc_dict = {k: v for k, v in doc_dict.items() if v is not None}
        if "sparse_embedding" in doc_dict:
            sparse_embedding = doc_dict.pop("sparse_embedding", None)
            if sparse_embedding:
                logger.warning(
                    "Document %s has the `sparse_embedding` field set,"
                    "but storing sparse embeddings in Couchbase is not currently supported."
                    "The `sparse_embedding` field will be ignored.",
                    doc.id,
                )
        cb_documents.append(doc_dict)
    written_docs = len(documents)

    operations = {doc["id"]: doc for doc in cb_documents}
    try:
        result: MultiMutationResult
        if policy == DuplicatePolicy.FAIL:
            result = self.collection.insert_multi(operations)
        else:
            result = self.collection.upsert_multi(operations)
    except Exception as e:
        logger.error(f"write error {e}")
        msg = f"Failed to write documents to Couchbase. Error: {e}"
        raise DocumentStoreError(msg) from e
    if not result.all_ok and result.exceptions:
        duplicate_ids = []
        other_errors = []
        for id, ex in result.exceptions.items():
            if isinstance(ex, DocumentExistsException):
                duplicate_ids.append(id)
            else:
                other_errors.append({"id": id, "exception": ex})
        if len(duplicate_ids) > 0:
            msg = f"IDs '{', '.join(duplicate_ids)}' already exist in the document store."
            raise DuplicateDocumentError(msg)
        if len(other_errors) > 0:
            msg = f"Failed to write documents to couchbase. Errors:\n{other_errors}"
            raise DocumentStoreError(msg)
    logger.debug("data written")
    return written_docs

delete_documents ¤

delete_documents(document_ids: List[str]) -> None

Deletes all documents with a matching document_ids from the document store.

Parameters:

  • document_ids (List[str]) –

    the document ids to delete

Source code in src/couchbase_haystack/document_stores/document_store.py 
341
342
343
344
345
346
347
348
349
def delete_documents(self, document_ids: List[str]) -> None:
    """Deletes all documents with a matching document_ids from the document store.

    Args:
        document_ids: the document ids to delete
    """
    if not document_ids:
        return
    self.collection.remove_multi(keys=document_ids)

CouchbaseSearchDocumentStore¤

Bases: CouchbaseDocumentStore

CouchbaseSearchDocumentStore is a DocumentStore implementation that uses Couchbase capella service that is easy to deploy, operate, and scale.

The document store supports both scope-level and global-level vector search indexes:

  • Scope-level indexes (default): The vector search index is created at the scope level and only searches documents within that scope
  • Global-level indexes: The vector search index is created at the bucket level and can search across all scopes and collections in the bucket

The index level is specified using the is_global_level_index parameter during initialization.

Source code in src/couchbase_haystack/document_stores/document_store.py 
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
class CouchbaseSearchDocumentStore(CouchbaseDocumentStore):
    """CouchbaseSearchDocumentStore is a DocumentStore implementation that uses
    [Couchbase capella](https://cloud.couchbase.com) service that is easy to deploy, operate, and scale.

    The document store supports both scope-level and global-level vector search indexes:

    - Scope-level indexes (default): The vector search index is created at the scope level and only searches
      documents within that scope
    - Global-level indexes: The vector search index is created at the bucket level and can search across all
      scopes and collections in the bucket

    The index level is specified using the `is_global_level_index` parameter during initialization.
    """

    def __init__(
        self,
        *,
        cluster_connection_string: Secret = Secret.from_env_var("CB_CONNECTION_STRING"),
        authenticator: Union[CouchbasePasswordAuthenticator, CouchbaseCertificateAuthenticator],
        cluster_options: CouchbaseClusterOptions = CouchbaseClusterOptions(),
        bucket: str,
        scope: str,
        collection: str,
        vector_search_index: str,
        is_global_level_index: bool = False,
        **kwargs: Dict[str, Any],
    ):
        """Creates a new CouchbaseSearchDocumentStore instance.

        Args:
            cluster_connection_string: Connection string for the Couchbase cluster
            authenticator: Authentication method (password or certificate based)
            cluster_options: Options for configuring the cluster connection
            bucket: Name of the Couchbase bucket to use
            scope: Name of the scope within the bucket
            collection: Name of the collection within the scope
            vector_search_index: Name of the FTS index (which must include vector indexing) to use for searches.
            is_global_level_index: If `True`, use a global (bucket-level) FTS index.
                                  If `False` (default), use a scope-level FTS index.
            kwargs: Additional keyword arguments passed to the Cluster constructor.
        """
        super().__init__(
            cluster_connection_string=cluster_connection_string,
            authenticator=authenticator,
            cluster_options=cluster_options,
            bucket=bucket,
            scope=scope,
            collection=collection,
            **kwargs,
        )
        self.vector_search_index = vector_search_index
        self.is_global_level_index = is_global_level_index

    def to_dict(self) -> Dict[str, Any]:
        """Serializes the component to a dictionary.

        Returns:
            Dictionary with serialized data.
        """
        return default_to_dict(
            self,
            **self._base_to_dict(),
            vector_search_index=self.vector_search_index,
            is_global_level_index=self.is_global_level_index,
        )

    @classmethod
    def from_dict(cls, data: Dict[str, Any]) -> "CouchbaseSearchDocumentStore":
        """Deserializes the component from a dictionary.

        Args:
            data: Dictionary to deserialize from.

        Returns:
            Deserialized component.
        """
        if data["init_parameters"]["authenticator"]["type"] == generate_qualified_class_name(CouchbasePasswordAuthenticator):
            data["init_parameters"]["authenticator"] = CouchbasePasswordAuthenticator.from_dict(
                data["init_parameters"]["authenticator"]
            )
        else:
            data["init_parameters"]["authenticator"] = CouchbaseCertificateAuthenticator.from_dict(
                data["init_parameters"]["authenticator"]
            )
        data["init_parameters"]["cluster_options"] = CouchbaseClusterOptions.from_dict(data["init_parameters"]["cluster_options"])
        deserialize_secrets_inplace(data["init_parameters"], keys=["cluster_connection_string"])
        return default_from_dict(cls, data)

    def _get_search_interface(self):
        """Returns the appropriate Couchbase search interface based on the `is_global_level_index` configuration.

        Returns:
            The Couchbase search index manager object.
        """
        if not self.is_global_level_index:
            return self.scope.search_indexes()
        return self.connection.search_indexes()

    def count_documents(self) -> int:
        """Returns how many documents are present in the document store.

        Returns:
            The number of documents in the document store.
        """
        search_interface = self._get_search_interface()
        return search_interface.get_indexed_documents_count(self.vector_search_index)

    def filter_documents(self, filters: Optional[Dict[str, Any]] = None) -> List[Document]:
        """Returns the documents that match the filters provided.

        For a detailed specification of the filters,
        refer to the Haystack [documentation](https://docs.haystack.deepset.ai/v2.0/docs/metadata-filtering).

        Args:
            filters: The filters to apply. It returns only the documents that match the filters.

        Returns:
            A list of Documents that match the given filters.

        Raises:
            DocumentStoreError: If the search request fails.
        """
        search_filters: SearchQuery
        if filters:
            search_filters = _normalize_filters(filters)
        else:
            search_filters = search.MatchAllQuery()
        logger.debug(search_filters.encodable)
        request = search.SearchRequest(search_filters)
        options = SearchOptions(fields=["*"], limit=10000)

        if not self.is_global_level_index:
            response = self.scope.search(self.vector_search_index, request, options)
        else:
            response = self.connection.search(self.vector_search_index, request, options)

        return self.__get_doc_from_kv(response)

    def _embedding_retrieval(
        self,
        query_embedding: List[float],
        top_k: int = 10,
        filters: Optional[Dict[str, Any]] = None,
        search_query: SearchQuery = None,
        limit: Optional[int] = None,
    ) -> List[Document]:
        """Find the documents that are most similar to the provided `query_embedding` by using a vector similarity metric.

        Args:
            query_embedding: Embedding of the query
            top_k: How many documents to be returned by the vector query
            filters: Optional dictionary of filters to apply before the vector search.
                     Refer to Haystack documentation for filter structure (https://docs.haystack.deepset.ai/v2.0/docs/metadata-filtering).
            search_query: Search filters param which is parsed to the Couchbase search query. The vector query and
                          search query are ORed operation.
            limit: Maximum number of Documents to return. Defaults to top_k if not specified.

        Returns:
            A list of Documents that are most similar to the given `query_embedding`

        Raises:
            ValueError: If `query_embedding` is empty
            DocumentStoreError: If the retrieval of documents from Couchbase fails
        """
        if not query_embedding:
            msg = "Query embedding must not be empty"
            raise ValueError(msg)
        pre_filter: Optional[SearchQuery] = None
        if filters is not None:
            pre_filter = _normalize_filters(filters)
            logger.debug(f"pre_filter.encodable: {pre_filter.encodable}")

        vector_search = VectorSearch.from_vector_query(
            VectorQuery(field_name="embedding", vector=query_embedding, num_candidates=top_k, prefilter=pre_filter)
        )
        request = search.SearchRequest.create(vector_search)
        if search_query:
            request.with_search_query(search_query)

        if limit is None:
            limit = top_k
        options = SearchOptions(fields=["*"], limit=limit)

        if not self.is_global_level_index:
            response = self.scope.search(self.vector_search_index, request, options)
        else:
            response = self.connection.search(self.vector_search_index, request, options)

        return self.__get_doc_from_kv(response)

    def __get_doc_from_kv(self, response: SearchResult) -> List[Document]:
        """Fetches the full document content from Couchbase KV storage based on IDs from a SearchResult.

        This helper method takes the results of an FTS/Vector search (which might only contain IDs and scores)
        and retrieves the complete documents using a multi-get operation for efficiency.

        Args:
            response: The `SearchResult` object containing document IDs and scores.

        Returns:
            A list of Haystack `Document` objects, populated with content and scores.

        Raises:
            DocumentStoreError: If fetching documents from KV fails for any ID.
        """
        documents: List[Document] = []
        ids: List[str] = []
        scores: List[float] = []
        for doc in response.rows():
            ids.append(doc.id)
            scores.append(doc.score)
        kv_response = self.collection.get_multi(keys=ids)
        if not kv_response.all_ok and kv_response.exceptions:
            errors = []
            for id, ex in kv_response.exceptions.items():
                errors.append({"id": id, "exception": ex})
            if len(errors) > 0:
                msg = f"Failed to write documents to couchbase. Errors:\n{errors}"
                raise DocumentStoreError(msg)
        for i, id in enumerate(ids):
            get_result = kv_response.results.get(id)
            if get_result is not None and get_result.success:
                value = get_result.value
                value["id"] = id
                value["score"] = scores[i]
            documents.append(Document.from_dict(value))
        return documents

__init__ ¤

__init__(
    *,
    cluster_connection_string: Secret = Secret.from_env_var(
        "CB_CONNECTION_STRING"
    ),
    authenticator: Union[
        CouchbasePasswordAuthenticator, CouchbaseCertificateAuthenticator
    ],
    cluster_options: CouchbaseClusterOptions = CouchbaseClusterOptions(),
    bucket: str,
    scope: str,
    collection: str,
    vector_search_index: str,
    is_global_level_index: bool = False,
    **kwargs: Dict[str, Any]
)

Parameters:

  • cluster_connection_string (Secret, default: from_env_var('CB_CONNECTION_STRING') ) –

    Connection string for the Couchbase cluster

  • authenticator (Union[CouchbasePasswordAuthenticator, CouchbaseCertificateAuthenticator]) –

    Authentication method (password or certificate based)

  • cluster_options (CouchbaseClusterOptions, default: CouchbaseClusterOptions() ) –

    Options for configuring the cluster connection

  • bucket (str) –

    Name of the Couchbase bucket to use

  • scope (str) –

    Name of the scope within the bucket

  • collection (str) –

    Name of the collection within the scope

  • vector_search_index (str) –

    Name of the FTS index (which must include vector indexing) to use for searches.

  • is_global_level_index (bool, default: False ) –

    If True, use a global (bucket-level) FTS index. If False (default), use a scope-level FTS index.

  • kwargs (Dict[str, Any], default: {} ) –

    Additional keyword arguments passed to the Cluster constructor.

Source code in src/couchbase_haystack/document_stores/document_store.py 
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
def __init__(
    self,
    *,
    cluster_connection_string: Secret = Secret.from_env_var("CB_CONNECTION_STRING"),
    authenticator: Union[CouchbasePasswordAuthenticator, CouchbaseCertificateAuthenticator],
    cluster_options: CouchbaseClusterOptions = CouchbaseClusterOptions(),
    bucket: str,
    scope: str,
    collection: str,
    vector_search_index: str,
    is_global_level_index: bool = False,
    **kwargs: Dict[str, Any],
):
    """Creates a new CouchbaseSearchDocumentStore instance.

    Args:
        cluster_connection_string: Connection string for the Couchbase cluster
        authenticator: Authentication method (password or certificate based)
        cluster_options: Options for configuring the cluster connection
        bucket: Name of the Couchbase bucket to use
        scope: Name of the scope within the bucket
        collection: Name of the collection within the scope
        vector_search_index: Name of the FTS index (which must include vector indexing) to use for searches.
        is_global_level_index: If `True`, use a global (bucket-level) FTS index.
                              If `False` (default), use a scope-level FTS index.
        kwargs: Additional keyword arguments passed to the Cluster constructor.
    """
    super().__init__(
        cluster_connection_string=cluster_connection_string,
        authenticator=authenticator,
        cluster_options=cluster_options,
        bucket=bucket,
        scope=scope,
        collection=collection,
        **kwargs,
    )
    self.vector_search_index = vector_search_index
    self.is_global_level_index = is_global_level_index

to_dict ¤

to_dict() -> Dict[str, Any]

Serializes the component to a dictionary.

Returns:

  • Dict[str, Any] –

    Dictionary with serialized data.

Source code in src/couchbase_haystack/document_stores/document_store.py 
405
406
407
408
409
410
411
412
413
414
415
416
def to_dict(self) -> Dict[str, Any]:
    """Serializes the component to a dictionary.

    Returns:
        Dictionary with serialized data.
    """
    return default_to_dict(
        self,
        **self._base_to_dict(),
        vector_search_index=self.vector_search_index,
        is_global_level_index=self.is_global_level_index,
    )

from_dict classmethod ¤

from_dict(data: Dict[str, Any]) -> CouchbaseSearchDocumentStore

Deserializes the component from a dictionary.

Parameters:

  • data (Dict[str, Any]) –

    Dictionary to deserialize from.

Returns:

Source code in src/couchbase_haystack/document_stores/document_store.py 
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
@classmethod
def from_dict(cls, data: Dict[str, Any]) -> "CouchbaseSearchDocumentStore":
    """Deserializes the component from a dictionary.

    Args:
        data: Dictionary to deserialize from.

    Returns:
        Deserialized component.
    """
    if data["init_parameters"]["authenticator"]["type"] == generate_qualified_class_name(CouchbasePasswordAuthenticator):
        data["init_parameters"]["authenticator"] = CouchbasePasswordAuthenticator.from_dict(
            data["init_parameters"]["authenticator"]
        )
    else:
        data["init_parameters"]["authenticator"] = CouchbaseCertificateAuthenticator.from_dict(
            data["init_parameters"]["authenticator"]
        )
    data["init_parameters"]["cluster_options"] = CouchbaseClusterOptions.from_dict(data["init_parameters"]["cluster_options"])
    deserialize_secrets_inplace(data["init_parameters"], keys=["cluster_connection_string"])
    return default_from_dict(cls, data)

_get_search_interface ¤

_get_search_interface()

Returns the appropriate Couchbase search interface based on the is_global_level_index configuration.

Returns:

  • –

    The Couchbase search index manager object.

Source code in src/couchbase_haystack/document_stores/document_store.py 
440
441
442
443
444
445
446
447
448
def _get_search_interface(self):
    """Returns the appropriate Couchbase search interface based on the `is_global_level_index` configuration.

    Returns:
        The Couchbase search index manager object.
    """
    if not self.is_global_level_index:
        return self.scope.search_indexes()
    return self.connection.search_indexes()

count_documents ¤

count_documents() -> int

Returns how many documents are present in the document store.

Returns:

  • int –

    The number of documents in the document store.

Source code in src/couchbase_haystack/document_stores/document_store.py 
450
451
452
453
454
455
456
457
def count_documents(self) -> int:
    """Returns how many documents are present in the document store.

    Returns:
        The number of documents in the document store.
    """
    search_interface = self._get_search_interface()
    return search_interface.get_indexed_documents_count(self.vector_search_index)

filter_documents ¤

filter_documents(filters: Optional[Dict[str, Any]] = None) -> List[Document]

Returns the documents that match the filters provided.

For a detailed specification of the filters, refer to the Haystack documentation.

Parameters:

  • filters (Optional[Dict[str, Any]], default: None ) –

    The filters to apply. It returns only the documents that match the filters.

Returns:

  • List[Document] –

    A list of Documents that match the given filters.

Raises:

  • DocumentStoreError –

    If the search request fails.

Source code in src/couchbase_haystack/document_stores/document_store.py 
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
def filter_documents(self, filters: Optional[Dict[str, Any]] = None) -> List[Document]:
    """Returns the documents that match the filters provided.

    For a detailed specification of the filters,
    refer to the Haystack [documentation](https://docs.haystack.deepset.ai/v2.0/docs/metadata-filtering).

    Args:
        filters: The filters to apply. It returns only the documents that match the filters.

    Returns:
        A list of Documents that match the given filters.

    Raises:
        DocumentStoreError: If the search request fails.
    """
    search_filters: SearchQuery
    if filters:
        search_filters = _normalize_filters(filters)
    else:
        search_filters = search.MatchAllQuery()
    logger.debug(search_filters.encodable)
    request = search.SearchRequest(search_filters)
    options = SearchOptions(fields=["*"], limit=10000)

    if not self.is_global_level_index:
        response = self.scope.search(self.vector_search_index, request, options)
    else:
        response = self.connection.search(self.vector_search_index, request, options)

    return self.__get_doc_from_kv(response)

_embedding_retrieval ¤

_embedding_retrieval(
    query_embedding: List[float],
    top_k: int = 10,
    filters: Optional[Dict[str, Any]] = None,
    search_query: SearchQuery = None,
    limit: Optional[int] = None,
) -> List[Document]

Find the documents that are most similar to the provided query_embedding by using a vector similarity metric.

Parameters:

  • query_embedding (List[float]) –

    Embedding of the query

  • top_k (int, default: 10 ) –

    How many documents to be returned by the vector query

  • filters (Optional[Dict[str, Any]], default: None ) –

    Optional dictionary of filters to apply before the vector search. Refer to Haystack documentation for filter structure (https://docs.haystack.deepset.ai/v2.0/docs/metadata-filtering).

  • search_query (SearchQuery, default: None ) –

    Search filters param which is parsed to the Couchbase search query. The vector query and search query are ORed operation.

  • limit (Optional[int], default: None ) –

    Maximum number of Documents to return. Defaults to top_k if not specified.

Returns:

  • List[Document] –

    A list of Documents that are most similar to the given query_embedding

Raises:

  • ValueError –

    If query_embedding is empty

  • DocumentStoreError –

    If the retrieval of documents from Couchbase fails

Source code in src/couchbase_haystack/document_stores/document_store.py 
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
def _embedding_retrieval(
    self,
    query_embedding: List[float],
    top_k: int = 10,
    filters: Optional[Dict[str, Any]] = None,
    search_query: SearchQuery = None,
    limit: Optional[int] = None,
) -> List[Document]:
    """Find the documents that are most similar to the provided `query_embedding` by using a vector similarity metric.

    Args:
        query_embedding: Embedding of the query
        top_k: How many documents to be returned by the vector query
        filters: Optional dictionary of filters to apply before the vector search.
                 Refer to Haystack documentation for filter structure (https://docs.haystack.deepset.ai/v2.0/docs/metadata-filtering).
        search_query: Search filters param which is parsed to the Couchbase search query. The vector query and
                      search query are ORed operation.
        limit: Maximum number of Documents to return. Defaults to top_k if not specified.

    Returns:
        A list of Documents that are most similar to the given `query_embedding`

    Raises:
        ValueError: If `query_embedding` is empty
        DocumentStoreError: If the retrieval of documents from Couchbase fails
    """
    if not query_embedding:
        msg = "Query embedding must not be empty"
        raise ValueError(msg)
    pre_filter: Optional[SearchQuery] = None
    if filters is not None:
        pre_filter = _normalize_filters(filters)
        logger.debug(f"pre_filter.encodable: {pre_filter.encodable}")

    vector_search = VectorSearch.from_vector_query(
        VectorQuery(field_name="embedding", vector=query_embedding, num_candidates=top_k, prefilter=pre_filter)
    )
    request = search.SearchRequest.create(vector_search)
    if search_query:
        request.with_search_query(search_query)

    if limit is None:
        limit = top_k
    options = SearchOptions(fields=["*"], limit=limit)

    if not self.is_global_level_index:
        response = self.scope.search(self.vector_search_index, request, options)
    else:
        response = self.connection.search(self.vector_search_index, request, options)

    return self.__get_doc_from_kv(response)

__get_doc_from_kv ¤

__get_doc_from_kv(response: SearchResult) -> List[Document]

Fetches the full document content from Couchbase KV storage based on IDs from a SearchResult.

This helper method takes the results of an FTS/Vector search (which might only contain IDs and scores) and retrieves the complete documents using a multi-get operation for efficiency.

Parameters:

  • response (SearchResult) –

    The SearchResult object containing document IDs and scores.

Returns:

  • List[Document] –

    A list of Haystack Document objects, populated with content and scores.

Raises:

  • DocumentStoreError –

    If fetching documents from KV fails for any ID.

Source code in src/couchbase_haystack/document_stores/document_store.py 
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
def __get_doc_from_kv(self, response: SearchResult) -> List[Document]:
    """Fetches the full document content from Couchbase KV storage based on IDs from a SearchResult.

    This helper method takes the results of an FTS/Vector search (which might only contain IDs and scores)
    and retrieves the complete documents using a multi-get operation for efficiency.

    Args:
        response: The `SearchResult` object containing document IDs and scores.

    Returns:
        A list of Haystack `Document` objects, populated with content and scores.

    Raises:
        DocumentStoreError: If fetching documents from KV fails for any ID.
    """
    documents: List[Document] = []
    ids: List[str] = []
    scores: List[float] = []
    for doc in response.rows():
        ids.append(doc.id)
        scores.append(doc.score)
    kv_response = self.collection.get_multi(keys=ids)
    if not kv_response.all_ok and kv_response.exceptions:
        errors = []
        for id, ex in kv_response.exceptions.items():
            errors.append({"id": id, "exception": ex})
        if len(errors) > 0:
            msg = f"Failed to write documents to couchbase. Errors:\n{errors}"
            raise DocumentStoreError(msg)
    for i, id in enumerate(ids):
        get_result = kv_response.results.get(id)
        if get_result is not None and get_result.success:
            value = get_result.value
            value["id"] = id
            value["score"] = scores[i]
        documents.append(Document.from_dict(value))
    return documents

CouchbaseQueryDocumentStore¤

Bases: CouchbaseDocumentStore

CouchbaseQueryDocumentStore uses Couchbase Global Secondary Index (GSI) for high-performance vector search.

Supports two types of vector indexes:

  • Hyperscale Vector Indexes: Optimized for pure vector searches, scales to billions of documents. Best for chatbot context (RAG), reverse image search, and anomaly detection.

  • Composite Vector Indexes: Combines vector and scalar indexing. Applies scalar filters before vector search. Best for filtered recommendations, job searches, and supply chain management.

Search types: ANN (fast, approximate) or KNN (exact). Similarity metrics: COSINE, DOT, L2/EUCLIDEAN, L2_SQUARED/EUCLIDEAN_SQUARED.

See Couchbase documentation.

Source code in src/couchbase_haystack/document_stores/document_store.py 
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
class CouchbaseQueryDocumentStore(CouchbaseDocumentStore):
    """CouchbaseQueryDocumentStore uses Couchbase Global Secondary Index (GSI) for high-performance vector search.

    Supports two types of vector indexes:

    - **Hyperscale Vector Indexes**: Optimized for pure vector searches, scales to billions of documents.
      Best for chatbot context (RAG), reverse image search, and anomaly detection.

    - **Composite Vector Indexes**: Combines vector and scalar indexing. Applies scalar filters before vector search.
      Best for filtered recommendations, job searches, and supply chain management.

    Search types: ANN (fast, approximate) or KNN (exact).
    Similarity metrics: COSINE, DOT, L2/EUCLIDEAN, L2_SQUARED/EUCLIDEAN_SQUARED.

    See [Couchbase documentation](https://docs.couchbase.com/server/current/vector-index/use-vector-indexes.html).
    """

    def __init__(
        self,
        *,
        cluster_connection_string: Secret = Secret.from_env_var("CB_CONNECTION_STRING"),
        authenticator: Union[CouchbasePasswordAuthenticator, CouchbaseCertificateAuthenticator],
        cluster_options: CouchbaseClusterOptions = CouchbaseClusterOptions(),
        bucket: str,
        scope: str,
        collection: str,
        search_type: QueryVectorSearchType,
        similarity: Union[QueryVectorSearchSimilarity, str],
        nprobes: Optional[int] = None,
        query_options: CouchbaseQueryOptions = CouchbaseQueryOptions(
            timeout=timedelta(seconds=60), scan_consistency=QueryScanConsistency.NOT_BOUNDED
        ),
        **kwargs: Dict[str, Any],
    ):
        """Creates a new CouchbaseGSIDocumentStore instance.

        Args:
            cluster_connection_string: Connection string for the Couchbase cluster
            authenticator: Authentication method (password or certificate based)
            cluster_options: Options for configuring the cluster connection
            bucket: Name of the Couchbase bucket to use
            scope: Name of the scope within the bucket
            collection: Name of the collection within the scope
            search_type: Type of vector search (ANN or KNN).
            similarity: Similarity metric to use (COSINE, DOT, L2 or EUCLIDEAN, L2_SQUARED or EUCLIDEAN_SQUARED) or
            string representation of the enum.
            nprobes: Number of probes for the ANN search.
                Defaults to None, uses the value set at index creation time.
            query_options: Options controlling SQL++ query execution (timeout, scan consistency).
            kwargs: Additional keyword arguments passed to the `CouchbaseDocumentStore` base class constructor.
        """
        super().__init__(
            cluster_connection_string=cluster_connection_string,
            authenticator=authenticator,
            cluster_options=cluster_options,
            bucket=bucket,
            scope=scope,
            collection=collection,
            **kwargs,
        )
        self.search_type = QueryVectorSearchType(search_type) if isinstance(search_type, str) else search_type
        self.similarity: str = (
            similarity.upper()
            if isinstance(similarity, str)
            else (similarity.value if isinstance(similarity, QueryVectorSearchSimilarity) else None)
        )
        if self.similarity is None:
            err_msg = f"Invalid similarity metric: {similarity}"
            raise ValueError(err_msg)
        self.nprobes = nprobes
        self.query_options = query_options

    def to_dict(self) -> Dict[str, Any]:
        """Serializes the component to a dictionary.

        Returns:
            Dictionary with serialized data.
        """
        return default_to_dict(
            self,
            **self._base_to_dict(),  # cluster details
            query_options=self.query_options.to_dict() if self.query_options else None,
            search_type=self.search_type.value if isinstance(self.search_type, QueryVectorSearchType) else self.search_type,
            similarity=self.similarity,
            nprobes=self.nprobes,
            **self._kwargs,
        )

    @classmethod
    def from_dict(cls, data: Dict[str, Any]) -> "CouchbaseQueryDocumentStore":
        """Deserializes the component from a dictionary.

        Args:
            data: Dictionary to deserialize from.

        Returns:
            Deserialized component.
        """
        init_params = data.get("init_parameters", {})

        # Handle authenticator deserialization
        if init_params["authenticator"]["type"] == generate_qualified_class_name(CouchbasePasswordAuthenticator):
            init_params["authenticator"] = CouchbasePasswordAuthenticator.from_dict(init_params["authenticator"])
        else:
            init_params["authenticator"] = CouchbaseCertificateAuthenticator.from_dict(init_params["authenticator"])

        # Handle cluster options deserialization
        init_params["cluster_options"] = CouchbaseClusterOptions.from_dict(init_params["cluster_options"])

        if init_params["query_options"]:
            init_params["query_options"] = CouchbaseQueryOptions.from_dict(init_params["query_options"])

        # Handle secrets
        deserialize_secrets_inplace(init_params, keys=["cluster_connection_string"])

        return default_from_dict(cls, data)

    def count_documents(self) -> int:
        """Returns how many documents are present in the document store.

        Returns:
            The number of documents in the document store.
        """
        query = f"SELECT COUNT(*) as count FROM `{self.collection_name}`"  # noqa: S608
        result = self.scope.query(query, self.query_options.cb_query_options()).execute()
        return result[0]["count"]

    def filter_documents(self, filters: Optional[Dict[str, Any]] = None) -> List[Document]:
        """Returns the documents that match the filters provided.

        For a detailed specification of the filters,
        refer to the Haystack [documentation](https://docs.haystack.deepset.ai/v2.0/docs/metadata-filtering).

        Args:
            filters: The filters to apply using SQL++ WHERE clause syntax.
                    Refer to the Haystack documentation for filter structure.

        Returns:
            A list of Documents that match the given filters.

        Raises:
            DocumentStoreError: If the SQL++ query execution fails.
        """
        query_str = f"SELECT d.*, meta().id as id FROM `{self.collection_name}` as d"  # noqa: S608
        where_clause = ""

        if filters:
            normalized_filters = normalize_sql_filters(filters)
            where_clause = f" WHERE {normalized_filters}"
            query_str += where_clause
        try:
            result = self.scope.query(query_str, self.query_options.cb_query_options())
            documents = []

            for row in result.rows():
                # Convert row to Document
                doc_dict = row.copy()
                documents.append(Document.from_dict(doc_dict))

            return documents

        except Exception as e:
            msg = f"Failed to filter documents: {e}"
            logger.error(msg)
            raise DocumentStoreError(msg) from e

    def _embedding_retrieval(
        self,
        query_embedding: List[float],
        top_k: int = 5,
        filters: Optional[Dict[str, Any]] = None,
        nprobes: Optional[int] = None,
    ) -> List[Document]:
        """Find the documents that are most similar to the provided `query_embedding` by using a vector similarity metric.

        Args:
            query_embedding: Embedding of the query
            top_k: How many documents to retrieve based on vector similarity.
            filters: Optional dictionary of filters to apply using a SQL++ WHERE clause before the vector search.
            nprobes: Number of probes for the ANN search. If None, uses the value set at index creation time
            or the value set at the document store level.

        Returns:
            A list of Documents most similar to the `query_embedding`, potentially pre-filtered.

        Raises:
            ValueError: If `query_embedding` is empty.
            DocumentStoreError: If the SQL++ query execution fails.
        """
        if not query_embedding:
            msg = "Query embedding must not be empty"
            raise ValueError(msg)

        # Construct the SQL++ query with vector search
        query_context = f"`{self.bucket_name}`.`{self.scope_name}`.`{self.collection_name}`"

        # Convert embedding to string representation for query
        query_vector_str = str(query_embedding)

        # Handle filters if provided
        where_clause = ""
        if filters:
            normalized_filters = normalize_sql_filters(filters)
            where_clause = f"WHERE {normalized_filters}"

        if nprobes is None:
            nprobes = self.nprobes
        # Determine the appropriate distance function based on search type
        if self.search_type == QueryVectorSearchType.ANN:
            nprobes_exp = f", {nprobes}" if nprobes else ""
            distance_function_exp = f"APPROX_VECTOR_DISTANCE(d.embedding, {query_vector_str}, '{self.similarity}'{nprobes_exp})"
        else:
            distance_function_exp = f"VECTOR_DISTANCE(d.embedding, {query_vector_str}, '{self.similarity}')"

        # Build the query
        query_str = f"""
        SELECT d.*, meta().id as id, {distance_function_exp} as score
        FROM {query_context} d
        {where_clause}
        ORDER BY score
        LIMIT {top_k}
        """  # noqa: S608  # query_vector_str is a float array, where_clause is normalized by normalize_sql_filters

        try:

            query_options = self.query_options.cb_query_options()
            # Execute the query
            result: QueryResult = self.connection.query(
                query_str,
                query_options,
            )

            # Process results
            documents = []
            for row in result.rows():
                # Convert row to Document
                doc_dict = row.copy()
                documents.append(Document.from_dict(doc_dict))

            return documents

        except Exception as e:
            msg = f"Failed to retrieve documents with vector search: {e}"
            logger.error(msg)
            raise DocumentStoreError(msg) from e

__init__ ¤

__init__(
    *,
    cluster_connection_string: Secret = Secret.from_env_var(
        "CB_CONNECTION_STRING"
    ),
    authenticator: Union[
        CouchbasePasswordAuthenticator, CouchbaseCertificateAuthenticator
    ],
    cluster_options: CouchbaseClusterOptions = CouchbaseClusterOptions(),
    bucket: str,
    scope: str,
    collection: str,
    search_type: QueryVectorSearchType,
    similarity: Union[QueryVectorSearchSimilarity, str],
    nprobes: Optional[int] = None,
    query_options: CouchbaseQueryOptions = CouchbaseQueryOptions(
        timeout=(timedelta(seconds=60)),
        scan_consistency=(QueryScanConsistency.NOT_BOUNDED),
    ),
    **kwargs: Dict[str, Any]
)

Parameters:

Source code in src/couchbase_haystack/document_stores/document_store.py 
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
def __init__(
    self,
    *,
    cluster_connection_string: Secret = Secret.from_env_var("CB_CONNECTION_STRING"),
    authenticator: Union[CouchbasePasswordAuthenticator, CouchbaseCertificateAuthenticator],
    cluster_options: CouchbaseClusterOptions = CouchbaseClusterOptions(),
    bucket: str,
    scope: str,
    collection: str,
    search_type: QueryVectorSearchType,
    similarity: Union[QueryVectorSearchSimilarity, str],
    nprobes: Optional[int] = None,
    query_options: CouchbaseQueryOptions = CouchbaseQueryOptions(
        timeout=timedelta(seconds=60), scan_consistency=QueryScanConsistency.NOT_BOUNDED
    ),
    **kwargs: Dict[str, Any],
):
    """Creates a new CouchbaseGSIDocumentStore instance.

    Args:
        cluster_connection_string: Connection string for the Couchbase cluster
        authenticator: Authentication method (password or certificate based)
        cluster_options: Options for configuring the cluster connection
        bucket: Name of the Couchbase bucket to use
        scope: Name of the scope within the bucket
        collection: Name of the collection within the scope
        search_type: Type of vector search (ANN or KNN).
        similarity: Similarity metric to use (COSINE, DOT, L2 or EUCLIDEAN, L2_SQUARED or EUCLIDEAN_SQUARED) or
        string representation of the enum.
        nprobes: Number of probes for the ANN search.
            Defaults to None, uses the value set at index creation time.
        query_options: Options controlling SQL++ query execution (timeout, scan consistency).
        kwargs: Additional keyword arguments passed to the `CouchbaseDocumentStore` base class constructor.
    """
    super().__init__(
        cluster_connection_string=cluster_connection_string,
        authenticator=authenticator,
        cluster_options=cluster_options,
        bucket=bucket,
        scope=scope,
        collection=collection,
        **kwargs,
    )
    self.search_type = QueryVectorSearchType(search_type) if isinstance(search_type, str) else search_type
    self.similarity: str = (
        similarity.upper()
        if isinstance(similarity, str)
        else (similarity.value if isinstance(similarity, QueryVectorSearchSimilarity) else None)
    )
    if self.similarity is None:
        err_msg = f"Invalid similarity metric: {similarity}"
        raise ValueError(err_msg)
    self.nprobes = nprobes
    self.query_options = query_options

to_dict ¤

to_dict() -> Dict[str, Any]

Serializes the component to a dictionary.

Returns:

  • Dict[str, Any] –

    Dictionary with serialized data.

Source code in src/couchbase_haystack/document_stores/document_store.py 
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
def to_dict(self) -> Dict[str, Any]:
    """Serializes the component to a dictionary.

    Returns:
        Dictionary with serialized data.
    """
    return default_to_dict(
        self,
        **self._base_to_dict(),  # cluster details
        query_options=self.query_options.to_dict() if self.query_options else None,
        search_type=self.search_type.value if isinstance(self.search_type, QueryVectorSearchType) else self.search_type,
        similarity=self.similarity,
        nprobes=self.nprobes,
        **self._kwargs,
    )

from_dict classmethod ¤

from_dict(data: Dict[str, Any]) -> CouchbaseQueryDocumentStore

Deserializes the component from a dictionary.

Parameters:

  • data (Dict[str, Any]) –

    Dictionary to deserialize from.

Returns:

Source code in src/couchbase_haystack/document_stores/document_store.py 
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
@classmethod
def from_dict(cls, data: Dict[str, Any]) -> "CouchbaseQueryDocumentStore":
    """Deserializes the component from a dictionary.

    Args:
        data: Dictionary to deserialize from.

    Returns:
        Deserialized component.
    """
    init_params = data.get("init_parameters", {})

    # Handle authenticator deserialization
    if init_params["authenticator"]["type"] == generate_qualified_class_name(CouchbasePasswordAuthenticator):
        init_params["authenticator"] = CouchbasePasswordAuthenticator.from_dict(init_params["authenticator"])
    else:
        init_params["authenticator"] = CouchbaseCertificateAuthenticator.from_dict(init_params["authenticator"])

    # Handle cluster options deserialization
    init_params["cluster_options"] = CouchbaseClusterOptions.from_dict(init_params["cluster_options"])

    if init_params["query_options"]:
        init_params["query_options"] = CouchbaseQueryOptions.from_dict(init_params["query_options"])

    # Handle secrets
    deserialize_secrets_inplace(init_params, keys=["cluster_connection_string"])

    return default_from_dict(cls, data)

count_documents ¤

count_documents() -> int

Returns how many documents are present in the document store.

Returns:

  • int –

    The number of documents in the document store.

Source code in src/couchbase_haystack/document_stores/document_store.py 
698
699
700
701
702
703
704
705
706
def count_documents(self) -> int:
    """Returns how many documents are present in the document store.

    Returns:
        The number of documents in the document store.
    """
    query = f"SELECT COUNT(*) as count FROM `{self.collection_name}`"  # noqa: S608
    result = self.scope.query(query, self.query_options.cb_query_options()).execute()
    return result[0]["count"]

filter_documents ¤

filter_documents(filters: Optional[Dict[str, Any]] = None) -> List[Document]

Returns the documents that match the filters provided.

For a detailed specification of the filters, refer to the Haystack documentation.

Parameters:

  • filters (Optional[Dict[str, Any]], default: None ) –

    The filters to apply using SQL++ WHERE clause syntax. Refer to the Haystack documentation for filter structure.

Returns:

  • List[Document] –

    A list of Documents that match the given filters.

Raises:

  • DocumentStoreError –

    If the SQL++ query execution fails.

Source code in src/couchbase_haystack/document_stores/document_store.py 
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
def filter_documents(self, filters: Optional[Dict[str, Any]] = None) -> List[Document]:
    """Returns the documents that match the filters provided.

    For a detailed specification of the filters,
    refer to the Haystack [documentation](https://docs.haystack.deepset.ai/v2.0/docs/metadata-filtering).

    Args:
        filters: The filters to apply using SQL++ WHERE clause syntax.
                Refer to the Haystack documentation for filter structure.

    Returns:
        A list of Documents that match the given filters.

    Raises:
        DocumentStoreError: If the SQL++ query execution fails.
    """
    query_str = f"SELECT d.*, meta().id as id FROM `{self.collection_name}` as d"  # noqa: S608
    where_clause = ""

    if filters:
        normalized_filters = normalize_sql_filters(filters)
        where_clause = f" WHERE {normalized_filters}"
        query_str += where_clause
    try:
        result = self.scope.query(query_str, self.query_options.cb_query_options())
        documents = []

        for row in result.rows():
            # Convert row to Document
            doc_dict = row.copy()
            documents.append(Document.from_dict(doc_dict))

        return documents

    except Exception as e:
        msg = f"Failed to filter documents: {e}"
        logger.error(msg)
        raise DocumentStoreError(msg) from e

_embedding_retrieval ¤

_embedding_retrieval(
    query_embedding: List[float],
    top_k: int = 5,
    filters: Optional[Dict[str, Any]] = None,
    nprobes: Optional[int] = None,
) -> List[Document]

Find the documents that are most similar to the provided query_embedding by using a vector similarity metric.

Parameters:

  • query_embedding (List[float]) –

    Embedding of the query

  • top_k (int, default: 5 ) –

    How many documents to retrieve based on vector similarity.

  • filters (Optional[Dict[str, Any]], default: None ) –

    Optional dictionary of filters to apply using a SQL++ WHERE clause before the vector search.

  • nprobes (Optional[int], default: None ) –

    Number of probes for the ANN search. If None, uses the value set at index creation time

Returns:

  • List[Document] –

    A list of Documents most similar to the query_embedding, potentially pre-filtered.

Raises:

  • ValueError –

    If query_embedding is empty.

  • DocumentStoreError –

    If the SQL++ query execution fails.

Source code in src/couchbase_haystack/document_stores/document_store.py 
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
def _embedding_retrieval(
    self,
    query_embedding: List[float],
    top_k: int = 5,
    filters: Optional[Dict[str, Any]] = None,
    nprobes: Optional[int] = None,
) -> List[Document]:
    """Find the documents that are most similar to the provided `query_embedding` by using a vector similarity metric.

    Args:
        query_embedding: Embedding of the query
        top_k: How many documents to retrieve based on vector similarity.
        filters: Optional dictionary of filters to apply using a SQL++ WHERE clause before the vector search.
        nprobes: Number of probes for the ANN search. If None, uses the value set at index creation time
        or the value set at the document store level.

    Returns:
        A list of Documents most similar to the `query_embedding`, potentially pre-filtered.

    Raises:
        ValueError: If `query_embedding` is empty.
        DocumentStoreError: If the SQL++ query execution fails.
    """
    if not query_embedding:
        msg = "Query embedding must not be empty"
        raise ValueError(msg)

    # Construct the SQL++ query with vector search
    query_context = f"`{self.bucket_name}`.`{self.scope_name}`.`{self.collection_name}`"

    # Convert embedding to string representation for query
    query_vector_str = str(query_embedding)

    # Handle filters if provided
    where_clause = ""
    if filters:
        normalized_filters = normalize_sql_filters(filters)
        where_clause = f"WHERE {normalized_filters}"

    if nprobes is None:
        nprobes = self.nprobes
    # Determine the appropriate distance function based on search type
    if self.search_type == QueryVectorSearchType.ANN:
        nprobes_exp = f", {nprobes}" if nprobes else ""
        distance_function_exp = f"APPROX_VECTOR_DISTANCE(d.embedding, {query_vector_str}, '{self.similarity}'{nprobes_exp})"
    else:
        distance_function_exp = f"VECTOR_DISTANCE(d.embedding, {query_vector_str}, '{self.similarity}')"

    # Build the query
    query_str = f"""
    SELECT d.*, meta().id as id, {distance_function_exp} as score
    FROM {query_context} d
    {where_clause}
    ORDER BY score
    LIMIT {top_k}
    """  # noqa: S608  # query_vector_str is a float array, where_clause is normalized by normalize_sql_filters

    try:

        query_options = self.query_options.cb_query_options()
        # Execute the query
        result: QueryResult = self.connection.query(
            query_str,
            query_options,
        )

        # Process results
        documents = []
        for row in result.rows():
            # Convert row to Document
            doc_dict = row.copy()
            documents.append(Document.from_dict(doc_dict))

        return documents

    except Exception as e:
        msg = f"Failed to retrieve documents with vector search: {e}"
        logger.error(msg)
        raise DocumentStoreError(msg) from e

Supporting Classes¤

QueryVectorSearchType¤

Bases: str, Enum

Enum for search types supported by Couchbase GSI.

Source code in src/couchbase_haystack/document_stores/document_store.py 
39
40
41
42
43
class QueryVectorSearchType(str, Enum):
    """Enum for search types supported by Couchbase GSI."""

    ANN = "ANN"
    KNN = "KNN"

QueryVectorSearchSimilarity¤

Bases: str, Enum

Enum for similarity metrics supported by Couchbase GSI.

Source code in src/couchbase_haystack/document_stores/document_store.py 
46
47
48
49
50
51
52
53
54
class QueryVectorSearchSimilarity(str, Enum):
    """Enum for similarity metrics supported by Couchbase GSI."""

    COSINE = "COSINE"
    DOT = "DOT"
    L2 = "L2"
    EUCLIDEAN = "EUCLIDEAN"
    L2_SQUARED = "L2_SQUARED"
    EUCLIDEAN_SQUARED = "EUCLIDEAN_SQUARED"

CouchbaseQueryOptions¤

Dataclass for storing query options specifically for Couchbase SQL++ (N1QL) queries.

Parameters:

  • timeout (timedelta, default: timedelta(seconds=60) ) –

    The timeout duration for the query. Defaults to 60 seconds.

  • scan_consistency (Optional[Union[QueryScanConsistency, str]], default: None ) –

    The scan consistency level for the query. See couchbase.n1ql.QueryScanConsistency. Defaults to None, which implies Couchbase's default behavior.

Source code in src/couchbase_haystack/document_stores/document_store.py 
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
@dataclass
class CouchbaseQueryOptions:
    """Dataclass for storing query options specifically for Couchbase SQL++ (N1QL) queries.

    Args:
        timeout: The timeout duration for the query. Defaults to 60 seconds.
        scan_consistency: The scan consistency level for the query. See `couchbase.n1ql.QueryScanConsistency`.
                          Defaults to None, which implies Couchbase's default behavior.
    """

    timeout: timedelta = timedelta(seconds=60)
    scan_consistency: Optional[Union[QueryScanConsistency, str]] = None

    __cb_query_options: Optional[QueryOptions] = None

    def to_dict(self) -> Dict[str, Any]:
        """Serializes the CouchbaseQueryOptions object to a dictionary.

        Returns:
            A dictionary representation of the object.
        """
        return default_to_dict(
            self,
            timeout=self.timeout.total_seconds(),
            scan_consistency=(
                self.scan_consistency.value if isinstance(self.scan_consistency, QueryScanConsistency) else self.scan_consistency
            ),
        )

    @classmethod
    def from_dict(cls, data: Dict[str, Any]) -> "CouchbaseQueryOptions":
        """Deserializes a dictionary into a CouchbaseQueryOptions object.

        Args:
            data: The dictionary to deserialize from.

        Returns:
            A CouchbaseQueryOptions instance.
        """
        init_parameters = data.get("init_parameters", {})
        init_parameters["scan_consistency"] = (
            QueryScanConsistency(init_parameters["scan_consistency"]) if init_parameters.get("scan_consistency") else None
        )
        init_parameters["timeout"] = timedelta(seconds=init_parameters.get("timeout")) if init_parameters.get("timeout") else None
        return default_from_dict(cls, data)

    def cb_query_options(self) -> QueryOptions:
        """Returns the underlying Couchbase SDK `QueryOptions` object.

        Returns:
            The configured `couchbase.options.QueryOptions` instance.
        """
        return QueryOptions(
            timeout=self.timeout.total_seconds(),
            scan_consistency=(
                self.scan_consistency.value if isinstance(self.scan_consistency, QueryScanConsistency) else self.scan_consistency
            ),
        )

to_dict ¤

to_dict() -> Dict[str, Any]

Serializes the CouchbaseQueryOptions object to a dictionary.

Returns:

  • Dict[str, Any] –

    A dictionary representation of the object.

Source code in src/couchbase_haystack/document_stores/document_store.py 
72
73
74
75
76
77
78
79
80
81
82
83
84
def to_dict(self) -> Dict[str, Any]:
    """Serializes the CouchbaseQueryOptions object to a dictionary.

    Returns:
        A dictionary representation of the object.
    """
    return default_to_dict(
        self,
        timeout=self.timeout.total_seconds(),
        scan_consistency=(
            self.scan_consistency.value if isinstance(self.scan_consistency, QueryScanConsistency) else self.scan_consistency
        ),
    )

from_dict classmethod ¤

from_dict(data: Dict[str, Any]) -> CouchbaseQueryOptions

Deserializes a dictionary into a CouchbaseQueryOptions object.

Parameters:

  • data (Dict[str, Any]) –

    The dictionary to deserialize from.

Returns:

Source code in src/couchbase_haystack/document_stores/document_store.py 
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
@classmethod
def from_dict(cls, data: Dict[str, Any]) -> "CouchbaseQueryOptions":
    """Deserializes a dictionary into a CouchbaseQueryOptions object.

    Args:
        data: The dictionary to deserialize from.

    Returns:
        A CouchbaseQueryOptions instance.
    """
    init_parameters = data.get("init_parameters", {})
    init_parameters["scan_consistency"] = (
        QueryScanConsistency(init_parameters["scan_consistency"]) if init_parameters.get("scan_consistency") else None
    )
    init_parameters["timeout"] = timedelta(seconds=init_parameters.get("timeout")) if init_parameters.get("timeout") else None
    return default_from_dict(cls, data)

cb_query_options ¤

cb_query_options() -> QueryOptions

Returns the underlying Couchbase SDK QueryOptions object.

Returns:

  • QueryOptions –

    The configured couchbase.options.QueryOptions instance.

Source code in src/couchbase_haystack/document_stores/document_store.py 
103
104
105
106
107
108
109
110
111
112
113
114
def cb_query_options(self) -> QueryOptions:
    """Returns the underlying Couchbase SDK `QueryOptions` object.

    Returns:
        The configured `couchbase.options.QueryOptions` instance.
    """
    return QueryOptions(
        timeout=self.timeout.total_seconds(),
        scan_consistency=(
            self.scan_consistency.value if isinstance(self.scan_consistency, QueryScanConsistency) else self.scan_consistency
        ),
    )