minor improvements; feedback from Christophe; agnostic language about…

… 'thread'-related constants

libs/astradb/langchain_astradb/storage.py

@@ -25,8 +25,8 @@
 from langchain_core.stores import BaseStore, ByteStore
 
 from langchain_astradb.utils.astradb import (
-    INSERT_DOCUMENT_MAX_THREADS,
-    REPLACE_DOCUMENTS_MAX_THREADS,
+    MAX_CONCURRENT_DOCUMENT_INSERTIONS,
+    MAX_CONCURRENT_DOCUMENT_REPLACEMENTS,
     SetupMode,
     _AstraDBCollectionEnvironment,
 )
@@ -94,7 +94,7 @@ def mset(self, key_value_pairs: Sequence[Tuple[str, V]]) -> None:
             self.collection.insert_many(
                 documents_to_insert,
                 ordered=False,
-                concurrency=INSERT_DOCUMENT_MAX_THREADS,
+                concurrency=MAX_CONCURRENT_DOCUMENT_INSERTIONS,
             )
             ids_to_replace = []
         except InsertManyException as err:
@@ -114,7 +114,7 @@ def mset(self, key_value_pairs: Sequence[Tuple[str, V]]) -> None:
             ]
 
             with ThreadPoolExecutor(
-                max_workers=REPLACE_DOCUMENTS_MAX_THREADS
+                max_workers=MAX_CONCURRENT_DOCUMENT_REPLACEMENTS
             ) as executor:
 
                 def _replace_document(document: Dict[str, Any]) -> UpdateResult:
@@ -167,7 +167,7 @@ async def amset(self, key_value_pairs: Sequence[Tuple[str, V]]) -> None:
                 if document["_id"] in ids_to_replace
             ]
 
-            sem = asyncio.Semaphore(REPLACE_DOCUMENTS_MAX_THREADS)
+            sem = asyncio.Semaphore(MAX_CONCURRENT_DOCUMENT_REPLACEMENTS)
 
             _async_collection = self.async_collection
 

libs/astradb/langchain_astradb/utils/astradb.py

@@ -25,11 +25,11 @@
 # Chunk size for many-document insertions (None meaning defer to astrapy):
 DEFAULT_DOCUMENT_CHUNK_SIZE = None
 # thread/coroutine count for bulk inserts
-INSERT_DOCUMENT_MAX_THREADS = 20
+MAX_CONCURRENT_DOCUMENT_INSERTIONS = 20
 # Thread/coroutine count for one-doc-at-a-time overwrites
-REPLACE_DOCUMENTS_MAX_THREADS = 20
+MAX_CONCURRENT_DOCUMENT_REPLACEMENTS = 20
 # Thread/coroutine count for one-doc-at-a-time deletes:
-DELETE_DOCUMENTS_MAX_THREADS = 20
+MAX_CONCURRENT_DOCUMENT_DELETIONS = 20
 
 logger = logging.getLogger()
 

libs/astradb/langchain_astradb/vectorstores.py

@@ -38,9 +38,9 @@
 
 from langchain_astradb.utils.astradb import (
     DEFAULT_DOCUMENT_CHUNK_SIZE,
-    DELETE_DOCUMENTS_MAX_THREADS,
-    INSERT_DOCUMENT_MAX_THREADS,
-    REPLACE_DOCUMENTS_MAX_THREADS,
+    MAX_CONCURRENT_DOCUMENT_DELETIONS,
+    MAX_CONCURRENT_DOCUMENT_INSERTIONS,
+    MAX_CONCURRENT_DOCUMENT_REPLACEMENTS,
     SetupMode,
     _AstraDBCollectionEnvironment,
 )
@@ -299,13 +299,13 @@ def __init__(
         # Concurrency settings
         self.batch_size: Optional[int] = batch_size or DEFAULT_DOCUMENT_CHUNK_SIZE
         self.bulk_insert_batch_concurrency: int = (
-            bulk_insert_batch_concurrency or INSERT_DOCUMENT_MAX_THREADS
+            bulk_insert_batch_concurrency or MAX_CONCURRENT_DOCUMENT_INSERTIONS
         )
         self.bulk_insert_overwrite_concurrency: int = (
-            bulk_insert_overwrite_concurrency or REPLACE_DOCUMENTS_MAX_THREADS
+            bulk_insert_overwrite_concurrency or MAX_CONCURRENT_DOCUMENT_REPLACEMENTS
         )
         self.bulk_delete_concurrency: int = (
-            bulk_delete_concurrency or DELETE_DOCUMENTS_MAX_THREADS
+            bulk_delete_concurrency or MAX_CONCURRENT_DOCUMENT_DELETIONS
         )
         # "vector-related" settings
         self.metric = metric
@@ -1626,6 +1626,7 @@ def from_texts(
 
         Args:
             texts: the texts to insert.
+            embedding: the embedding function to use in the store.
             metadatas: metadata dicts for the texts.
             ids: ids to associate to the texts.
             **kwargs: you can pass any argument that you would