langroid 0.1.157__py3-none-any.whl → 0.1.159__py3-none-any.whl

langroid/agent/special/doc_chat_agent.py

@@ -14,7 +14,7 @@ pip install "langroid[hf-embeddings]"
 """
 import logging
 from contextlib import ExitStack
-from typing import List, Optional, Tuple, no_type_check
+from typing import Dict, List, Optional, Set, Tuple, no_type_check
 
 import numpy as np
 import pandas as pd
@@ -77,6 +77,7 @@ class DocChatAgentConfig(ChatAgentConfig):
     # extra fields to include in content as key=value pairs
     # (helps retrieval for table-like data)
     add_fields_to_content: List[str] = []
+    filter_fields: List[str] = []  # fields usable in filter
     retrieve_only: bool = False  # only retr relevant extracts, don't gen summary answer
     extraction_granularity: int = 1  # granularity (in sentences) for relev extraction
     filter: str | None = (
@@ -335,7 +336,9 @@ class DocChatAgent(ChatAgent):
         Ingest a dataframe into vecdb.
         """
         self.from_dataframe = True
-        self.df_description = describe_dataframe(df, sample_size=3)
+        self.df_description = describe_dataframe(
+            df, filter_fields=self.config.filter_fields, n_vals=5
+        )
         df, metadata = DocChatAgent.document_compatible_dataframe(df, content, metadata)
         docs = dataframe_to_documents(df, content="content", metadata=metadata)
         # When ingesting a dataframe we will no longer do any chunking,
@@ -374,6 +377,42 @@ class DocChatAgent(ChatAgent):
             for d in self.chunked_docs
         ]
 
+    def get_field_values(self, fields: list[str]) -> Dict[str, str]:
+        """Get string-listing of possible values of each filterable field,
+        e.g.
+        {
+            "genre": "crime, drama, mystery, ... (10 more)",
+            "certificate": "R, PG-13, PG, R",
+        }
+        """
+        field_values: Dict[str, Set[str]] = {}
+        # make empty set for each field
+        for f in fields:
+            field_values[f] = set()
+        if self.vecdb is None:
+            raise ValueError("VecDB not set")
+        # get all documents and accumulate possible values of each field until 10
+        docs = self.vecdb.get_all_documents()  # only works for vecdbs that support this
+        for d in docs:
+            # extract fields from d
+            doc_field_vals = extract_fields(d, fields)
+            for field, val in doc_field_vals.items():
+                field_values[field].add(val)
+        # For each field make a string showing list of possible values,
+        # truncate to 20 values, and if there are more, indicate how many
+        # more there are, e.g. Genre: crime, drama, mystery, ... (20 more)
+        field_values_list = {}
+        for f in fields:
+            vals = list(field_values[f])
+            n = len(vals)
+            remaining = n - 20
+            vals = vals[:20]
+            if n > 20:
+                vals.append(f"(...{remaining} more)")
+            # make a string of the values, ensure they are strings
+            field_values_list[f] = ", ".join(str(v) for v in vals)
+        return field_values_list
+
     def doc_length(self, docs: List[Document]) -> int:
         """
         Calc token-length of a list of docs

langroid/agent/special/lance_doc_chat_agent.py

@@ -44,6 +44,24 @@ class LanceDocChatAgent(DocChatAgent):
             self.vecdb.schema,
             excludes=["id", "vector"],
         )
+        # intersect config.filter_fields with schema_dict.keys() in case
+        # there are extraneous fields in config.filter_fields
+        filter_fields_set = set(
+            self.config.filter_fields or schema_dict.keys()
+        ).intersection(schema_dict.keys())
+
+        # possible values of filterable fields
+        filter_field_values = self.get_field_values(list(filter_fields_set))
+
+        # add field values to schema_dict as another field `values` for each field
+        for field, values in filter_field_values.items():
+            if field in schema_dict:
+                schema_dict[field]["values"] = values
+        # if self.config.filter_fields is set, restrict to these:
+        if len(self.config.filter_fields) > 0:
+            schema_dict = {
+                k: v for k, v in schema_dict.items() if k in self.config.filter_fields
+            }
         schema = json.dumps(schema_dict, indent=4)
         if len(fields := self.config.add_fields_to_content) > 0:
             schema += f"""
@@ -122,7 +140,11 @@ class LanceDocChatAgent(DocChatAgent):
             )
 
         df, metadata = DocChatAgent.document_compatible_dataframe(df, content, metadata)
-        self.df_description = describe_dataframe(df, sample_size=3)
+        self.df_description = describe_dataframe(
+            df,
+            filter_fields=self.config.filter_fields,
+            n_vals=10,
+        )
         self.vecdb.add_dataframe(df, content="content", metadata=metadata)
 
         tbl = self.vecdb.client.open_table(self.vecdb.config.collection_name)

langroid/agent/special/lance_rag/critic_agent.py

@@ -27,59 +27,66 @@ from langroid.agent.special.lance_rag.query_planner_agent import (
     LanceQueryPlanAgentConfig,
 )
 from langroid.mytypes import Entity
-from langroid.utils.constants import DONE, PASS
+from langroid.utils.constants import DONE, NO_ANSWER, PASS
 
 logger = logging.getLogger(__name__)
 
 
 class QueryPlanCriticConfig(LanceQueryPlanAgentConfig):
     name = "QueryPlanCritic"
-    system_message = """
+    system_message = f"""
     You are an expert at carefully planning a query that needs to be answered
     based on a large collection of documents. These docs have a special `content` field
     and additional FILTERABLE fields in the SCHEMA below:
     
-    {doc_schema}
-    
-    You will receive a QUERY PLAN consisting of a 
-    ORIGINAL QUERY, SQL-Like FILTER, REPHRASED QUERY, 
-    a DATAFRAME CALCULATION, and an ANSWER which is the 
-    answer received from an assistant that used this QUERY PLAN.
+    {{doc_schema}}
     
+    You will receive a QUERY PLAN consisting of:
+    - ORIGINAL QUERY, 
+    - SQL-Like FILTER, WHICH CAN BE EMPTY (and it's fine if results sound reasonable)
+      FILTER SHOULD ONLY BE USED IF EXPLICITLY REQUIRED BY THE QUERY.
+    - REPHRASED QUERY that will be used to match against the CONTENT (not filterable)
+         of the documents.
+      In general the REPHRASED QUERY should be relied upon to match the CONTENT 
+      of the docs. Thus the REPHRASED QUERY itself acts like a 
+      SEMANTIC/LEXICAL/FUZZY FILTER since the Assistant is able to use it to match 
+      the CONTENT of the docs in various ways (semantic, lexical, fuzzy, etc.).
+         
+    - DATAFRAME CALCULATION, and 
+    - ANSWER recieved from an assistant that used this QUERY PLAN.
+
+    In addition to the above SCHEMA fields there is a `content` field which:
+    - CANNOT appear in a FILTER, 
+    - CAN appear in the DATAFRAME CALCULATION.
+    THERE ARE NO OTHER FIELDS IN THE DOCUMENTS or in the RESULTING DATAFRAME.
+        
     Your job is to act as a CRITIC and provide feedback, 
     ONLY using the `query_plan_feedback` tool, and DO NOT SAY ANYTHING ELSE.
-    You must take `answer` field into account
-    and judge whether it is a reasonable answer, and accordingly give your feedback.
     
-    VERY IMPORTANT: IF THE ANSWER seems reasonable, then you should consider
-    the query plan to be fine, and only ask for a revision if you notice 
-    something obviously wrong with the query plan.
-    Typically, when there is a non-empty answer, 
-    you SHOULD NOT ASK FOR A REVISION (i.e. set `feedback` field as empty string),
-    unless there is some glaring problem.
-    
-    OTHERWISE:
-        
-    When giving feedback, SUGGEST CHANGES TO:
-    - the FILTER, if it appears too restrictive, e.g. prefer "LIKE" over "=", 
-        e.g. "CEO LIKE '%Jobs%'" instead of "CEO = 'Steve Jobs'" 
-    - the QUERY (i.e. possibly REPHRASED QUERY), and/or 
-    - the DATAFRAME CALCULATION, if any
+    Here is how you must examine the QUERY PLAN + ANSWER:
+    - If the ANSWER is in the expected form, then the QUERY PLAN is likely VALID,
+      and your feedback should be EMPTY.
+    - If the ANSWER is {NO_ANSWER} or of the wrong form, 
+      then try to DIAGNOSE the problem IN THE FOLLOWING ORDER:
+      - DATAFRAME CALCULATION -- is it doing the right thing?
+        Is it finding the Index of a row instead of the value in a column?
+        Or another example: mmaybe it is finding the maximum population
+           rather than the CITY with the maximum population?
+        If you notice a problem with the DATAFRAME CALCULATION, then
+        ONLY SUBMIT FEEDBACK ON THE DATAFRAME CALCULATION, and DO NOT
+        SUGGEST ANYTHING ELSE.
+      - If the DATAFRAME CALCULATION looks correct, then check if 
+        the REPHRASED QUERY makes sense given the ORIGINAL QUERY and FILTER.
+        If this is the problem, then ONLY SUBMIT FEEDBACK ON THE REPHRASED QUERY,
+        and DO NOT SUGGEST ANYTHING ELSE.
+      - If the REPHRASED QUERY looks correct, then check if the FILTER makes sense.
+        REMEMBER: A filter should ONLY be used if EXPLICITLY REQUIRED BY THE QUERY.
      
-    Keep these in mind:
-    * The FILTER must only use fields in the SCHEMA above, EXCEPT `content`
-    * The FILTER can be improved by RELAXING it, e.g. using "LIKE" instead of "=",
-        e.g. "CEO LIKE '%Jobs%'" instead of "CEO = 'Steve Jobs'" 
-    * The DATAFRAME CALCULATION must only use fields in the SCHEMA above. 
-    * The REPHRASED QUERY should NOT refer to any FILTER fields, and should
-        make sense with respect to the intended purpose, i.e. to be used to 
-        MATCH the CONTENT of the docs.
-    * The ASSISTANT does NOT know anything about the FILTER fields
-    * The DATAFRAME CALCULATION, if any, should be suitable to answer 
-       the user's ORIGINAL QUERY.
     
     ALWAYS use `query_plan_feedback` tool/fn to present your feedback!
     and DO NOT SAY ANYTHING ELSE OUTSIDE THE TOOL/FN.
+    IF NO REVISION NEEDED, simply give EMPTY FEEBACK, SAY NOTHING ELSE
+    and DO NOT EXPLAIN YOURSELF.
         
     """
 

langroid/agent/special/lance_rag/query_planner_agent.py

@@ -50,6 +50,7 @@ class LanceQueryPlanAgentConfig(ChatAgentConfig):
     -  a FILTER (can be empty string) that would help the ASSISTANT to answer the query.
         Remember the FILTER can only refer ANY fields in the above SCHEMA
         EXCEPT the `content` field of the documents. 
+        ONLY USE A FILTER IF EXPLICITLY MENTIONED IN THE QUERY.
         TO get good results, for STRING MATCHES, consider using LIKE instead of =, e.g.
         "CEO LIKE '%Jobs%'" instead of "CEO = 'Steve Jobs'"
     - a possibly REPHRASED QUERY to be answerable given the FILTER.

langroid/language_models/openai_gpt.py

@@ -169,7 +169,7 @@ class OpenAIGPTConfig(LLMConfig):
         local_model = "api_base" in kwargs and kwargs["api_base"] is not None
 
         chat_model = kwargs.get("chat_model", "")
-        if chat_model.startswith("litellm") or chat_model.startswith("local"):
+        if chat_model.startswith("litellm/") or chat_model.startswith("local/"):
             local_model = True
 
         warn_gpt_3_5 = (

langroid/parsing/table_loader.py

@@ -1,4 +1,5 @@
 from csv import Sniffer
+from typing import List
 
 import pandas as pd
 
@@ -50,32 +51,44 @@ def read_tabular_data(path_or_url: str, sep: None | str = None) -> pd.DataFrame:
         )
 
 
-def describe_dataframe(df: pd.DataFrame, sample_size: int = 5) -> str:
+def describe_dataframe(
+    df: pd.DataFrame, filter_fields: List[str] = [], n_vals: int = 10
+) -> str:
     """
-    Generates a description of the columns in the dataframe, along with typical values.
+    Generates a description of the columns in the dataframe,
+    along with a listing of up to `n_vals` unique values for each column.
     Intended to be used to insert into an LLM context so it can generate
     appropriate queries or filters on the df.
 
     Args:
-        df (pd.DataFrame): The dataframe to describe.
-        sample_size (int): The number of sample values to show for each column.
+    df (pd.DataFrame): The dataframe to describe.
+    filter_fields (list): A list of fields that can be used for filtering.
+        When non-empty, the values-list will be restricted to these.
+    n_vals (int): How many unique values to show for each column.
 
     Returns:
-        str: A description of the dataframe.
+    str: A description of the dataframe.
     """
     description = []
-    for column in df.columns:
-        sample_values = df[column].dropna().head(sample_size).tolist()
-        if len(sample_values) > 0 and isinstance(sample_values[0], str):
-            # truncate to 100 chars
-            sample_values = [v[:100] for v in sample_values]
+    for column in df.columns.to_list():
+        unique_values = df[column].dropna().unique()
+        unique_count = len(unique_values)
+        if column not in filter_fields:
+            values_desc = f"{unique_count} unique values"
+        else:
+            if unique_count > n_vals:
+                displayed_values = unique_values[:n_vals]
+                more_count = unique_count - n_vals
+                values_desc = f" Values - {displayed_values}, ... {more_count} more"
+            else:
+                values_desc = f" Values - {unique_values}"
         col_type = "string" if df[column].dtype == "object" else df[column].dtype
-        col_desc = f"* {column} ({col_type}): {sample_values}"
+        col_desc = f"* {column} ({col_type}); {values_desc}"
         description.append(col_desc)
 
     all_cols = "\n".join(description)
 
     return f"""
-        Name of each field, its type and some typical values:
+        Name of each field, its type and unique values (up to {n_vals}):
         {all_cols}
         """

langroid/utils/logging.py

@@ -72,12 +72,9 @@ def setup_file_logger(
     propagate: bool = False,
 ) -> logging.Logger:
     os.makedirs(os.path.dirname(filename), exist_ok=True)
-    if not append:
-        if os.path.exists(filename):
-            os.remove(filename)
-
+    file_mode = "a" if append else "w"
     logger = setup_logger(name)
-    handler = logging.FileHandler(filename)
+    handler = logging.FileHandler(filename, mode=file_mode)
     handler.setLevel(logging.INFO)
     if log_format:
         formatter = logging.Formatter(

langroid/utils/pandas_utils.py

@@ -8,7 +8,7 @@ def stringify(x: Any) -> str:
     if isinstance(x, pd.Series):
         df = x.to_frame()
     elif not isinstance(x, pd.DataFrame):
-        df = pd.DataFrame([x], columns=["Result"])
+        return str(x)
     else:
         df = x
 
@@ -16,7 +16,9 @@ def stringify(x: Any) -> str:
     for col in df.columns:
         if df[col].dtype == object:
             df[col] = df[col].apply(
-                lambda text: (text[:1000] + "...") if len(text) > 1000 else text
+                lambda item: (item[:1000] + "...")
+                if isinstance(item, str) and len(item) > 1000
+                else item
             )
 
     # Limit to 10 rows

{langroid-0.1.157.dist-info → langroid-0.1.159.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: langroid
-Version: 0.1.157
+Version: 0.1.159
 Summary: Harness LLMs with Multi-Agent Programming
 License: MIT
 Author: Prasad Chalasani
@@ -205,6 +205,9 @@ teacher_task.run()
 <summary> <b>Click to expand</b></summary>
 
 - **Jan 2024:**
+  - **[0.1.157](https://github.com/langroid/langroid/releases/tag/0.1.157):** `DocChatAgentConfig` 
+     has a new param: `add_fields_to_content`, to specify additional document fields to insert into 
+     the main `content` field, to help improve retrieval.
   - **[0.1.156](https://github.com/langroid/langroid/releases/tag/0.1.156):** New Task control signals
      PASS_TO, SEND_TO; VectorStore: Compute Pandas expression on documents; LanceRAGTaskCreator creates 3-agent RAG system with Query Planner, Critic and RAG Agent.
 - **Dec 2023:**

{langroid-0.1.157.dist-info → langroid-0.1.159.dist-info}/RECORD

@@ -8,13 +8,13 @@ langroid/agent/helpers.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 langroid/agent/junk,sha256=LxfuuW7Cijsg0szAzT81OjWWv1PMNI-6w_-DspVIO2s,339
 langroid/agent/openai_assistant.py,sha256=yBtxis64XOnxtJzlkwUoTm-wCyvKr4DGo9-laXYMok0,32654
 langroid/agent/special/__init__.py,sha256=PLGt63iwNnSCCigRYcqicD8AUXwdV_HRUWffRHllUC4,510
-langroid/agent/special/doc_chat_agent.py,sha256=eInd-HuhKV4V9VS4hZmjqF15xrLnv9twiTCMQPe889I,42901
-langroid/agent/special/lance_doc_chat_agent.py,sha256=7wg7UoUsUQt8FW4Qa8wkJE2IVBvKT3hYMSc0MYZcom8,6615
+langroid/agent/special/doc_chat_agent.py,sha256=0XMUTyIAmL-looB_Xb6TsljoywiiDfF6GjgldlL2K3E,44585
+langroid/agent/special/lance_doc_chat_agent.py,sha256=xZvB4adgGBp8L79wi1Z77kGkZmTTJg8dZU60heeO5cc,7570
 langroid/agent/special/lance_rag/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-langroid/agent/special/lance_rag/critic_agent.py,sha256=7JA3cBE6a9uBC33IN567mpmh8Jw1fGuLfbBwgMjnjmA,5228
+langroid/agent/special/lance_rag/critic_agent.py,sha256=9izW4keCxVZEqrFOgyVUHD7N1vTXLkRynXYYd1Vpwzw,5785
 langroid/agent/special/lance_rag/lance_rag_task.py,sha256=l_HQgrYY-CX2FwIsS961aEF3bYog3GDYo98fj0C0mSk,2889
 langroid/agent/special/lance_rag/lance_tools.py,sha256=WypIS-3ZMDqY_PZEGB2K80-o4RfS43_OnER0dyFlsDY,1339
-langroid/agent/special/lance_rag/query_planner_agent.py,sha256=BZJtTxXJS8vP_qGeoTqQHFZdqp8n5SUmKWkJvxECNiI,7816
+langroid/agent/special/lance_rag/query_planner_agent.py,sha256=Tf0-70Di4mw1B6efSp6dZvGcUPZO2lXjHkwEheSd_gQ,7880
 langroid/agent/special/relevance_extractor_agent.py,sha256=Z0OJmpCOESRX0Viar3JRcjjnDBhSA8useD1ZDflIo0s,4571
 langroid/agent/special/retriever_agent.py,sha256=uu6vqFg85uCVM-_DrXesYe2gH_-WcoHhlsKRlLuZPXk,1867
 langroid/agent/special/sql/__init__.py,sha256=3kR5nC0wnYIzmMrr9L8RJa7JAJpbwBLx7KKygiwz0v0,111
@@ -48,7 +48,7 @@ langroid/language_models/azure_openai.py,sha256=_OOEoZOziI3NDOH_8t3qmh8IDWoHESQe
 langroid/language_models/base.py,sha256=jUEUqDWJBVxIxmG6U4Ysg2QKGOnP_CLmRuEMicsSwUw,20596
 langroid/language_models/config.py,sha256=PXcmEUq52GCDj2sekt8F9E1flWyyNjP2S0LTRs7T6Kg,269
 langroid/language_models/openai_assistants.py,sha256=9K-DEAL2aSWHeXj2hwCo2RAlK9_1oCPtqX2u1wISCj8,36
-langroid/language_models/openai_gpt.py,sha256=T_PkzK0zSbyh1hA_Pvao2RGbeVhJsaJpjTMJKyLdVK8,42074
+langroid/language_models/openai_gpt.py,sha256=EF10UqGuAj-fZC6eAnssWGwLtA_pgW4e4ih64EP_LLA,42076
 langroid/language_models/prompt_formatter/__init__.py,sha256=wj2e6j7R9d3m63HCbSDY1vosjFuhHLQVlgBrq8iqF38,197
 langroid/language_models/prompt_formatter/base.py,sha256=2y_GcwhstvB5ih3haS7l5Fv79jVnFJ_vEw1jqWJzB9k,1247
 langroid/language_models/prompt_formatter/llama2_formatter.py,sha256=YdcO88qyBeuMENVIVvVqSYuEpvYSTndUe_jd6hVTko4,2899
@@ -66,7 +66,7 @@ langroid/parsing/parser.py,sha256=BwVJboobG71N08w5LC7Tu36LI4pEJoSgAdiBSLChWGY,10
 langroid/parsing/repo_loader.py,sha256=hhMfQBBSo-HvsZDQEcgmk_idKQQAeDQ_MMPd38x2ACU,29338
 langroid/parsing/search.py,sha256=xmQdAdTIwZ0REEUeQVFlGZlqf7k8Poah7-ALuyW7Ov0,8440
 langroid/parsing/spider.py,sha256=w_mHR1B4KOmxsBLoVI8kMkMTEbwTzeK3ath9fOMJrTk,3043
-langroid/parsing/table_loader.py,sha256=txnv_41z_fF0YDwwBWEtXqK8JjvWT5j45iakbhjel_o,2890
+langroid/parsing/table_loader.py,sha256=qNM4obT_0Y4tjrxNBCNUYjKQ9oETCZ7FbolKBTcz-GM,3410
 langroid/parsing/url_loader.py,sha256=RZCX1RJuQpTatJjBOU74_gJ5Ab7xwarRmFh5ON4n_G4,2279
 langroid/parsing/url_loader_cookies.py,sha256=Lg4sNpRz9MByWq2mde6T0hKv68VZSV3mtMjNEHuFeSU,2327
 langroid/parsing/urls.py,sha256=Nv4yCWQLLBEjaiRdaZZVQNBEl_cfK_V6cVuPm91wGtU,7686
@@ -86,10 +86,10 @@ langroid/utils/docker.py,sha256=kJQOLTgM0x9j9pgIIqp0dZNZCTvoUDhp6i8tYBq1Jr0,1105
 langroid/utils/globals.py,sha256=VkTHhlqSz86oOPq65sjul0XU8I52UNaFC5vwybMQ74w,1343
 langroid/utils/llms/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 langroid/utils/llms/strings.py,sha256=CSAX9Z6FQOLXOzbLMe_Opqtc3ruDAKTTk7cPqc6Blh0,263
-langroid/utils/logging.py,sha256=xXpohbvK74_reomdkIWTeyDjGG8GT1fuU7zcLL3Ngt8,3951
+langroid/utils/logging.py,sha256=R8TN-FqVpwZ4Ajgls9TDMthLvPpQd0QVNXK-PJDj1Z8,3917
 langroid/utils/output/__init__.py,sha256=IpfqnCkfXa4HaOx39EMUhXuA7GPZFd7N_QMm1n43C_I,174
 langroid/utils/output/printing.py,sha256=5EsYB1O4qKhocW19aebOUzK82RD9U5nygbY21yo8gfg,2872
-langroid/utils/pandas_utils.py,sha256=XcOgM6Ur3W6kycuyZQcxkVIDTZcOXAiYJ4iwlK_8es4,678
+langroid/utils/pandas_utils.py,sha256=nSA1tIgOUTkRDn-IKq7HP8XGJcL6bA110LcPfRF7h8I,707
 langroid/utils/pydantic_utils.py,sha256=E6miRLZRU61nkGL1Y55eXE79auCdXMji5QRsPfvwEt8,16554
 langroid/utils/system.py,sha256=x9204H1-6EubDe8-9yX87KZSmRCkf0puXQv91QOetF4,3326
 langroid/utils/web/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -103,7 +103,7 @@ langroid/vector_store/meilisearch.py,sha256=d2huA9P-NoYRuAQ9ZeXJmMKr7ry8u90RUSR2
 langroid/vector_store/momento.py,sha256=j6Eo6oIDN2fe7lsBOlCXJn3uvvERHHTFL5QJfeREeOM,10044
 langroid/vector_store/qdrant_cloud.py,sha256=3im4Mip0QXLkR6wiqVsjV1QvhSElfxdFSuDKddBDQ-4,188
 langroid/vector_store/qdrantdb.py,sha256=qt7Dye6rcgoe0551WzmOxRGIlJfL87D4MX7HdqxuEok,13393
-langroid-0.1.157.dist-info/LICENSE,sha256=EgVbvA6VSYgUlvC3RvPKehSg7MFaxWDsFuzLOsPPfJg,1065
-langroid-0.1.157.dist-info/METADATA,sha256=jETw_wkd-JN8NXlZzvQdPp92orlrTo7eAHopHUUUZ94,42442
-langroid-0.1.157.dist-info/WHEEL,sha256=FMvqSimYX_P7y0a7UY-_Mc83r5zkBZsCYPm7Lr0Bsq4,88
-langroid-0.1.157.dist-info/RECORD,,
+langroid-0.1.159.dist-info/LICENSE,sha256=EgVbvA6VSYgUlvC3RvPKehSg7MFaxWDsFuzLOsPPfJg,1065
+langroid-0.1.159.dist-info/METADATA,sha256=-qIZRl58PZFOsLnKyeFFktfjVedKKQufuogeH6TN2qw,42701
+langroid-0.1.159.dist-info/WHEEL,sha256=FMvqSimYX_P7y0a7UY-_Mc83r5zkBZsCYPm7Lr0Bsq4,88
+langroid-0.1.159.dist-info/RECORD,,