cwyodmodules 0.3.44__py3-none-any.whl → 0.3.46__py3-none-any.whl

cwyodmodules/graphrag/query/generate.py

@@ -1,92 +1,107 @@
-from graphrag.llm.llm import generate_response
-from graphrag.query.graph_search import local_query_graph, global_query_graph
-from graphrag.query.vector_search import _similarity_search
-
-from graphrag.database.base import get_db
-from graphrag.database.models import Chunk
-
-from typing import List, Tuple, Any, Dict, Set
-
-import asyncio
-
-
-async def _local_query(
-    query: str, top_k: int, max_nodes: int=3, order_range: int=5
-) -> Tuple[str | None, List[str], Dict[str, Dict[str, Any]], List[str]]:
-    
-    db = next(get_db())
-    nodes, keywords = await local_query_graph(query=query, top_k=top_k, order_range=order_range)
-    chunk_texts: List[str] = []
-    for chunk_id in nodes:
-        if len(chunk_texts) >= max_nodes:
-            break
-        chunk_texts.append(
-            db.get(Chunk, chunk_id).text
-        )
-        
-    context = "\n".join(chunk_texts)
-    response = await generate_response(context=context, query=query)
-    db.close()
-    return response, chunk_texts, nodes, keywords
-
-
-async def _global_query(
-    query: str, top_k: int, max_nodes: int=3, alpha: float=0.7
-) -> Tuple[str | None, List[str], Dict[str, Dict[str, Any]], List[str]]:
-    
-    db = next(get_db())
-    chunks, keywords = await global_query_graph(query=query, top_k=top_k, alpha=alpha, edge_nodes=max_nodes)
-    chunk_texts: List[str] = []
-    for chunk_id in chunks:
-        chunk_texts.append(
-            db.get(Chunk, chunk_id).text
-        )
-
-    context = "\n".join(chunk_texts)
-    response = await generate_response(context=context, query=query)
-    db.close()
-    return response, chunk_texts, chunks, keywords
-
-
-async def _hybrid_query(query: str, top_k: int, max_nodes: int=3, alpha: float=0.7, order_range: int=5
-) -> Tuple[str | None, List[str], Tuple[Dict[str, Any], Dict[str, Dict[str, Any]]], Set[str]]:
-    db = next(get_db())
-    local_query_task = asyncio.create_task(local_query_graph(query=query, top_k=top_k, order_range=order_range))
-    global_query_task = asyncio.create_task(global_query_graph(query=query, top_k=top_k, alpha=alpha, edge_nodes=max_nodes))
-    
-    (local_nodes, local_keywords), (global_chunks, global_keywords) = await asyncio.gather(*[local_query_task, global_query_task])
-    chunk_texts: List[str] = []
-    
-    seen_ids = set()
-    for chunk_id in local_nodes:
-        if len(chunk_texts) >= max_nodes:
-            break
-        seen_ids.add(chunk_id)
-        chunk_texts.append(
-            db.get(Chunk, chunk_id).text
-        )
-        
-    for chunk_id in global_chunks:
-        if chunk_id in seen_ids: continue
-        chunk_texts.append(
-            db.get(Chunk, chunk_id).text
-        )
-        
-    context = "\n".join(chunk_texts)
-    response = await generate_response(context=context, query=query)
-    db.close()
-    return response, chunk_texts, (global_chunks, local_nodes), set(local_keywords + global_keywords)
-
-    
-async def _naive_rag(query: str, top_k: int) -> Tuple[str, List[str]]:
-    db = next(get_db())
-    nodes = await _similarity_search(text=query, table='chunk', top_k=top_k)
-    chunk_texts: List[str] = []
-    for (node, _) in nodes:
-        chunk_texts.append(
-            node.text
-        )
-    context = "\n".join(chunk_texts)
-    response = await generate_response(context=context, query=query)
-    db.close()
+from ..llm.llm import generate_response
+from .graph_search import local_query_graph, global_query_graph
+from .vector_search import _similarity_search
+from ..database.base import get_db, for_notebook
+from ..database.models import Chunk
+from .utils import (
+    separate_by_rank,
+    get_nodes,
+    get_text_from_nodes,
+    get_all_connected_nodes,
+    get_final_text,
+    get_context_from_db,
+    process_chunks,
+    get_node_by_id,
+    get_node_from_chunk,
+)
+from ..llm.query_generation import (
+    generate_query_from_question_prompt,
+    generate_response_from_context_prompt,
+)
+import asyncio
+from typing import List, Dict, Tuple, Any, Set
+from collections import defaultdict
+import networkx as nx
+import pandas as pd
+
+
+async def _local_query(
+    query: str, top_k: int, max_nodes: int=3, order_range: int=5
+) -> Tuple[str | None, List[str], Dict[str, Dict[str, Any]], List[str]]:
+    
+    db = next(get_db())
+    nodes, keywords = await local_query_graph(query=query, top_k=top_k, order_range=order_range)
+    chunk_texts: List[str] = []
+    for chunk_id in nodes:
+        if len(chunk_texts) >= max_nodes:
+            break
+        chunk_texts.append(
+            db.get(Chunk, chunk_id).text
+        )
+        
+    context = "\n".join(chunk_texts)
+    response = await generate_response(context=context, query=query)
+    db.close()
+    return response, chunk_texts, nodes, keywords
+
+
+async def _global_query(
+    query: str, top_k: int, max_nodes: int=3, alpha: float=0.7
+) -> Tuple[str | None, List[str], Dict[str, Dict[str, Any]], List[str]]:
+    
+    db = next(get_db())
+    chunks, keywords = await global_query_graph(query=query, top_k=top_k, alpha=alpha, edge_nodes=max_nodes)
+    chunk_texts: List[str] = []
+    for chunk_id in chunks:
+        chunk_texts.append(
+            db.get(Chunk, chunk_id).text
+        )
+
+    context = "\n".join(chunk_texts)
+    response = await generate_response(context=context, query=query)
+    db.close()
+    return response, chunk_texts, chunks, keywords
+
+
+async def _hybrid_query(query: str, top_k: int, max_nodes: int=3, alpha: float=0.7, order_range: int=5
+) -> Tuple[str | None, List[str], Tuple[Dict[str, Any], Dict[str, Dict[str, Any]]], Set[str]]:
+    db = next(get_db())
+    local_query_task = asyncio.create_task(local_query_graph(query=query, top_k=top_k, order_range=order_range))
+    global_query_task = asyncio.create_task(global_query_graph(query=query, top_k=top_k, alpha=alpha, edge_nodes=max_nodes))
+    
+    (local_nodes, local_keywords), (global_chunks, global_keywords) = await asyncio.gather(*[local_query_task, global_query_task])
+    chunk_texts: List[str] = []
+    
+    seen_ids = set()
+    for chunk_id in local_nodes:
+        if len(chunk_texts) >= max_nodes:
+            break
+        seen_ids.add(chunk_id)
+        chunk_texts.append(
+            db.get(Chunk, chunk_id).text
+        )
+        
+    for chunk_id in global_chunks:
+        if chunk_id in seen_ids: continue
+        chunk_texts.append(
+            db.get(Chunk, chunk_id).text
+        )
+        
+    context = "\n".join(chunk_texts)
+    response = await generate_response(context=context, query=query)
+    db.close()
+    return response, chunk_texts, (global_chunks, local_nodes), set(local_keywords + global_keywords)
+
+    
+async def _naive_rag(query: str, top_k: int) -> Tuple[str, List[str]]:
+    db = next(get_db())
+    nodes = await _similarity_search(text=query, table='chunk', top_k=top_k)
+    chunk_texts: List[str] = []
+    for (node, _) in nodes:
+        chunk_texts.append(
+            node.text
+        )
+    context = "\n".join(chunk_texts)
+    response = await generate_response(context=context, query=query)
+    db.close()
     return response, chunk_texts

{cwyodmodules-0.3.44.dist-info → cwyodmodules-0.3.46.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: cwyodmodules
-Version: 0.3.44
+Version: 0.3.46
 Summary: Add your description here
 Author-email: Patrik <patrikhartl@gmail.com>
 Classifier: Operating System :: OS Independent

{cwyodmodules-0.3.44.dist-info → cwyodmodules-0.3.46.dist-info}/RECORD

@@ -1,3 +1,4 @@
+cwyodmodules/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 cwyodmodules/mgmt_config.py,sha256=xRp9absSmQifcKKCn6fyqcMSJ9hDvMTqPPAViezJHZw,5308
 cwyodmodules/api/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 cwyodmodules/api/chat_history.py,sha256=bVXFmhTHIfEiHv_nBrfizO-cQRHhKgrdcZ07OD1b0Tw,20683
@@ -40,13 +41,13 @@ cwyodmodules/batch/utilities/helpers/env_helper.py,sha256=S0EG0XzILrJ_iPCAwKN-kz
 cwyodmodules/batch/utilities/helpers/lightrag_helper.py,sha256=7lb9JMm5IohsO73LWo5bWmlzWCGYNsx_fYl-aFdwATQ,3845
 cwyodmodules/batch/utilities/helpers/llm_helper.py,sha256=lHLYrUidtaemmKrVbWoo7oIvwluUoPUk16U5lV-YIX8,8282
 cwyodmodules/batch/utilities/helpers/orchestrator_helper.py,sha256=mCcZyMFG0otnw9gzWd-PYocHmDdFDVg-RT9oDPiDZPk,897
-cwyodmodules/batch/utilities/helpers/secret_helper.py,sha256=RYDPR-1c7uChfmAJfzvN_GiI4u6czyWZ67XQah0hla4,2756
+cwyodmodules/batch/utilities/helpers/secret_helper.py,sha256=KacIc2hHpWtgButUnR3sFC9StLS36HjYvNpTmf2f4gc,2834
 cwyodmodules/batch/utilities/helpers/config/agent_mode.py,sha256=8XMbu8dwMXva_xxeZNDlwOjDaZwIcwc-xJK1-QsaJ3w,82
 cwyodmodules/batch/utilities/helpers/config/assistant_strategy.py,sha256=uT8h646zEURU9x8oDOH7pWoZKb0Mw6dA2nJtA2M-ufg,171
-cwyodmodules/batch/utilities/helpers/config/config_helper.py,sha256=GJBmSrnSnKhJQHA9Brj3CIxuPdJQ__XNjQvWQa4iLkY,15323
+cwyodmodules/batch/utilities/helpers/config/config_helper.py,sha256=26na6YrLqRLhdSZxTSlOnJOkIcPbTcbFVuPEQTPT3WY,14908
 cwyodmodules/batch/utilities/helpers/config/conversation_flow.py,sha256=4nP8a-I-sME5-2unzWWBNpTzWdfpfc5_EAYU6Pn6LAQ,94
 cwyodmodules/batch/utilities/helpers/config/database_type.py,sha256=Zmmlh1NAKDdd-2ei478boncRKcx8v3lDkPf4kO2j4ss,132
-cwyodmodules/batch/utilities/helpers/config/default.json,sha256=LIQNj5TpzmN6YPNZOmwngO4p9us3GDc2N18w-KQ4mw8,15282
+cwyodmodules/batch/utilities/helpers/config/default.json,sha256=865avujlLpO2to1dlh7yXfs0F57noJNrld2zwVYirwM,15406
 cwyodmodules/batch/utilities/helpers/config/default_contract_assistant_prompt.txt,sha256=X39WGcxzQPIvqG7NpAMPsgmSwSyMEoK1DVWiuEHEHRg,3210
 cwyodmodules/batch/utilities/helpers/config/default_employee_assistant_prompt.txt,sha256=toQFo0wXYrEK7zAItAS9rbtyhT6DJZKBhiL6C9VPUQk,3942
 cwyodmodules/batch/utilities/helpers/config/embedding_config.py,sha256=9pCJxpsouln9dngjVHaKGFYP14PrwmSts_UFDytSiVk,950
@@ -91,13 +92,13 @@ cwyodmodules/batch/utilities/tools/question_answer_tool.py,sha256=8GWurj50bUND_4
 cwyodmodules/batch/utilities/tools/text_processing_tool.py,sha256=zCF83SQaC8k6z6-Gevxj6AdDKzhU3P_6FG0hAnrG5Zk,1824
 cwyodmodules/graphrag/__init__.py,sha256=O5fi4Q3RC9Gt8ItNZGV1HjyIJvEz7grYXTFMmUWstxw,111
 cwyodmodules/graphrag/config.py,sha256=TvG45dezHkqgZzWnPxhpyWciIiKOrD_3JQ3APkoD6Hw,1525
-cwyodmodules/graphrag/main.py,sha256=8_z5X3b00J1FYnq43LrbJpOC3bKGkuyZQglgtBtVO2I,2201
+cwyodmodules/graphrag/main.py,sha256=Ot4hu0muElQGge9L9-8lZSTswgG8YPRoQjXE6wnZ2Gw,2211
 cwyodmodules/graphrag/database/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 cwyodmodules/graphrag/database/base.py,sha256=d8CCJYAyEqu3F9m4qfvszMKJGOP0tD0FpTEUbjDDHqU,751
 cwyodmodules/graphrag/database/models.py,sha256=4NZt2ZSeyaR1bkrUZQHYWPxWHK49-XNV6Wt3n3zVTcg,2803
 cwyodmodules/graphrag/indexing/__init__.py,sha256=TdI1oChZYljm_ao4GPQ3Oe8ESeeDKGsyheqoHFf5sJQ,125
 cwyodmodules/graphrag/indexing/chunking.py,sha256=hSEH6KoHemQg1tB4-_BN4r7ZMLBQ94ctV9EUwgKiQcc,7216
-cwyodmodules/graphrag/indexing/extraction.py,sha256=Pe1gukVxpq7lmV62oiIMVc8TTn4IGNQtI5MFRifbhe4,10377
+cwyodmodules/graphrag/indexing/extraction.py,sha256=LwV29UVyhdAIkcd30GSQrfx-ilxsg4tHHCsPmNqrA9M,10822
 cwyodmodules/graphrag/indexing/types.py,sha256=wIn-q-eGVdmcNvDt0gGXIdqJcvwK5exRqtuWtLu2z0Y,2457
 cwyodmodules/graphrag/indexing/upsert.py,sha256=BhinBPaq2z0ApHWSG0XzmqXFsrXuiCJcQyL1hhnM1C0,4814
 cwyodmodules/graphrag/indexing/utils.py,sha256=kvjpBSyRsJcdCybz1Eqd2itFEe4cm76AxqJ_F7n3JEs,169
@@ -105,12 +106,12 @@ cwyodmodules/graphrag/llm/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJW
 cwyodmodules/graphrag/llm/llm.py,sha256=sxWqiiMG-iuVvNPYnlqC9mDdld2JP8iPGhzaBq-zk6M,4850
 cwyodmodules/graphrag/llm/prompt.py,sha256=h4F4j-i6MxpeSz1Kiyhr_hMdiIQOcNosR_HR-zmSoMU,6674
 cwyodmodules/graphrag/query/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-cwyodmodules/graphrag/query/generate.py,sha256=xBnZs2U9xFWtPk4AfAZgYKbHdcxNcIO6Q_D5dWJ_8qg,3323
+cwyodmodules/graphrag/query/generate.py,sha256=BZiB6iw7PkIovw-CyYFogMHnDxK0Qu_4uCe494nYEFY,3816
 cwyodmodules/graphrag/query/graph_search.py,sha256=95h3ecSWx4864XgKABtG0fh3Nk8HkqJVzoCrO8daJ-Y,7724
 cwyodmodules/graphrag/query/types.py,sha256=1Iq1dp4I4a56_cuFjOZ0NTgd0A2_MpVFznp_czgt6cI,617
 cwyodmodules/graphrag/query/vector_search.py,sha256=9Gwu9LPjtoAYUU8WKqCvbCHAIg3dpk71reoYd1scLnQ,1807
-cwyodmodules-0.3.44.dist-info/licenses/LICENSE,sha256=UqBDTipijsSW2ZSOXyTZnMsXmLoEHTgNEM0tL4g-Sso,1150
-cwyodmodules-0.3.44.dist-info/METADATA,sha256=oGvSCI0ocoOWPVt2ADEKOOIa_JTsM_5v6ekG_TNuBRo,2002
-cwyodmodules-0.3.44.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-cwyodmodules-0.3.44.dist-info/top_level.txt,sha256=99RENLbkdRX-qpJvsxZ5AfmTL5s6shSaKOWYpz1vwzg,13
-cwyodmodules-0.3.44.dist-info/RECORD,,
+cwyodmodules-0.3.46.dist-info/licenses/LICENSE,sha256=UqBDTipijsSW2ZSOXyTZnMsXmLoEHTgNEM0tL4g-Sso,1150
+cwyodmodules-0.3.46.dist-info/METADATA,sha256=RAKWb18OzQioSxPjwq-xR7ZX9NBrRL4jELP1p5gUmMM,2002
+cwyodmodules-0.3.46.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+cwyodmodules-0.3.46.dist-info/top_level.txt,sha256=99RENLbkdRX-qpJvsxZ5AfmTL5s6shSaKOWYpz1vwzg,13
+cwyodmodules-0.3.46.dist-info/RECORD,,