auto-coder 0.1.280__py3-none-any.whl → 0.1.282__py3-none-any.whl

{auto_coder-0.1.280.dist-info → auto_coder-0.1.282.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: auto-coder
-Version: 0.1.280
+Version: 0.1.282
 Summary: AutoCoder: AutoCoder
 Author: allwefantasy
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
@@ -26,7 +26,7 @@ Requires-Dist: tabulate
 Requires-Dist: jupyter-client
 Requires-Dist: prompt-toolkit
 Requires-Dist: tokenizers
-Requires-Dist: byzerllm[saas] >=0.1.170
+Requires-Dist: byzerllm[saas] >=0.1.171
 Requires-Dist: patch
 Requires-Dist: diff-match-patch
 Requires-Dist: GitPython

{auto_coder-0.1.280.dist-info → auto_coder-0.1.282.dist-info}/RECORD

@@ -1,7 +1,7 @@
 autocoder/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-autocoder/auto_coder.py,sha256=SgAAoSpUaPZiiNnEWRYIULEsKBJaKuQK_b7pEl_mKY8,65516
+autocoder/auto_coder.py,sha256=cxH77xjLPwCmE6R-o1KpONOZlVNOvbTYV4ilAjL8w8A,65592
 autocoder/auto_coder_lang.py,sha256=Rtupq6N3_HT7JRhDKdgCBcwRaiAnyCOR_Gsp4jUomrI,3229
-autocoder/auto_coder_rag.py,sha256=mX-szIG9T7Mzwoc4QwKp_GyYBcVf6dfsNJnKzYHHl6U,30329
+autocoder/auto_coder_rag.py,sha256=cwsCpudP6d2CHOUTD-RXw7CCgx-c7sG7_cKqnjzphlk,32973
 autocoder/auto_coder_rag_client_mcp.py,sha256=QRxUbjc6A8UmDMQ8lXgZkjgqtq3lgKYeatJbDY6rSo0,6270
 autocoder/auto_coder_rag_mcp.py,sha256=-RrjNwFaS2e5v8XDIrKR-zlUNUE8UBaeOtojffBrvJo,8521
 autocoder/auto_coder_runner.py,sha256=w-4MCKhOFaoABcDfVoZoonF59UyRso3kghimQYLz3NA,100851
@@ -12,7 +12,7 @@ autocoder/chat_auto_coder_lang.py,sha256=ShOQVOnMA-WlT-fB9OrOer-xQkbcWxJGl-WMPuZ
 autocoder/command_args.py,sha256=9aYJ-AmPxP1sQh6ciw04FWHjSn31f2W9afXFwo8wgx4,30441
 autocoder/lang.py,sha256=U6AjVV8Rs1uLyjFCZ8sT6WWuNUxMBqkXXIOs4S120uk,14511
 autocoder/models.py,sha256=PlG1tKHSHwB57cKLOl5gTl5yTzFUDzCgeHPJU3N9F6Q,9106
-autocoder/version.py,sha256=mNnPow60dgdANkDcEoYTXr9_lpoMQZSEy1-LRu7QFHs,23
+autocoder/version.py,sha256=MuZdShxBktD9b-QDa1Sdv3QS_FjwPsAMhpVaUVk1n9A,23
 autocoder/agent/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autocoder/agent/auto_demand_organizer.py,sha256=NWSAEsEk94vT3lGjfo25kKLMwYdPcpy9e-i21txPasQ,6942
 autocoder/agent/auto_filegroup.py,sha256=CW7bqp0FW1GIEMnl-blyAc2UGT7O9Mom0q66ITz1ckM,6635
@@ -55,7 +55,7 @@ autocoder/common/command_templates.py,sha256=WAixVjue5QmCFAD13K4ElfcOEjdeGr8tFb0
 autocoder/common/conf_import_export.py,sha256=w__WsIobe6nmsGns2pV-laU7R5ZvtQNuIbXebxhbY7A,3967
 autocoder/common/conf_validator.py,sha256=EzSmadpZ22D9e8iWmfeWodUeYJt0IgMoaAOmCleXliI,8795
 autocoder/common/const.py,sha256=eTjhjh4Aj4CUzviJ81jaf3Y5cwqsLATySn2wJxaS6RQ,2911
-autocoder/common/context_pruner.py,sha256=_JUpN8naoGNF2JupirM-5xn8Bl9uQgIVmZP3dXH4F1g,21725
+autocoder/common/context_pruner.py,sha256=HlU5BmxpCX7uVTJUsTFLlXvkwcOQuidI9uCKZaFxh6s,19874
 autocoder/common/conversation_pruner.py,sha256=pzmrQEa7pFzA66eYSS_h7VqP6ZwUABeooDQzm0PGu0A,5770
 autocoder/common/files.py,sha256=nPiKcnUcYZbSUn3TskKeTVnAxCJRtuehPuB_5d2imX8,4618
 autocoder/common/git_utils.py,sha256=EK8gekbXsG6BNDVrd1Nsan_7kJ71dd8_w9FiOFxjsVI,26276
@@ -95,7 +95,7 @@ autocoder/dispacher/actions/plugins/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQ
 autocoder/dispacher/actions/plugins/action_regex_project.py,sha256=AqGIkjbqV1eOS3vBoZUTSOpyOlkv1p5h35mI2Kcvekw,6906
 autocoder/dispacher/actions/plugins/action_translate.py,sha256=GEn7dZA22jy5WyzINomjmzzB795p2Olg-CJla97lRF8,7744
 autocoder/index/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-autocoder/index/entry.py,sha256=l-RVlKSLDZkWaWi3jDzMbyWpZ97ZHIEZufsF5gae4Lo,14430
+autocoder/index/entry.py,sha256=YwwKLxgre2xIbes7f_S0Dh7mAN7RoUSph9ORCmtTun8,14395
 autocoder/index/for_command.py,sha256=BFvljE4t6VaMBGboZAuhUCzVK0EitCy_n5D_7FEnihw,3204
 autocoder/index/index.py,sha256=3-SHlmeQMv6SFxNj7vVcNRDAYj9ZshuJJ1zXwBi0cDc,30873
 autocoder/index/symbols_utils.py,sha256=_EP7E_qWXxluAxq3FGZLlLfdrfwx3FmxCdulI8VGuac,2244
@@ -109,10 +109,10 @@ autocoder/pyproject/__init__.py,sha256=ms-A_pocgGv0oZPEW8JAdXi7G-VSVhkQ6CnWFe535
 autocoder/rag/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autocoder/rag/api_server.py,sha256=xiypCkdbclY0Z3Cmq5FTvtKrfQUV7yKcDaFFUttA2n0,7242
 autocoder/rag/doc_filter.py,sha256=UduVO2mlrngwJICrefjDJTYfdmQ4GcRXrfWDQ7xXksk,14206
-autocoder/rag/document_retriever.py,sha256=5oThtxukGuRFF96o3pHKsk306a8diXbhgSrbqyU2BvM,8894
+autocoder/rag/document_retriever.py,sha256=MGn6oIPo49BbRC99xmLMFkZrpHfcDfKoGYqWxXF554U,8051
 autocoder/rag/lang.py,sha256=TVNx5m7OtBcdfahzI29tMj9m1yrEm32G1c1zc4ZNIPs,3130
 autocoder/rag/llm_wrapper.py,sha256=Ht5GF5yJtrztoliujsZzx_ooWZmHkd5xLZKcGEiicZw,4303
-autocoder/rag/long_context_rag.py,sha256=3CAlf7GM-LgewS5j9XGKvsKSO4MM6M8TTkKxAGzqVY0,39308
+autocoder/rag/long_context_rag.py,sha256=ZvTT3yO5FmJwWXHqqFrQgkb9YrVajrXrtB_tcFdDwAs,40172
 autocoder/rag/rag_config.py,sha256=8LwFcTd8OJWWwi1_WY4IzjqgtT6RyE2j4PjxS5cCTDE,802
 autocoder/rag/rag_entry.py,sha256=6TKtErZ0Us9XSV6HgRKXA6yR3SiZGPHpynOKSaR1wgE,2463
 autocoder/rag/raw_rag.py,sha256=BOr0YGf3umjqXOIDVO1LXQ0bIHx8hzBdiubND2ezyxc,2946
@@ -130,6 +130,7 @@ autocoder/rag/cache/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSu
 autocoder/rag/cache/base_cache.py,sha256=EaYYYbclMBvnlOUoM7qonnluwZX5oSvUjdvGvFun8_8,742
 autocoder/rag/cache/byzer_storage_cache.py,sha256=gK90pf741CgccCzQ73urBorCqVyAfwU1FAqMtSorWVk,17232
 autocoder/rag/cache/file_monitor_cache.py,sha256=2TnOW8Y81Zc0WA1upRrkmQH18IMdv40CeNccmnTvd3c,4981
+autocoder/rag/cache/local_byzer_storage_cache.py,sha256=Uhmu5JK0tfZ8NvlcjJzcwtQRhZDpbGp_U6qLXZxVwss,17495
 autocoder/rag/cache/simple_cache.py,sha256=8FMmBAfhAPcdSNUWC6Ga43LBFGXD-klwabVbzm_bciI,9347
 autocoder/rag/loaders/__init__.py,sha256=EQHEZ5Cmz-mGP2SllUTvcIbYCnF7W149dNpNItfs0yE,304
 autocoder/rag/loaders/docx_loader.py,sha256=ZswPqiiLngUEpzLhNNm1nmwEYV7ZHFEfIoXoG7c5GDU,614
@@ -167,9 +168,9 @@ autocoder/utils/types.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autocoder/utils/auto_coder_utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autocoder/utils/auto_coder_utils/chat_stream_out.py,sha256=lkJ_A-sYU36JMzjFWkk3pR6uos8oZHYt9GPsPe_CPAo,11766
 autocoder/utils/chat_auto_coder_utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-auto_coder-0.1.280.dist-info/LICENSE,sha256=HrhfyXIkWY2tGFK11kg7vPCqhgh5DcxleloqdhrpyMY,11558
-auto_coder-0.1.280.dist-info/METADATA,sha256=SDBMvUk6v6YP7RSwlAWHFGfa3LTOUj3fky1Yz0hlFB0,2643
-auto_coder-0.1.280.dist-info/WHEEL,sha256=GV9aMThwP_4oNCtvEC2ec3qUYutgWeAzklro_0m4WJQ,91
-auto_coder-0.1.280.dist-info/entry_points.txt,sha256=0nzHtHH4pNcM7xq4EBA2toS28Qelrvcbrr59GqD_0Ak,350
-auto_coder-0.1.280.dist-info/top_level.txt,sha256=Jqc0_uJSw2GwoFQAa9iJxYns-2mWla-9ok_Y3Gcznjk,10
-auto_coder-0.1.280.dist-info/RECORD,,
+auto_coder-0.1.282.dist-info/LICENSE,sha256=HrhfyXIkWY2tGFK11kg7vPCqhgh5DcxleloqdhrpyMY,11558
+auto_coder-0.1.282.dist-info/METADATA,sha256=L2wd7XGt-KURDWFp-mn4HAo7K87iqwuPXKfepSBC3JA,2643
+auto_coder-0.1.282.dist-info/WHEEL,sha256=GV9aMThwP_4oNCtvEC2ec3qUYutgWeAzklro_0m4WJQ,91
+auto_coder-0.1.282.dist-info/entry_points.txt,sha256=0nzHtHH4pNcM7xq4EBA2toS28Qelrvcbrr59GqD_0Ak,350
+auto_coder-0.1.282.dist-info/top_level.txt,sha256=Jqc0_uJSw2GwoFQAa9iJxYns-2mWla-9ok_Y3Gcznjk,10
+auto_coder-0.1.282.dist-info/RECORD,,

autocoder/auto_coder.py

@@ -1345,7 +1345,8 @@ def main(input_args: Optional[List[str]] = None):
                 v = learner.learn_from_commit(query=args.query,conversations=loaded_conversations)
             else:                
                 # 预估token数量
-                estimated_input_tokens = count_tokens(json.dumps(loaded_conversations, ensure_ascii=False))
+                dumped_conversations = json.dumps(loaded_conversations, ensure_ascii=False)                
+                estimated_input_tokens = count_tokens(dumped_conversations)
                 printer = Printer()
                 printer.print_in_terminal("estimated_chat_input_tokens", style="yellow",
                                   estimated_input_tokens=estimated_input_tokens

autocoder/auto_coder_rag.py

@@ -22,7 +22,7 @@ from loguru import logger
 import asyncio
 from datetime import datetime
 
-from autocoder.rag.document_retriever import process_file_local
+from autocoder.rag.utils import process_file_local
 import pkg_resources
 from autocoder.rag.token_counter import TokenCounter
 from autocoder.rag.types import RAGServiceInfo
@@ -186,6 +186,11 @@ def main(input_args: Optional[List[str]] = None):
     build_index_parser.add_argument(
         "--model", default="v3_chat", help=desc["model"]
     )
+    
+    build_index_parser.add_argument(
+        "--on_ray", action="store_true", help="Run on Ray"
+    )
+
     build_index_parser.add_argument(
         "--index_model", default="", help=desc["index_model"]
     )
@@ -216,8 +221,7 @@ def main(input_args: Optional[List[str]] = None):
     )
     serve_parser.add_argument("--file", default="", help=desc["file"])
     serve_parser.add_argument("--model", default="v3_chat", help=desc["model"])
-    serve_parser.add_argument("--index_model", default="", help=desc["index_model"])
-    serve_parser.add_argument("--emb_model", default="", help=desc["emb_model"])
+    serve_parser.add_argument("--index_model", default="", help=desc["index_model"])    
     serve_parser.add_argument("--ray_address", default="auto", help=desc["ray_address"])
     serve_parser.add_argument(
         "--index_filter_workers",
@@ -377,6 +381,12 @@ def main(input_args: Optional[List[str]] = None):
         help="The model used for question answering",
     )
 
+    serve_parser.add_argument(
+        "--emb_model",
+        default="",
+        help="The model used for embedding documents",
+    )
+
     # Benchmark command
     benchmark_parser = subparsers.add_parser(
         "benchmark", help="Benchmark LLM client performance"
@@ -510,10 +520,7 @@ def main(input_args: Optional[List[str]] = None):
         # Generate unique name for RAG build if doc_dir exists
         if server_args.doc_dir:
             auto_coder_args.rag_build_name = generate_unique_name_from_path(server_args.doc_dir)
-            logger.info(f"Generated RAG build name: {auto_coder_args.rag_build_name}")
-
-        if auto_coder_args.enable_hybrid_index and args.product_mode == "lite":
-            raise Exception("Hybrid index is not supported in lite mode")
+            logger.info(f"Generated RAG build name: {auto_coder_args.rag_build_name}")        
 
         if auto_coder_args.enable_hybrid_index and args.product_mode == "pro":
             # 尝试连接storage
@@ -555,16 +562,22 @@ def main(input_args: Optional[List[str]] = None):
                 qa_model.skip_nontext_check = True
                 llm.setup_sub_client("qa_model", qa_model)
 
+            if args.emb_model:
+                emb_model = byzerllm.ByzerLLM()
+                emb_model.setup_default_model_name(args.emb_model)
+                emb_model.skip_nontext_check = True
+                llm.setup_sub_client("emb_model", emb_model)
+
             # 当启用hybrid_index时,检查必要的组件
             if auto_coder_args.enable_hybrid_index:
-                if not llm.is_model_exist("emb"):
+                if not args.emb_model and not llm.is_model_exist("emb"):
                     logger.error(
                         "When enable_hybrid_index is true, an 'emb' model must be deployed"
                     )
                     return
-                llm.setup_default_emb_model_name("emb")
+                llm.setup_default_emb_model_name(args.emb_model or "emb")
 
-        elif args.product_mode == "lite":
+        if args.product_mode == "lite":
             from autocoder import models as models_module
             model_info = models_module.get_model_by_name(args.model)
             llm = byzerllm.SimpleByzerLLM(default_model_name=args.model)
@@ -629,6 +642,26 @@ def main(input_args: Optional[List[str]] = None):
                 )
                 llm.setup_sub_client("qa_model", qa_model)
 
+            if args.emb_model:
+                model_info = models_module.get_model_by_name(args.emb_model)
+                emb_model = byzerllm.SimpleByzerLLM(default_model_name=args.emb_model)
+                emb_model.deploy(
+                    model_path="",
+                    pretrained_model_type=model_info["model_type"],
+                    udf_name=args.emb_model,
+                    infer_params={
+                        "saas.base_url": model_info["base_url"],
+                        "saas.api_key": model_info["api_key"],
+                        "saas.model": model_info["model_name"],
+                        "saas.is_reasoning": False
+                    }
+                )
+                llm.setup_sub_client("emb_model", emb_model)
+
+            if args.enable_hybrid_index:
+                if not args.emb_model:
+                    raise Exception("When enable_hybrid_index is true, an 'emb' model must be specified")                
+
         if server_args.doc_dir:
             auto_coder_args.rag_type = "simple"
             auto_coder_args.rag_build_name = generate_unique_name_from_path(server_args.doc_dir)
@@ -675,31 +708,62 @@ def main(input_args: Optional[List[str]] = None):
         auto_coder_args.enable_hybrid_index = True
         auto_coder_args.rag_type = "simple"
 
-        try:
-            from byzerllm.apps.byzer_storage.simple_api import ByzerStorage
-
-            storage = ByzerStorage("byzerai_store", "rag", "files")
-            storage.retrieval.cluster_info("byzerai_store")
-        except Exception as e:
-            logger.error(
-                "When enable_hybrid_index is true, ByzerStorage must be started"
-            )
-            logger.error("Please run 'byzerllm storage start' first")
-            return
+        if args.on_ray:
 
-        llm = byzerllm.ByzerLLM()
-        llm.setup_default_model_name(args.model)
+            try:
+                from byzerllm.apps.byzer_storage.simple_api import ByzerStorage
 
-        # 当启用hybrid_index时,检查必要的组件
-        if auto_coder_args.enable_hybrid_index:
-            if not llm.is_model_exist("emb"):
+                storage = ByzerStorage("byzerai_store", "rag", "files")
+                storage.retrieval.cluster_info("byzerai_store")
+            except Exception as e:
                 logger.error(
-                    "When enable_hybrid_index is true, an 'emb' model must be deployed"
+                    "When enable_hybrid_index is true, ByzerStorage must be started"
                 )
+                logger.error("Please run 'byzerllm storage start' first")
                 return
-            llm.setup_default_emb_model_name("emb")
 
-        auto_coder_args.rag_build_name = generate_unique_name_from_path(args.doc_dir)
+            llm = byzerllm.ByzerLLM()
+            llm.setup_default_model_name(args.model)
+
+            # 当启用hybrid_index时,检查必要的组件
+            if auto_coder_args.enable_hybrid_index:
+                if not llm.is_model_exist("emb"):
+                    logger.error(
+                        "When enable_hybrid_index is true, an 'emb' model must be deployed"
+                    )
+                    return
+                llm.setup_default_emb_model_name("emb")
+        else:
+            from autocoder import models as models_module
+            model_info = models_module.get_model_by_name(args.model)
+            llm = byzerllm.SimpleByzerLLM(default_model_name=args.model)
+            llm.deploy(
+                model_path="",
+                pretrained_model_type=model_info["model_type"],
+                udf_name=args.model,
+                infer_params={
+                    "saas.base_url": model_info["base_url"],
+                    "saas.api_key": model_info["api_key"],
+                    "saas.model": model_info["model_name"],
+                    "saas.is_reasoning": model_info["is_reasoning"]
+                }
+            )
+
+            model_info = models_module.get_model_by_name(args.emb_model)
+            emb_model = byzerllm.SimpleByzerLLM(default_model_name=args.emb_model)
+            emb_model.deploy(
+                model_path="",
+                pretrained_model_type=model_info["model_type"],
+                udf_name=args.emb_model,
+                infer_params={
+                    "saas.base_url": model_info["base_url"],
+                    "saas.api_key": model_info["api_key"],
+                    "saas.model": model_info["model_name"],
+                    "saas.is_reasoning": False
+                }
+            )
+            llm.setup_sub_client("emb_model", emb_model)
+        
         rag = RAGFactory.get_rag(
             llm=llm,
             args=auto_coder_args,