auto-coder 0.1.274__py3-none-any.whl → 0.1.276__py3-none-any.whl

{auto_coder-0.1.274.dist-info → auto_coder-0.1.276.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: auto-coder
-Version: 0.1.274
+Version: 0.1.276
 Summary: AutoCoder: AutoCoder
 Author: allwefantasy
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence

{auto_coder-0.1.274.dist-info → auto_coder-0.1.276.dist-info}/RECORD

@@ -1,10 +1,10 @@
 autocoder/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autocoder/auto_coder.py,sha256=SgAAoSpUaPZiiNnEWRYIULEsKBJaKuQK_b7pEl_mKY8,65516
 autocoder/auto_coder_lang.py,sha256=Rtupq6N3_HT7JRhDKdgCBcwRaiAnyCOR_Gsp4jUomrI,3229
-autocoder/auto_coder_rag.py,sha256=nwgsXO2-scssWStjX3S910tDp-OZXZRddSYrpyC4Nq0,29021
+autocoder/auto_coder_rag.py,sha256=mX-szIG9T7Mzwoc4QwKp_GyYBcVf6dfsNJnKzYHHl6U,30329
 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=1SGmVP_X10vNJujdaDc3QuiTeHa4wDIpLp41poUJW7w,100773
+autocoder/auto_coder_runner.py,sha256=1M74XhbmmJg0z5rC_TjjlPaqyEPfg2z6N7mLN7z8qFw,100773
 autocoder/auto_coder_server.py,sha256=6YQweNEKUrGAZ3yPvw8_qlNZJYLVSVUXGrn1K6udLts,20413
 autocoder/benchmark.py,sha256=Ypomkdzd1T3GE6dRICY3Hj547dZ6_inqJbBJIp5QMco,4423
 autocoder/chat_auto_coder.py,sha256=G7_CIgDOTdGGPzRQDo0hEOh5p8A36oJQaYJc514xBkk,16842
@@ -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=MbulDxnyvkGLh88mZIIUxOAp1Ni82jbdAr2EllSoNPw,23
+autocoder/version.py,sha256=P8t0JVsPFr7ktiosnhBO0A_KC2ivastmjzH6pWwESMU,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
@@ -26,11 +26,11 @@ autocoder/agent/planner.py,sha256=SZTSZHxHzDmuWZo3K5fs79RwvJLWurg-nbJRRNbX65o,91
 autocoder/agent/project_reader.py,sha256=tWLaPoLw1gI6kO_NzivQj28KbobU2ceOLuppHMbfGl8,18234
 autocoder/chat/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autocoder/commands/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-autocoder/commands/auto_command.py,sha256=ymAYnaxiKfYFegKxE5FhdD8XqXMMf4mbDHJsIOxPaUU,52396
+autocoder/commands/auto_command.py,sha256=Qv5hObiIioIOAVkucFB2na6aLUuxSIgLVO9PfnXwyIo,52496
 autocoder/commands/tools.py,sha256=lanjoBGR6H8HDJSY3KrM6ibrtHZbgKX6mKJHSSE66dg,20493
 autocoder/common/JupyterClient.py,sha256=O-wi6pXeAEYhAY24kDa0BINrLYvKS6rKyWe98pDClS0,2816
 autocoder/common/ShellClient.py,sha256=fM1q8t_XMSbLBl2zkCNC2J9xuyKN3eXzGm6hHhqL2WY,2286
-autocoder/common/__init__.py,sha256=d1AmAAYhm4b17dVhAJFwV3Vv2r1lUhMejpjr32Poyg8,12948
+autocoder/common/__init__.py,sha256=Z6gvzhzLWYnXGVCnek2UoWic5DRiqWGQh4AiGZL3XVQ,12989
 autocoder/common/anything2images.py,sha256=0ILBbWzY02M-CiWB-vzuomb_J1hVdxRcenAfIrAXq9M,25283
 autocoder/common/anything2img.py,sha256=4TREa-sOA-iargieUy7MpyCYVUE-9Mmq0wJtwomPqnE,7662
 autocoder/common/audio.py,sha256=Kn9nWKQddWnUrAz0a_ZUgjcu4VUU_IcZBigT7n3N3qc,7439
@@ -44,13 +44,13 @@ autocoder/common/code_auto_generate.py,sha256=i0f0BPTvt8QN5-arC2PqfMs4sukGJqFw58
 autocoder/common/code_auto_generate_diff.py,sha256=EYGHf24fVAk3M8kXllXukdUvzo2I2Fb8qNrTiU_Qtug,18759
 autocoder/common/code_auto_generate_editblock.py,sha256=1XmoLZl-XrrPOyHkSfqe0pOxi0Lp4d69oDQX6zzqs_c,20523
 autocoder/common/code_auto_generate_strict_diff.py,sha256=68fAks4S6Hz0wAeXrdI1PjdeCiy0gXRhgEeIlgxB04E,17539
-autocoder/common/code_auto_merge.py,sha256=cMEX44QT59iMPEo5B8OQKU9fAzn4iIahEPk2n3rNFvg,8643
-autocoder/common/code_auto_merge_diff.py,sha256=jl6rYvioAuN5CWaYRPY4uBFKsDB0_yW6g6KeVlIxPJA,17948
-autocoder/common/code_auto_merge_editblock.py,sha256=D8petP0X74lScsVktFXEQ4_9u5UDVBRDFF16nkjWGZg,20310
-autocoder/common/code_auto_merge_strict_diff.py,sha256=VxasaD8IcxOEEilXmxT0WeLLA16JDlvvI7cV5GyyHAI,12086
+autocoder/common/code_auto_merge.py,sha256=wFxv0iHciqTwzy8osGPmyF0LcdK7ysFwnzXnsWr-gwM,8642
+autocoder/common/code_auto_merge_diff.py,sha256=clVxvDlo-SN3QxARipUfOFiSEomTrC41ha7_AOmufI4,17948
+autocoder/common/code_auto_merge_editblock.py,sha256=JzVTAqvqxVMkXOOPY5Ti2Wc_nL4YABYi470sCHKTOgg,20335
+autocoder/common/code_auto_merge_strict_diff.py,sha256=oYVEs1dmObBqK_1PgFX0LeuVti7es2IZtD00iNOn91g,12116
 autocoder/common/code_modification_ranker.py,sha256=e1i8oNPN_PfG4O8HQnQWxh9tyOd-ur6jFTXfkHPVXYo,13152
 autocoder/common/command_completer.py,sha256=Nw_EFXLDuVXbOKGlmzTODNRJUV9mSUEWV73GUmO_WLw,35166
-autocoder/common/command_generator.py,sha256=kxdkRtnPl4O51ChUmvLTXJqJlto5G02eG6xyx8fcLdA,2669
+autocoder/common/command_generator.py,sha256=t1o1d7xEyfnPvY_MpG5usyx0eKW7WCUR0urKoqCex60,2761
 autocoder/common/command_templates.py,sha256=WAixVjue5QmCFAD13K4ElfcOEjdeGr8tFb0atDAbEoo,8658
 autocoder/common/conf_import_export.py,sha256=w__WsIobe6nmsGns2pV-laU7R5ZvtQNuIbXebxhbY7A,3967
 autocoder/common/conf_validator.py,sha256=EzSmadpZ22D9e8iWmfeWodUeYJt0IgMoaAOmCleXliI,8795
@@ -58,7 +58,7 @@ autocoder/common/const.py,sha256=eTjhjh4Aj4CUzviJ81jaf3Y5cwqsLATySn2wJxaS6RQ,291
 autocoder/common/context_pruner.py,sha256=_JUpN8naoGNF2JupirM-5xn8Bl9uQgIVmZP3dXH4F1g,21725
 autocoder/common/conversation_pruner.py,sha256=pzmrQEa7pFzA66eYSS_h7VqP6ZwUABeooDQzm0PGu0A,5770
 autocoder/common/files.py,sha256=nPiKcnUcYZbSUn3TskKeTVnAxCJRtuehPuB_5d2imX8,4618
-autocoder/common/git_utils.py,sha256=qeuF_IB3G3M72asHxWokROU3hINCuFA1nar-UtF9wIU,26022
+autocoder/common/git_utils.py,sha256=EK8gekbXsG6BNDVrd1Nsan_7kJ71dd8_w9FiOFxjsVI,26276
 autocoder/common/global_cancel.py,sha256=hT7J7J5ChThIhk2x11_v4v9ASIn4HtwyPD26t2s-fwc,418
 autocoder/common/image_to_page.py,sha256=yWiTJQ49Lm3j0FngiJhQ9u7qayqE_bOGb8Rk0TmSWx0,14123
 autocoder/common/index_import_export.py,sha256=h758AYY1df6JMTKUXYmMkSgxItfymDt82XT7O-ygEuw,4565
@@ -75,7 +75,7 @@ autocoder/common/result_manager.py,sha256=nBcFRj5reBC7vp13M91f4B8iPW8B8OehayHlUd
 autocoder/common/screenshots.py,sha256=_gA-z1HxGjPShBrtgkdideq58MG6rqFB2qMUJKjrycs,3769
 autocoder/common/search.py,sha256=245iPFgWhMldoUK3CqCP89ltaxZiNPK73evoG6Fp1h8,16518
 autocoder/common/search_replace.py,sha256=GphFkc57Hb673CAwmbiocqTbw8vrV7TrZxtOhD0332g,22147
-autocoder/common/shells.py,sha256=NOgsKbAJVBti_5lSp7kEKAzjH0AJWufWpgXQdJYQF7E,18312
+autocoder/common/shells.py,sha256=-5j45qb1SVmkZaORqDZ5EM2zJ16b5QGM1wHDfBfGejk,18944
 autocoder/common/stats_panel.py,sha256=wGl9O45pjVVDxhNumLv4_NfLYSlUP_18Tw4hcJSjw50,4596
 autocoder/common/sys_prompt.py,sha256=JlexfjZt554faqbgkCmzOJqYUzDHfbnxly5ugFfHfEE,26403
 autocoder/common/text.py,sha256=KGRQq314GHBmY4MWG8ossRoQi1_DTotvhxchpn78c-k,1003
@@ -111,7 +111,7 @@ autocoder/rag/api_server.py,sha256=dRbhAZVRAOlZ64Cnxf4_rKb4iJwHnrWS9Zr67IVORw0,7
 autocoder/rag/doc_filter.py,sha256=yEXaBw1XJH57Gtvk4-RFQtd5eawA6SBjzxeRZrIsQew,11623
 autocoder/rag/document_retriever.py,sha256=5oThtxukGuRFF96o3pHKsk306a8diXbhgSrbqyU2BvM,8894
 autocoder/rag/llm_wrapper.py,sha256=wf56ofQNOaBkLhnoxK9VoVnHWD0gsj0pP8mUBfS92RI,2737
-autocoder/rag/long_context_rag.py,sha256=CzPC-ct6PVIKBkHsKon4s92YXmi8jZOlGgcquOwWQlI,31802
+autocoder/rag/long_context_rag.py,sha256=qFlNmbgQnstCSCb0SxfkMEYtZRr8p6YEc6u0jpve4Q0,32002
 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
@@ -123,11 +123,11 @@ autocoder/rag/token_counter.py,sha256=C-Lwc4oIjJpZDEqp9WLHGOe6hb4yhrdJpMtkrtp_1q
 autocoder/rag/token_limiter.py,sha256=3VgJF4may3ESyATmBIiOe05oc3VsidJcJTJ5EhoSvH8,18854
 autocoder/rag/token_limiter_utils.py,sha256=FATNEXBnFJy8IK3PWNt1pspIv8wuTgy3F_ACNvqoc4I,404
 autocoder/rag/types.py,sha256=WPgLpUTwbk0BAikyDOc0NOEwV5k73myF38zWdOuYdC4,2499
-autocoder/rag/utils.py,sha256=x5L8gskxUUGkVEP5K-0C-iYntE84asuDXOCqdzdNUoA,4956
+autocoder/rag/utils.py,sha256=f21ybCAlYVgr3tJP9MkVoM9d82-uG5NHu2gsv2oaVBQ,4961
 autocoder/rag/variable_holder.py,sha256=PFvBjFcR7-fNDD4Vcsc8CpH2Te057vcpwJMxtrfUgKI,75
 autocoder/rag/cache/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-autocoder/rag/cache/base_cache.py,sha256=KxrPKP-PunL7reXh5txT97DYLzGKm9M5O4CnOLDyX-c,403
-autocoder/rag/cache/byzer_storage_cache.py,sha256=iYRLKIrTOCmoHI1eUqEfYn_LK1fbU3vF5oYTWIC_HmM,15443
+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/simple_cache.py,sha256=8FMmBAfhAPcdSNUWC6Ga43LBFGXD-klwabVbzm_bciI,9347
 autocoder/rag/loaders/__init__.py,sha256=EQHEZ5Cmz-mGP2SllUTvcIbYCnF7W149dNpNItfs0yE,304
@@ -165,9 +165,9 @@ 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-0.1.274.dist-info/LICENSE,sha256=HrhfyXIkWY2tGFK11kg7vPCqhgh5DcxleloqdhrpyMY,11558
-auto_coder-0.1.274.dist-info/METADATA,sha256=7P2eJSSQDVvZ-_9VRfwKjGFB3OgLVmbYbQrRibp7HvQ,2643
-auto_coder-0.1.274.dist-info/WHEEL,sha256=GV9aMThwP_4oNCtvEC2ec3qUYutgWeAzklro_0m4WJQ,91
-auto_coder-0.1.274.dist-info/entry_points.txt,sha256=0nzHtHH4pNcM7xq4EBA2toS28Qelrvcbrr59GqD_0Ak,350
-auto_coder-0.1.274.dist-info/top_level.txt,sha256=Jqc0_uJSw2GwoFQAa9iJxYns-2mWla-9ok_Y3Gcznjk,10
-auto_coder-0.1.274.dist-info/RECORD,,
+auto_coder-0.1.276.dist-info/LICENSE,sha256=HrhfyXIkWY2tGFK11kg7vPCqhgh5DcxleloqdhrpyMY,11558
+auto_coder-0.1.276.dist-info/METADATA,sha256=515XHnua8SHruzQVf5S4pH2wboUIiasDzY92WnETZoc,2643
+auto_coder-0.1.276.dist-info/WHEEL,sha256=GV9aMThwP_4oNCtvEC2ec3qUYutgWeAzklro_0m4WJQ,91
+auto_coder-0.1.276.dist-info/entry_points.txt,sha256=0nzHtHH4pNcM7xq4EBA2toS28Qelrvcbrr59GqD_0Ak,350
+auto_coder-0.1.276.dist-info/top_level.txt,sha256=Jqc0_uJSw2GwoFQAa9iJxYns-2mWla-9ok_Y3Gcznjk,10
+auto_coder-0.1.276.dist-info/RECORD,,

autocoder/auto_coder_rag.py

@@ -17,6 +17,7 @@ import shlex
 from rich.console import Console
 from rich.table import Table
 import os
+import hashlib
 from loguru import logger
 import asyncio
 from datetime import datetime
@@ -32,6 +33,21 @@ if platform.system() == "Windows":
     init()
 
 
+def generate_unique_name_from_path(path: str) -> str:
+    """
+    Generate a unique name (MD5 hash) from a path after normalizing it.
+    For Linux/Unix systems, trailing path separators are removed.
+    """
+    if not path:
+        return ""
+    
+    # Normalize the path (resolve absolute path and remove trailing separators)
+    normalized_path = os.path.normpath(os.path.abspath(path))
+    
+    # Generate MD5 hash from the normalized path
+    return hashlib.md5(normalized_path.encode("utf-8")).hexdigest()
+
+
 def initialize_system(args):
     if args.product_mode == "lite":
         return
@@ -491,6 +507,10 @@ 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")
@@ -500,7 +520,7 @@ def main(input_args: Optional[List[str]] = None):
             try:
                 from byzerllm.apps.byzer_storage.simple_api import ByzerStorage
 
-                storage = ByzerStorage("byzerai_store", "rag", "files")
+                storage = ByzerStorage("byzerai_store", "rag", auto_coder_args.rag_build_name)
                 storage.retrieval.cluster_info("byzerai_store")
             except Exception as e:
                 logger.error(
@@ -611,6 +631,7 @@ def main(input_args: Optional[List[str]] = None):
 
         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)
             rag = RAGFactory.get_rag(
                 llm=llm,
                 args=auto_coder_args,
@@ -618,6 +639,7 @@ def main(input_args: Optional[List[str]] = None):
                 tokenizer_path=server_args.tokenizer_path,
             )
         else:
+            auto_coder_args.rag_build_name = generate_unique_name_from_path("")
             rag = RAGFactory.get_rag(llm=llm, args=auto_coder_args, path="")
 
         llm_wrapper = LLWrapper(llm=llm, rag=rag)
@@ -636,10 +658,7 @@ def main(input_args: Optional[List[str]] = None):
             logger.warning(f"Failed to save service info: {str(e)}")
         
         serve(llm=llm_wrapper, args=server_args)
-    elif args.command == "build_hybrid_index":
-        if not args.quick:
-            initialize_system()
-
+    elif args.command == "build_hybrid_index":        
         auto_coder_args = AutoCoderArgs(
             **{
                 arg: getattr(args, arg)
@@ -648,6 +667,11 @@ def main(input_args: Optional[List[str]] = None):
             }
         )
 
+        # Generate unique name for RAG build if doc_dir exists
+        if args.doc_dir:
+            auto_coder_args.rag_build_name = generate_unique_name_from_path(args.doc_dir)
+            logger.info(f"Generated RAG build name: {auto_coder_args.rag_build_name}")
+
         auto_coder_args.enable_hybrid_index = True
         auto_coder_args.rag_type = "simple"
 
@@ -675,6 +699,7 @@ def main(input_args: Optional[List[str]] = None):
                 return
             llm.setup_default_emb_model_name("emb")
 
+        auto_coder_args.rag_build_name = generate_unique_name_from_path(args.doc_dir)
         rag = RAGFactory.get_rag(
             llm=llm,
             args=auto_coder_args,

autocoder/auto_coder_runner.py

@@ -1384,7 +1384,7 @@ def commit(query: str):
             md5 = hashlib.md5(file_content.encode("utf-8")).hexdigest()
             file_name = os.path.basename(execute_file)
             commit_result = git_utils.commit_changes(
-                ".", f"auto_coder_{file_name}_{md5}\n{commit_message}"
+                ".", f"{commit_message}\nauto_coder_{file_name}_{md5}"
             )
             git_utils.print_commit_info(commit_result=commit_result)
             if commit_message:

autocoder/commands/auto_command.py

@@ -175,6 +175,7 @@ class CommandAutoTuner:
         Python版本: {{ env_info.python_version }}
         终端类型: {{ env_info.shell_type }}
         终端编码: {{ env_info.shell_encoding }}
+        当前用户: {{ current_user }}
         
         {%- if shell_type %}
         脚本类型：{{ shell_type }}
@@ -284,7 +285,8 @@ class CommandAutoTuner:
             "shell_type": shell_type,
             "shell_encoding": shells.get_terminal_encoding(),
             "conversation_safe_zone_tokens": self.args.conversation_prune_safe_zone_tokens,
-            "os_distribution": shells.get_os_distribution()
+            "os_distribution": shells.get_os_distribution(),
+            "current_user": shells.get_current_username()
         }
     
     @byzerllm.prompt()

autocoder/common/__init__.py

@@ -320,6 +320,7 @@ class AutoCoderArgs(pydantic.BaseModel):
 
     monitor_mode: bool = False
     enable_hybrid_index: bool = False
+    rag_build_name: Optional[str] = None
     disable_auto_window: bool = False
     filter_batch_size: Optional[int] = 5
     disable_segment_reorder: bool = False

autocoder/common/code_auto_merge.py

@@ -1,4 +1,3 @@
-
 import os
 from byzerllm.utils.client import code_utils
 from autocoder.common import AutoCoderArgs, git_utils
@@ -197,5 +196,5 @@ class CodeAutoMerge:
 
         self.printer.print_in_terminal("files_merged", total=total)
         if not force_skip_git and not self.args.skip_commit:
-            commit_result = git_utils.commit_changes(self.args.source_dir, f"auto_coder_{file_name}_{md5}\n{self.args.query}")
+            commit_result = git_utils.commit_changes(self.args.source_dir, f"{self.args.query}\nauto_coder_{file_name}_{md5}")
             git_utils.print_commit_info(commit_result=commit_result)

autocoder/common/code_auto_merge_diff.py

@@ -579,7 +579,7 @@ class CodeAutoMergeDiff:
 
         self.printer.print_in_terminal("files_merged_total", total=total)
         if not force_skip_git and not self.args.skip_commit:
-            commit_result = git_utils.commit_changes(self.args.source_dir, f"auto_coder_{file_name}_{md5}\n{self.args.query}")
+            commit_result = git_utils.commit_changes(self.args.source_dir, f"{self.args.query}\nauto_coder_{file_name}_{md5}")
             git_utils.print_commit_info(commit_result=commit_result)
         else:
             # Print edits for review

autocoder/common/code_auto_merge_editblock.py

@@ -425,7 +425,8 @@ class CodeAutoMergeEditBlock:
             if not force_skip_git and not self.args.skip_commit:
                 try:
                     commit_result = git_utils.commit_changes(
-                        self.args.source_dir, f"auto_coder_{file_name}_{md5}\n{self.args.query}"
+                        self.args.source_dir,
+                        f"{self.args.query}\nauto_coder_{file_name}_{md5}",
                     )
                     git_utils.print_commit_info(commit_result=commit_result)
                 except Exception as e:

autocoder/common/code_auto_merge_strict_diff.py

@@ -285,7 +285,9 @@ class CodeAutoMergeStrictDiff:
                             
         self.printer.print_in_terminal("files_merged_total", total=total)
         if not force_skip_git and not self.args.skip_commit:
-            commit_result = git_utils.commit_changes(self.args.source_dir, f"auto_coder_{file_name}_{md5}\n{self.args.query}")
+            commit_result = git_utils.commit_changes(
+                self.args.source_dir, f"{self.args.query}\nauto_coder_{file_name}_{md5}"
+            )
             git_utils.print_commit_info(commit_result=commit_result)
         else:
             # Print diff blocks for review

autocoder/common/command_generator.py

@@ -16,6 +16,7 @@ def _generate_shell_script(user_input: str) -> str:
     Python版本: {{ env_info.python_version }}
     终端类型: {{ env_info.shell_type }}
     终端编码: {{ env_info.shell_encoding }}
+    当前用户: {{ current_user }}
     
     {%- if shell_type %}
     脚本类型：{{ shell_type }}
@@ -51,7 +52,8 @@ def _generate_shell_script(user_input: str) -> str:
         "env_info": env_info,
         "shell_type": shell_type,
         "shell_encoding": shells.get_terminal_encoding(),
-        "os_distribution": shells.get_os_distribution()
+        "os_distribution": shells.get_os_distribution(),
+        "current_user": shells.get_current_username()
     }
 
 

autocoder/common/git_utils.py

@@ -118,7 +118,8 @@ def revert_changes(repo_path: str, message: str) -> bool:
             return False
 
         # 通过message定位到commit_hash
-        commit = repo.git.log("--all", f"--grep={message}", "--format=%H", "-n", "1")
+        # --grep 默认只搜索第一行 -F 参数将搜索模式视为固定字符串而非正则表达式
+        commit = repo.git.log("--all", f"--grep={message}", "-F", "--format=%H", "-n", "1")
         if not commit:
             logger.warning(f"No commit found with message: {message}")
             return False
@@ -170,7 +171,8 @@ def revert_change(repo_path: str, message: str) -> bool:
     repo = get_repo(repo_path)
     if repo is None:
         return False
-    commit = repo.git.log("--all", f"--grep={message}", "--format=%H", "-n", "1")
+    # --grep 默认只搜索第一行 -F 参数将搜索模式视为固定字符串而非正则表达式
+    commit = repo.git.log("--all", f"--grep={message}", "-F", "--format=%H", "-n", "1")
     if commit:
         repo.git.revert(commit, no_edit=True)
         logger.info(f"Reverted changes with commit message: {message}")
@@ -618,7 +620,9 @@ def generate_commit_message(changes_report: str) -> str:
 def get_commit_by_message(repo_path: str, message: str):
     repo = get_repo(repo_path)
     try:
-        commit_hash = repo.git.log("--all", f"--grep={message}", "--format=%H", "-n", "1")
+        commit_hash = repo.git.log(
+            "--all", f"--grep={message}", "-F", "--format=%H", "-n", "1"
+        )
         if not commit_hash:
             return None
         return repo.commit(commit_hash.strip())

autocoder/common/shells.py

@@ -10,6 +10,7 @@ from rich.console import Console
 from rich.panel import Panel
 from rich.text import Text
 from rich.live import Live
+import getpass
 
 from autocoder.common.result_manager import ResultManager
 
@@ -547,4 +548,25 @@ set PYTHONIOENCODING=utf-8
             try:
                 os.unlink(temp_file.name)
             except Exception:
-                pass
+                pass
+
+def get_current_username():
+    """
+    Get the current username across different operating systems.
+    
+    Returns:
+        str: The current username.
+    """
+    try:
+        # getpass.getuser() works on Windows, macOS, and Linux
+        username = getpass.getuser()
+        return username
+    except Exception:
+        # Fallback methods if getpass.getuser() fails
+        try:
+            if platform.system() == 'Windows':
+                return os.environ.get('USERNAME', '')
+            else:  # macOS/Linux
+                return os.environ.get('USER', '')
+        except Exception:
+            return ''

autocoder/rag/cache/base_cache.py

@@ -2,11 +2,26 @@ from pydantic import BaseModel
 from typing import List, Tuple,Dict,Optional,Any
 from abc import ABC, abstractmethod
 
+# New model class for file information
+class FileInfo(BaseModel):
+    file_path: str
+    relative_path: str
+    modify_time: float
+    file_md5: str
+
+# New model class for cache items
+class CacheItem(BaseModel):
+    file_path: str
+    relative_path: str
+    content: List[Dict[str, Any]]  # Serialized SourceCode objects
+    modify_time: float
+    md5: str
+
 class DeleteEvent(BaseModel):
     file_paths: List[str]
 
 class AddOrUpdateEvent(BaseModel):
-    file_infos: List[Tuple[str, str, float, str]]
+    file_infos: List[FileInfo]
 
 class BaseCacheManager(ABC):
     @abstractmethod

autocoder/rag/cache/byzer_storage_cache.py

@@ -2,6 +2,8 @@ from autocoder.rag.cache.base_cache import (
     BaseCacheManager,
     DeleteEvent,
     AddOrUpdateEvent,
+    FileInfo,
+    CacheItem
 )
 from typing import Generator, List, Dict, Any, Optional, Tuple
 from autocoder.common import SourceCode
@@ -25,12 +27,15 @@ from autocoder.rag.variable_holder import VariableHolder
 import platform
 import hashlib
 from typing import Union
+from pydantic import BaseModel
 
 if platform.system() != "Windows":
     import fcntl
 else:
     fcntl = None
 
+
+
 def generate_file_md5(file_path: str) -> str:
     md5_hash = hashlib.md5()
     with open(file_path, "rb") as f:
@@ -62,7 +67,8 @@ class ByzerStorageCache(BaseCacheManager):
         self.path = path
         self.ignore_spec = ignore_spec
         self.required_exts = required_exts
-        self.storage = ByzerStorage("byzerai_store", "rag", "files")
+        self.rag_build_name = extra_params.rag_build_name
+        self.storage = ByzerStorage("byzerai_store", "rag", self.rag_build_name)
         self.queue = []
         self.chunk_size = 1000
         self._init_schema()
@@ -71,11 +77,12 @@ class ByzerStorageCache(BaseCacheManager):
             raise ValueError("extra_params is required for ByzerStorageCache")
 
         self.max_output_tokens = extra_params.hybrid_index_max_output_tokens
+        
 
         # 设置缓存文件路径
         self.cache_dir = os.path.join(self.path, ".cache")
         self.cache_file = os.path.join(self.cache_dir, "byzer_storage_speedup.jsonl")
-        self.cache = {}
+        self.cache: Dict[str, CacheItem] = {}
 
         self.lock = threading.Lock()
         self.stop_event = threading.Event()
@@ -122,18 +129,20 @@ class ByzerStorageCache(BaseCacheManager):
             .execute()
         )
 
-    def _load_cache(self) -> dict:
+    def _load_cache(self) -> Dict[str, CacheItem]:
         """Load cache from file"""
         if os.path.exists(self.cache_file):
             try:
-                with open(self.cache_file, "r",encoding="utf-8") as f:
+                with open(self.cache_file, "r", encoding="utf-8") as f:
                     lines = f.readlines()
                     cache = {}
                     for line in lines:
                         try:
                             data = json.loads(line.strip())
                             if isinstance(data, dict) and "file_path" in data:
-                                cache[data["file_path"]] = data
+                                # 转换为 CacheItem 对象
+                                cache_item = CacheItem.model_validate(data)
+                                cache[data["file_path"]] = cache_item
                         except json.JSONDecodeError:
                             continue
                     return cache
@@ -147,38 +156,42 @@ class ByzerStorageCache(BaseCacheManager):
 
         if not fcntl:
             try:
-                with open(cache_file, "w",encoding="utf-8") as f:
-                    for data in self.cache.values():
-                        json.dump(data, f, ensure_ascii=False)
+                with open(cache_file, "w", encoding="utf-8") as f:
+                    for cache_item in self.cache.values():
+                        # 确保序列化 Pydantic 模型
+                        json.dump(cache_item.model_dump(), f, ensure_ascii=False)
                         f.write("\n")
             except IOError as e:
                 logger.error(f"Error writing cache file: {str(e)}")
         else:
             lock_file = cache_file + ".lock"
-            with open(lock_file, "w",encoding="utf-8") as lockf:
+            with open(lock_file, "w", encoding="utf-8") as lockf:
                 try:
                     # 获取文件锁
                     fcntl.flock(lockf, fcntl.LOCK_EX | fcntl.LOCK_NB)
                     # 写入缓存文件
-                    with open(cache_file, "w",encoding="utf-8") as f:
-                        for data in self.cache.values():
-                            json.dump(data, f, ensure_ascii=False)
+                    with open(cache_file, "w", encoding="utf-8") as f:
+                        for cache_item in self.cache.values():
+                            # 确保序列化 Pydantic 模型
+                            json.dump(cache_item.model_dump(), f, ensure_ascii=False)
                             f.write("\n")
 
                 finally:
                     # 释放文件锁
                     fcntl.flock(lockf, fcntl.LOCK_UN)
 
+    def fileinfo_to_tuple(self, file_info: FileInfo) -> Tuple[str, str, float, str]:
+        return (file_info.file_path, file_info.relative_path, file_info.modify_time, file_info.file_md5)
+
     def build_cache(self):
         """Build the cache by reading files and storing in Byzer Storage"""
         logger.info(f"Building cache for path: {self.path}")
 
         files_to_process = []
-        for file_info in self.get_all_files():
-            file_path, _, modify_time, file_md5 = file_info
+        for file_info in self.get_all_files():            
             if (
-                file_path not in self.cache                
-                or self.cache[file_path]["md5"] != file_md5
+                file_info.file_path not in self.cache                
+                or self.cache[file_info.file_path].md5 != file_info.file_md5
             ):
                 files_to_process.append(file_info)
                 
@@ -192,19 +205,21 @@ class ByzerStorageCache(BaseCacheManager):
             initializer=initialize_tokenizer,
             initargs=(VariableHolder.TOKENIZER_PATH,),
         ) as pool:
-            results = pool.map(process_file_in_multi_process, files_to_process)
+            target_files_to_process = []
+            for file_info in files_to_process:
+                target_files_to_process.append(self.fileinfo_to_tuple(file_info))
+            results = pool.map(process_file_in_multi_process, target_files_to_process)
 
         items = []
         for file_info, result in zip(files_to_process, results):            
-            file_path, relative_path, modify_time, file_md5 = file_info
             content: List[SourceCode] = result
-            self.cache[file_path] = {
-                "file_path": file_path,
-                "relative_path": relative_path,
-                "content": [c.model_dump() for c in content],
-                "modify_time": modify_time,
-                "md5": file_md5,
-            }
+            self.cache[file_info.file_path] = CacheItem(
+                file_path=file_info.file_path,
+                relative_path=file_info.relative_path,
+                content=[c.model_dump() for c in content],
+                modify_time=file_info.modify_time,
+                md5=file_info.file_md5,
+            )
 
             for doc in content:
                 logger.info(f"Processing file: {doc.module_name}")
@@ -213,11 +228,11 @@ class ByzerStorageCache(BaseCacheManager):
                 for chunk_idx, chunk in enumerate(chunks):
                     chunk_item = {
                         "_id": f"{doc.module_name}_{chunk_idx}",
-                        "file_path": file_path,
+                        "file_path": file_info.file_path,
                         "content": chunk,
                         "raw_content": chunk,
                         "vector": chunk,
-                        "mtime": modify_time,
+                        "mtime": file_info.modify_time,
                     }
                     items.append(chunk_item)
 
@@ -260,9 +275,9 @@ class ByzerStorageCache(BaseCacheManager):
             
             self.storage.commit()
 
-    def update_storage(self, file_path, is_delete: bool):
+    def update_storage(self, file_info: FileInfo, is_delete: bool):
         query = self.storage.query_builder()
-        query.and_filter().add_condition("file_path", file_path).build()
+        query.and_filter().add_condition("file_path", file_info.file_path).build()
         results = query.execute()
         if results:
             for result in results:
@@ -272,9 +287,9 @@ class ByzerStorageCache(BaseCacheManager):
         if not is_delete:
             content = [
                 SourceCode.model_validate(doc)
-                for doc in self.cache[file_path]["content"]
+                for doc in self.cache[file_info.file_path].content
             ]
-            modify_time = self.cache[file_path]["modify_time"]
+            modify_time = self.cache[file_info.file_path].modify_time
             for doc in content:
                 logger.info(f"Processing file: {doc.module_name}")
                 doc.module_name
@@ -282,7 +297,7 @@ class ByzerStorageCache(BaseCacheManager):
                 for chunk_idx, chunk in enumerate(chunks):
                     chunk_item = {
                         "_id": f"{doc.module_name}_{chunk_idx}",
-                        "file_path": file_path,
+                        "file_path": file_info.file_path,
                         "content": chunk,
                         "raw_content": chunk,
                         "vector": chunk,
@@ -302,26 +317,34 @@ class ByzerStorageCache(BaseCacheManager):
                 for item in file_list.file_paths:
                     logger.info(f"{item} is detected to be removed")
                     del self.cache[item]
-                    self.update_storage(item, is_delete=True)
+                    # 创建一个临时的 FileInfo 对象
+                    file_info = FileInfo(file_path=item, relative_path="", modify_time=0, file_md5="")
+                    self.update_storage(file_info, is_delete=True)
 
             elif isinstance(file_list, AddOrUpdateEvent):
                 for file_info in file_list.file_infos:
-                    logger.info(f"{file_info[0]} is detected to be updated")
-                    result = process_file_local(file_info[0])
-                    self.cache[file_info[0]] = result
-                    self.update_storage(file_info[0], is_delete=False)
+                    logger.info(f"{file_info.file_path} is detected to be updated")
+                    # 处理文件并创建 CacheItem
+                    content = process_file_local(self.fileinfo_to_tuple(file_info))
+                    self.cache[file_info.file_path] = CacheItem(
+                        file_path=file_info.file_path,
+                        relative_path=file_info.relative_path,
+                        content=[c.model_dump() for c in content],
+                        modify_time=file_info.modify_time,
+                        md5=file_info.file_md5,
+                    )
+                    self.update_storage(file_info, is_delete=False)
             self.write_cache()
 
     def trigger_update(self):
         logger.info("检查文件是否有更新.....")
         files_to_process = []
         current_files = set()
-        for file_info in self.get_all_files():
-            file_path, _, _, file_md5 = file_info
-            current_files.add(file_path)
+        for file_info in self.get_all_files():            
+            current_files.add(file_info.file_path)
             if (
-                file_path not in self.cache
-                or self.cache[file_path]["md5"] != file_md5
+                file_info.file_path not in self.cache
+                or self.cache[file_info.file_path].md5 != file_info.file_md5
             ):
                 files_to_process.append(file_info)
 
@@ -341,7 +364,7 @@ class ByzerStorageCache(BaseCacheManager):
         self.trigger_update()
 
         if options is None or "query" not in options:
-            return self.cache
+            return {file_path: self.cache[file_path].model_dump() for file_path in self.cache}
         
         query = options.get("query", "")
         total_tokens = 0    
@@ -361,6 +384,7 @@ class ByzerStorageCache(BaseCacheManager):
         results = query_builder.execute()
 
         # Group results by file_path and reconstruct documents while preserving order
+        # 这里还可以有排序优化，综合考虑一篇内容出现的次数以及排序位置
         file_paths = []
         seen = set()
         for result in results:
@@ -374,17 +398,17 @@ class ByzerStorageCache(BaseCacheManager):
         for file_path in file_paths:
             if file_path in self.cache:
                 cached_data = self.cache[file_path]
-                for doc in cached_data["content"]:                    
+                for doc in cached_data.content:                    
                     if total_tokens + doc["tokens"] > self.max_output_tokens:
                         return result
                     total_tokens += doc["tokens"]
-                result[file_path] = cached_data
+                result[file_path] = cached_data.model_dump()
 
         return result
 
                 
 
-    def get_all_files(self) -> List[Tuple[str, str, float, str]]:
+    def get_all_files(self) -> List[FileInfo]:
         all_files = []
         for root, dirs, files in os.walk(self.path,followlinks=True):
             dirs[:] = [d for d in dirs if not d.startswith(".") and d not in default_ignore_dirs]
@@ -412,6 +436,9 @@ class ByzerStorageCache(BaseCacheManager):
                 relative_path = os.path.relpath(file_path, self.path)
                 modify_time = os.path.getmtime(file_path)
                 file_md5 = generate_file_md5(file_path)
-                all_files.append((file_path, relative_path, modify_time, file_md5))
+                all_files.append(FileInfo(file_path=file_path, 
+                                          relative_path=relative_path, 
+                                          modify_time=modify_time, 
+                                          file_md5=file_md5))
 
         return all_files

autocoder/rag/long_context_rag.py

@@ -71,15 +71,21 @@ class LongContextRAG:
         tokenizer_path: Optional[str] = None,
     ) -> None:
         self.llm = llm
-        self.args = args
-        if args.product_mode == "pro":
-            self.index_model = byzerllm.ByzerLLM()
-            self.index_model.setup_default_model_name(
-                args.index_model or self.llm.default_model_name
-            )
-        else:
-            self.index_model = self.llm
+        self.recall_llm = self.llm
+        self.chunk_llm = self.llm
+        self.qa_llm = self.llm
+
+        if self.llm.get_sub_client("qa_model"):
+            self.qa_llm = self.llm.get_sub_client("qa_model")
+
+        if self.llm.get_sub_client("recall_model"):
+            self.recall_llm = self.llm.get_sub_client("recall_model")
 
+        if self.llm.get_sub_client("chunk_model"):
+            self.chunk_llm = self.llm.get_sub_client("chunk_model")
+
+        self.args = args
+        
         self.path = path
         self.relevant_score = self.args.rag_doc_filter_relevance or 5
 
@@ -162,7 +168,7 @@ class LongContextRAG:
         )
 
         self.doc_filter = DocFilter(
-            self.index_model, self.args, on_ray=self.on_ray, path=self.path
+            self.llm, self.args, on_ray=self.on_ray, path=self.path
         )
 
         doc_num = 0
@@ -459,22 +465,23 @@ class LongContextRAG:
 
             logger.info(f"Query: {query} only_contexts: {only_contexts}")
             start_time = time.time()
+            
 
             rag_stat = RAGStat(
                 recall_stat=RecallStat(
                     total_input_tokens=0,
                     total_generated_tokens=0,
-                    model_name=self.llm.default_model_name,
+                    model_name=self.recall_llm.default_model_name,
                 ),
                 chunk_stat=ChunkStat(
                     total_input_tokens=0,
                     total_generated_tokens=0,
-                    model_name=self.llm.default_model_name,
+                    model_name=self.chunk_llm.default_model_name,
                 ),
                 answer_stat=AnswerStat(
                     total_input_tokens=0,
                     total_generated_tokens=0,
-                    model_name=self.llm.default_model_name,
+                    model_name=self.qa_llm.default_model_name,
                 ),
             )
 
@@ -624,7 +631,7 @@ class LongContextRAG:
 
             # 记录令牌统计
             request_tokens = sum([doc.tokens for doc in relevant_docs])
-            target_model = model or self.llm.default_model_name
+            target_model = target_llm.default_model_name
             logger.info(
                 f"=== LLM Request ===\n"
                 f"  * Target model: {target_model}\n"

autocoder/rag/utils.py

@@ -11,7 +11,7 @@ import traceback
 
 
 def process_file_in_multi_process(
-    file_info: Tuple[str, str, float]
+    file_info: Tuple[str, str, float, str]
 ) -> List[SourceCode]:
     start_time = time.time()
     file_path, relative_path, _, _ = file_info

autocoder/version.py

@@ -1 +1 @@
-__version__ = "0.1.274"
+__version__ = "0.1.276"