auto-coder 0.1.274__tar.gz → 0.1.275__tar.gz

{auto_coder-0.1.274 → auto_coder-0.1.275}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: auto-coder
-Version: 0.1.274
+Version: 0.1.275
 Summary: AutoCoder: AutoCoder
 Author: allwefantasy
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence

{auto_coder-0.1.274 → auto_coder-0.1.275}/src/auto_coder.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: auto-coder
-Version: 0.1.274
+Version: 0.1.275
 Summary: AutoCoder: AutoCoder
 Author: allwefantasy
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence

{auto_coder-0.1.274 → auto_coder-0.1.275}/src/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,

{auto_coder-0.1.274 → auto_coder-0.1.275}/src/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:

{auto_coder-0.1.274 → auto_coder-0.1.275}/src/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()

{auto_coder-0.1.274 → auto_coder-0.1.275}/src/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

{auto_coder-0.1.274 → auto_coder-0.1.275}/src/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)

{auto_coder-0.1.274 → auto_coder-0.1.275}/src/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

{auto_coder-0.1.274 → auto_coder-0.1.275}/src/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:

{auto_coder-0.1.274 → auto_coder-0.1.275}/src/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

{auto_coder-0.1.274 → auto_coder-0.1.275}/src/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()
     }
 
 

{auto_coder-0.1.274 → auto_coder-0.1.275}/src/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())

{auto_coder-0.1.274 → auto_coder-0.1.275}/src/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 ''

auto_coder-0.1.275/src/autocoder/rag/cache/base_cache.py

@@ -0,0 +1,29 @@
+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[FileInfo]
+
+class BaseCacheManager(ABC):
+    @abstractmethod
+    def get_cache(self,options:Optional[Dict[str,Any]]=None) -> Dict[str, Dict]:
+        pass

{auto_coder-0.1.274 → auto_coder-0.1.275}/src/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

{auto_coder-0.1.274 → auto_coder-0.1.275}/src/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

auto_coder-0.1.275/src/autocoder/version.py

@@ -0,0 +1 @@
+__version__ = "0.1.275"

auto_coder-0.1.274/src/autocoder/rag/cache/base_cache.py

@@ -1,14 +0,0 @@
-from pydantic import BaseModel
-from typing import List, Tuple,Dict,Optional,Any
-from abc import ABC, abstractmethod
-
-class DeleteEvent(BaseModel):
-    file_paths: List[str]
-
-class AddOrUpdateEvent(BaseModel):
-    file_infos: List[Tuple[str, str, float, str]]
-
-class BaseCacheManager(ABC):
-    @abstractmethod
-    def get_cache(self,options:Optional[Dict[str,Any]]=None) -> Dict[str, Dict]:
-        pass

auto_coder-0.1.274/src/autocoder/version.py

@@ -1 +0,0 @@
-__version__ = "0.1.274"