auto-coder 0.1.212__py3-none-any.whl → 0.1.213__py3-none-any.whl

{auto_coder-0.1.212.dist-info → auto_coder-0.1.213.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: auto-coder
-Version: 0.1.212
+Version: 0.1.213
 Summary: AutoCoder: AutoCoder
 Author: allwefantasy
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence

{auto_coder-0.1.212.dist-info → auto_coder-0.1.213.dist-info}/RECORD

@@ -1,5 +1,5 @@
 autocoder/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-autocoder/auto_coder.py,sha256=JD8wsideQo1TYAs7tw4sHETRh4EKgB2TqGcrKjzVz2g,41215
+autocoder/auto_coder.py,sha256=fnm1dhTj5AidkYf65qt4XjrXZz0D3OrkHW8y7QOMjmQ,41765
 autocoder/auto_coder_lang.py,sha256=Rtupq6N3_HT7JRhDKdgCBcwRaiAnyCOR_Gsp4jUomrI,3229
 autocoder/auto_coder_rag.py,sha256=hU7NmCUIM__918p1RbGFlJKPVEqdr4NXjD6b1dBgEVU,21743
 autocoder/auto_coder_server.py,sha256=XU9b4SBH7zjPPXaTWWHV4_zJm-XYa6njuLQaplYJH_c,20290
@@ -8,7 +8,7 @@ autocoder/chat_auto_coder.py,sha256=WhaS6uiG5lIBrnif_Sa_M4aTBMTiVn8-eGhlDWjtdBc,
 autocoder/chat_auto_coder_lang.py,sha256=zU9VRY-l80fZnLJ0Op8A3wq27UhQHh9WcpSYU4SmnqU,8708
 autocoder/command_args.py,sha256=BpMbPceBzjCftPB0yOVsSmTmt61xS3gtc1WGKtcDHQs,30449
 autocoder/lang.py,sha256=U6AjVV8Rs1uLyjFCZ8sT6WWuNUxMBqkXXIOs4S120uk,14511
-autocoder/version.py,sha256=GdkOhsXexDY7uY8Ip0dBIeDsIU8IrxsHi7LvRzcCZEk,24
+autocoder/version.py,sha256=V-hdG0k1-v5-WK9KQgzySDX22RDvdwG4vsIdj8042Cg,24
 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
@@ -36,7 +36,7 @@ autocoder/common/code_auto_merge.py,sha256=1tiKrram8GCCuPYrNvHFbr-Tic-f84Y8H13fD
 autocoder/common/code_auto_merge_diff.py,sha256=5SI6ggklJ0QDHvsS0cpNXFuIkFRQxp1i08k6iwWbPJc,15242
 autocoder/common/code_auto_merge_editblock.py,sha256=l6yEiZqXyIlUNIIPXvkHOnLCIInXR78TzSjF-jtJkkg,17035
 autocoder/common/code_auto_merge_strict_diff.py,sha256=ABYOTDUQYA4Bn4BwT1Rw812y49cHW3UH_JSpM9uJ6ig,9399
-autocoder/common/code_modification_ranker.py,sha256=CbajZlsLSXt6D26zka0ECc228tJTHzXZMPgRq7nMhXw,4885
+autocoder/common/code_modification_ranker.py,sha256=DFlbwgdg8GK47zVcvfZSzkyniEKmTVLTOWejjcVIgaw,5121
 autocoder/common/command_completer.py,sha256=Eo3V9fRIIX74YdLxz1pZ3wAqdXhFirWoQ0zXkB5R198,8999
 autocoder/common/command_generator.py,sha256=v4LmU7sO-P7jEZIXCWHUC6P-vT7AvBi_x_PTwCqBAE8,1323
 autocoder/common/command_templates.py,sha256=3G-pCNbL6iHbnkG6v1JZpbIK3Mc9d373_RYGmCcDPMY,8548
@@ -67,7 +67,7 @@ autocoder/dispacher/actions/plugins/action_regex_project.py,sha256=ZvU3hsxFPSZTf
 autocoder/dispacher/actions/plugins/action_translate.py,sha256=nVAtRSQpdGNmZxg1R_9zXG3AuTv3CHf2v7ODgj8u65c,7727
 autocoder/index/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 autocoder/index/for_command.py,sha256=zfbvQnhHjsAqBc4Ce1kMGIu0jPEk_rtH7fntg89_4z0,3092
-autocoder/index/index.py,sha256=EpS3ZsCL6WuMk2hLiUe5J_XIny5uvlEuQLw68WzYxKk,31823
+autocoder/index/index.py,sha256=X9pAQj-rZar-OhYrx3BGAxNfM1hObz_SG3bQdz5efQg,34145
 autocoder/index/symbols_utils.py,sha256=CjcjUVajmJZB75Ty3a7kMv1BZphrm-tIBAdOJv6uo-0,2037
 autocoder/pyproject/__init__.py,sha256=7ZuIxD2QBYIwhjmpva8eL2knorKo03yNqUhSyecpt7c,14448
 autocoder/rag/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -119,9 +119,9 @@ autocoder/utils/request_event_queue.py,sha256=r3lo5qGsB1dIjzVQ05dnr0z_9Z3zOkBdP1
 autocoder/utils/request_queue.py,sha256=nwp6PMtgTCiuwJI24p8OLNZjUiprC-TsefQrhMI-yPE,3889
 autocoder/utils/rest.py,sha256=HawagAap3wMIDROGhY1730zSZrJR_EycODAA5qOj83c,8807
 autocoder/utils/tests.py,sha256=BqphrwyycGAvs-5mhH8pKtMZdObwhFtJ5MC_ZAOiLq8,1340
-auto_coder-0.1.212.dist-info/LICENSE,sha256=HrhfyXIkWY2tGFK11kg7vPCqhgh5DcxleloqdhrpyMY,11558
-auto_coder-0.1.212.dist-info/METADATA,sha256=FtesXVxtmVzhaNmIqFZJEf7p6zPnXANbdlALh88wzyo,2590
-auto_coder-0.1.212.dist-info/WHEEL,sha256=GV9aMThwP_4oNCtvEC2ec3qUYutgWeAzklro_0m4WJQ,91
-auto_coder-0.1.212.dist-info/entry_points.txt,sha256=0nzHtHH4pNcM7xq4EBA2toS28Qelrvcbrr59GqD_0Ak,350
-auto_coder-0.1.212.dist-info/top_level.txt,sha256=Jqc0_uJSw2GwoFQAa9iJxYns-2mWla-9ok_Y3Gcznjk,10
-auto_coder-0.1.212.dist-info/RECORD,,
+auto_coder-0.1.213.dist-info/LICENSE,sha256=HrhfyXIkWY2tGFK11kg7vPCqhgh5DcxleloqdhrpyMY,11558
+auto_coder-0.1.213.dist-info/METADATA,sha256=7JrELvSZVoZAD3daumw7vNKywrDnYZaqtw8QkuFczEE,2590
+auto_coder-0.1.213.dist-info/WHEEL,sha256=GV9aMThwP_4oNCtvEC2ec3qUYutgWeAzklro_0m4WJQ,91
+auto_coder-0.1.213.dist-info/entry_points.txt,sha256=0nzHtHH4pNcM7xq4EBA2toS28Qelrvcbrr59GqD_0Ak,350
+auto_coder-0.1.213.dist-info/top_level.txt,sha256=Jqc0_uJSw2GwoFQAa9iJxYns-2mWla-9ok_Y3Gcznjk,10
+auto_coder-0.1.213.dist-info/RECORD,,

autocoder/auto_coder.py

@@ -295,9 +295,20 @@ def main(input_args: Optional[List[str]] = None):
                 llm.setup_sub_client("code_model", code_model)
 
         if args.generate_rerank_model:
-            generate_rerank_model = byzerllm.ByzerLLM()
-            generate_rerank_model.setup_default_model_name(args.generate_rerank_model)
-            llm.setup_sub_client("generate_rerank_model", generate_rerank_model)
+            if "," in args.generate_rerank_model:
+                # Multiple rerank models specified
+                model_names = args.generate_rerank_model.split(",")
+                models = []
+                for _, model_name in enumerate(model_names):
+                    rerank_model = byzerllm.ByzerLLM()
+                    rerank_model.setup_default_model_name(model_name.strip())
+                    models.append(rerank_model)
+                llm.setup_sub_client("generate_rerank_model", models)
+            else:
+                # Single rerank model
+                rerank_model = byzerllm.ByzerLLM()
+                rerank_model.setup_default_model_name(args.generate_rerank_model)
+                llm.setup_sub_client("generate_rerank_model", rerank_model)
 
         if args.inference_model:
             inference_model = byzerllm.ByzerLLM()

autocoder/common/code_modification_ranker.py

@@ -14,10 +14,9 @@ class CodeModificationRanker:
     def __init__(self, llm: byzerllm.ByzerLLM, args: AutoCoderArgs):
         self.llm = llm
         self.args = args     
-        if self.llm.get_sub_client("generate_rerank_model"):
-            self.rerank_llm = self.llm.get_sub_client("generate_rerank_model")
-        else:
-            self.rerank_llm = self.llm
+        self.llms = self.llm.get_sub_client("generate_rerank_model") or [self.llm]
+        if not isinstance(self.llms, list):
+            self.llms = [self.llms]
     
     @byzerllm.prompt()
     def _rank_modifications(self, s:CodeGenerateResult) -> str:
@@ -63,17 +62,20 @@ class CodeModificationRanker:
             
         logger.info(f"Start ranking {len(generate_result.contents)} candidates")
         generate_times = self.args.generate_times_same_model
-        
+        total_tasks = len(self.llms) * generate_times
         try:
-            # Create a thread pool with generate_times workers
-            with ThreadPoolExecutor(max_workers=generate_times) as executor:
-                # Submit tasks
-                futures = [
-                    executor.submit(
-                        self._rank_modifications.with_llm(self.rerank_llm).with_return_type(RankResult).run,
-                        generate_result
-                    ) for _ in range(generate_times)
-                ]
+            # Create a thread pool with (number of models * generate_times) workers            
+            with ThreadPoolExecutor(max_workers=total_tasks) as executor:
+                # Submit tasks for each model and generate_times
+                futures = []
+                for llm in self.llms:
+                    for _ in range(generate_times):
+                        futures.append(
+                            executor.submit(
+                                self._rank_modifications.with_llm(llm).with_return_type(RankResult).run,
+                                generate_result
+                            )
+                        )
                 
                 # Collect all results
                 results = []
@@ -104,7 +106,7 @@ class CodeModificationRanker:
                 elapsed = time.time() - start_time
                 # Format scores for logging
                 score_details = ", ".join([f"candidate {i}: {candidate_scores[i]:.2f}" for i in sorted_candidates])
-                logger.info(f"Ranking completed in {elapsed:.2f}s, best candidate index: {sorted_candidates[0]}, scores: {score_details}")
+                logger.info(f"Ranking completed in {elapsed:.2f}s, total voters: {total_tasks}, best candidate index: {sorted_candidates[0]}, scores: {score_details}")
                 
                 rerank_contents = [generate_result.contents[i] for i in sorted_candidates]
                 rerank_conversations = [generate_result.conversations[i] for i in sorted_candidates]

autocoder/index/index.py

@@ -578,8 +578,18 @@ def build_index_and_filter_files(
         "indexed_files": 0,
         "level1_filtered": 0,
         "level2_filtered": 0,
+        "verified_files": 0,
         "final_files": 0,
-        "timings": {}
+        "timings": {
+            "process_tagged_sources": 0.0,
+            "build_index": 0.0,
+            "level1_filter": 0.0,
+            "level2_filter": 0.0,
+            "relevance_verification": 0.0,
+            "file_selection": 0.0,
+            "prepare_output": 0.0,
+            "total": 0.0
+        }
     }
 
     def get_file_path(file_path):
@@ -597,7 +607,8 @@ def build_index_and_filter_files(
             final_files[get_file_path(source.module_name)] = TargetFile(
                 file_path=source.module_name, reason="Rest/Rag/Search"
             )
-    stats["timings"]["process_tagged_sources"] = time.monotonic() - phase_start
+    phase_end = time.monotonic()
+    stats["timings"]["process_tagged_sources"] = phase_end - phase_start
 
     if not args.skip_build_index and llm:
         # Phase 2: Build index
@@ -615,7 +626,8 @@ def build_index_and_filter_files(
         index_manager = IndexManager(llm=llm, sources=sources, args=args)
         index_data = index_manager.build_index()
         stats["indexed_files"] = len(index_data) if index_data else 0
-        stats["timings"]["build_index"] = time.monotonic() - phase_start
+        phase_end = time.monotonic()
+        stats["timings"]["build_index"] = phase_end - phase_start
 
         if args.request_id and not args.skip_events:
             queue_communicate.send_event(
@@ -650,7 +662,8 @@ def build_index_and_filter_files(
                     file_path = file.file_path.strip()
                     final_files[get_file_path(file_path)] = file
                 stats["level1_filtered"] = len(target_files.file_list)
-            stats["timings"]["level1_filter"] = time.monotonic() - phase_start
+            phase_end = time.monotonic()
+            stats["timings"]["level1_filter"] = phase_end - phase_start
 
             # Phase 4: Level 2 filtering - Related files
             if target_files is not None and args.index_filter_level >= 2:
@@ -673,8 +686,8 @@ def build_index_and_filter_files(
                         file_path = file.file_path.strip()
                         final_files[get_file_path(file_path)] = file
                     stats["level2_filtered"] = len(related_files.file_list)
-                stats["timings"]["level2_filter"] = time.monotonic() - \
-                    phase_start
+                phase_end = time.monotonic()
+                stats["timings"]["level2_filter"] = phase_end - phase_start
 
             if not final_files:
                 logger.warning("No related files found, using all files")
@@ -689,6 +702,28 @@ def build_index_and_filter_files(
             phase_start = time.monotonic()
             verified_files = {}
             temp_files = list(final_files.values())
+            verification_results = []
+            
+            def print_verification_results(results):
+                from rich.table import Table
+                from rich.console import Console
+                
+                console = Console()
+                table = Table(title="File Relevance Verification Results", show_header=True, header_style="bold magenta")
+                table.add_column("File Path", style="cyan", no_wrap=True)
+                table.add_column("Score", justify="right", style="green")
+                table.add_column("Status", style="yellow")
+                table.add_column("Reason/Error")
+                
+                for file_path, score, status, reason in results:
+                    table.add_row(
+                        file_path,
+                        str(score) if score is not None else "N/A",
+                        status,
+                        reason
+                    )
+                
+                console.print(table)
 
             def verify_single_file(file: TargetFile):
                 for source in sources:
@@ -700,13 +735,20 @@ def build_index_and_filter_files(
                                 query=args.query
                             )
                             if result.relevant_score >= args.verify_file_relevance_score:
-                                return file.file_path, TargetFile(
+                                verified_files[file.file_path] = TargetFile(
                                     file_path=file.file_path,
                                     reason=f"Score:{result.relevant_score}, {result.reason}"
                                 )
+                                return file.file_path, result.relevant_score, "PASS", result.reason
+                            else:
+                                return file.file_path, result.relevant_score, "FAIL", result.reason
                         except Exception as e:
-                            logger.warning(
-                                f"Failed to verify file {file.file_path}: {str(e)}")
+                            error_msg = str(e)
+                            verified_files[file.file_path] = TargetFile(
+                                file_path=file.file_path,
+                                reason=f"Verification failed: {error_msg}"
+                            )
+                            return file.file_path, None, "ERROR", error_msg
                 return None
 
             with ThreadPoolExecutor(max_workers=args.index_filter_workers) as executor:
@@ -715,15 +757,18 @@ def build_index_and_filter_files(
                 for future in as_completed(futures):
                     result = future.result()
                     if result:
-                        file_path, target_file = result
-                        verified_files[file_path] = target_file
+                        verification_results.append(result)
                         time.sleep(args.anti_quota_limit)
 
+            # Print verification results in a table
+            print_verification_results(verification_results)
+            
             stats["verified_files"] = len(verified_files)
-            stats["timings"]["relevance_verification"] = time.monotonic() - \
-                phase_start
+            phase_end = time.monotonic()
+            stats["timings"]["relevance_verification"] = phase_end - phase_start
 
-            final_files = verified_files if verified_files else final_files
+            # Keep all files, not just verified ones
+            final_files = verified_files
 
     def display_table_and_get_selections(data):
         from prompt_toolkit.shortcuts import checkboxlist_dialog
@@ -755,7 +800,7 @@ def build_index_and_filter_files(
         console = Console()
 
         table = Table(
-            title="Target Files You Selected",
+            title="Files Used as Context",
             show_header=True,
             header_style="bold magenta",
         )
@@ -802,7 +847,8 @@ def build_index_and_filter_files(
         if args.index_filter_file_num > 0:
             final_filenames = final_filenames[: args.index_filter_file_num]
 
-    stats["timings"]["file_selection"] = time.monotonic() - phase_start
+    phase_end = time.monotonic()
+    stats["timings"]["file_selection"] = phase_end - phase_start
 
     # Phase 7: Display results and prepare output
     logger.info("Phase 7: Preparing final output...")
@@ -848,26 +894,42 @@ def build_index_and_filter_files(
         )        
 
     stats["final_files"] = len(depulicated_sources)
-    stats["timings"]["prepare_output"] = time.monotonic() - phase_start
+    phase_end = time.monotonic()
+    stats["timings"]["prepare_output"] = phase_end - phase_start
 
     # Calculate total time and print summary
-    total_time = time.monotonic() - total_start_time
+    total_end_time = time.monotonic()
+    total_time = total_end_time - total_start_time
     stats["timings"]["total"] = total_time
+    
+    # Calculate total filter time
+    total_filter_time = (
+        stats["timings"]["level1_filter"] +
+        stats["timings"]["level2_filter"] +
+        stats["timings"]["relevance_verification"]
+    )
 
-    # Print final statistics
-    logger.info("\n=== Build Index and Filter Files Summary ===")
-    logger.info(f"Total files in project: {stats['total_files']}")
-    logger.info(f"Files indexed: {stats['indexed_files']}")
-    logger.info(f"Files after Level 1 filter: {stats['level1_filtered']}")
-    logger.info(f"Files after Level 2 filter: {stats['level2_filtered']}")
-    logger.info(
-        f"Files after relevance verification: {stats.get('verified_files', 0)}")
-    logger.info(f"Final files selected: {stats['final_files']}")
-    logger.info("\nTime breakdown:")
-    for phase, duration in stats["timings"].items():
-        logger.info(f"  - {phase}: {duration:.2f}s")
-    logger.info(f"Total execution time: {total_time:.2f}s")
-    logger.info("==========================================\n")
+    # Print final statistics in a more structured way
+    summary = f"""
+=== Indexing and Filtering Summary ===
+• Total files scanned: {stats['total_files']}
+• Files indexed: {stats['indexed_files']}
+• Files filtered:
+  - Level 1 (query-based): {stats['level1_filtered']}
+  - Level 2 (related files): {stats['level2_filtered']}
+  - Relevance verified: {stats.get('verified_files', 0)}
+• Final files selected: {stats['final_files']}
+
+=== Time Breakdown ===
+• Index build: {stats['timings'].get('build_index', 0):.2f}s
+• Level 1 filter: {stats['timings'].get('level1_filter', 0):.2f}s
+• Level 2 filter: {stats['timings'].get('level2_filter', 0):.2f}s
+• Relevance check: {stats['timings'].get('relevance_verification', 0):.2f}s
+• File selection: {stats['timings'].get('file_selection', 0):.2f}s
+• Total time: {total_time:.2f}s
+====================================
+"""
+    logger.info(summary)
 
     if args.request_id and not args.skip_events:
         queue_communicate.send_event(
@@ -876,7 +938,7 @@ def build_index_and_filter_files(
                 event_type=CommunicateEventType.CODE_INDEX_FILTER_END.value,
                 data=json.dumps({
                     "filtered_files": stats["final_files"],
-                    "filter_time": stats['level1_filtered'] + stats['level2_filtered'] + stats.get('verified_files', 0)
+                    "filter_time": total_filter_time
                 })
             )
         )

autocoder/version.py

@@ -1 +1 @@
-__version__ = "0.1.212"
+__version__ = "0.1.213"