PraisonAI 4.6.142__tar.gz → 4.6.144__tar.gz

{praisonai-4.6.142 → praisonai-4.6.144}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: PraisonAI
-Version: 4.6.142
+Version: 4.6.144
 Summary: PraisonAI is an AI Agents Framework with Self Reflection. PraisonAI application combines PraisonAI Agents, AutoGen, and CrewAI into a low-code solution for building and managing multi-agent LLM systems, focusing on simplicity, customisation, and efficient human-agent collaboration.
 Author: Mervin Praison
 License: MIT
@@ -11,9 +11,9 @@ Description-Content-Type: text/markdown
 Requires-Dist: rich>=13.7
 Requires-Dist: markdown>=3.5
 Requires-Dist: pyparsing>=3.0.0
-Requires-Dist: praisonaiagents>=1.6.145
-Requires-Dist: praisonai-code>=0.0.41
-Requires-Dist: praisonai-bot>=0.0.26
+Requires-Dist: praisonaiagents>=1.6.147
+Requires-Dist: praisonai-code>=0.0.43
+Requires-Dist: praisonai-bot>=0.0.28
 Requires-Dist: python-dotenv>=0.19.0
 Requires-Dist: litellm<2,>=1.83.14
 Requires-Dist: PyYAML>=6.0

{praisonai-4.6.142 → praisonai-4.6.144}/PraisonAI.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: PraisonAI
-Version: 4.6.142
+Version: 4.6.144
 Summary: PraisonAI is an AI Agents Framework with Self Reflection. PraisonAI application combines PraisonAI Agents, AutoGen, and CrewAI into a low-code solution for building and managing multi-agent LLM systems, focusing on simplicity, customisation, and efficient human-agent collaboration.
 Author: Mervin Praison
 License: MIT
@@ -11,9 +11,9 @@ Description-Content-Type: text/markdown
 Requires-Dist: rich>=13.7
 Requires-Dist: markdown>=3.5
 Requires-Dist: pyparsing>=3.0.0
-Requires-Dist: praisonaiagents>=1.6.145
-Requires-Dist: praisonai-code>=0.0.41
-Requires-Dist: praisonai-bot>=0.0.26
+Requires-Dist: praisonaiagents>=1.6.147
+Requires-Dist: praisonai-code>=0.0.43
+Requires-Dist: praisonai-bot>=0.0.28
 Requires-Dist: python-dotenv>=0.19.0
 Requires-Dist: litellm<2,>=1.83.14
 Requires-Dist: PyYAML>=6.0

{praisonai-4.6.142 → praisonai-4.6.144}/PraisonAI.egg-info/SOURCES.txt

@@ -14,7 +14,6 @@ praisonai/_bootstrap.py
 praisonai/_entrypoint.py
 praisonai/_framework_availability.py
 praisonai/_lazy_cache.py
-praisonai/_lockmap.py
 praisonai/_logging.py
 praisonai/_registry.py
 praisonai/_safe_loader.py

{praisonai-4.6.142 → praisonai-4.6.144}/PraisonAI.egg-info/requires.txt

@@ -1,9 +1,9 @@
 rich>=13.7
 markdown>=3.5
 pyparsing>=3.0.0
-praisonaiagents>=1.6.145
-praisonai-code>=0.0.41
-praisonai-bot>=0.0.26
+praisonaiagents>=1.6.147
+praisonai-code>=0.0.43
+praisonai-bot>=0.0.28
 python-dotenv>=0.19.0
 litellm<2,>=1.83.14
 PyYAML>=6.0

{praisonai-4.6.142 → praisonai-4.6.144}/praisonai/auto.py

@@ -10,7 +10,7 @@ This module uses FULL LAZY LOADING for all heavy dependencies:
 
 This ensures minimal import-time overhead.
 """
-from typing import Any, Dict, List, Optional, Type, TypeVar
+from typing import Any, Awaitable, Dict, List, Optional, Type, TypeVar
 import os
 import json
 import asyncio
@@ -620,7 +620,61 @@ class BaseAutoGenerator:
             ImportError: If neither litellm nor openai is installed
         """
         return await self._completion_impl(response_model, messages, is_async=True, **kwargs)
-    
+
+    async def _structured_completion_async(
+        self, response_model: Type[T], messages: List[Dict], *, is_async: bool, **kwargs
+    ) -> T:
+        """Await the structured-completion ladder with an explicit provider path.
+
+        Lets a single generation coroutine be driven either through the async
+        provider clients (``is_async=True``, from ``agenerate()``) or the sync
+        provider clients (``is_async=False``, from the sync ``generate()``
+        bridge). This preserves the historical behavior where a sync
+        ``generate()`` call used the blocking sync client/network stack, while
+        still routing both entry points through one shared body.
+        """
+        return await self._completion_impl(
+            response_model, messages, is_async=is_async, **kwargs
+        )
+
+    def _run_coro_sync(self, coro: Awaitable[T]) -> T:
+        """Drive an async coroutine to completion from a sync caller.
+
+        Single source of truth so the sync ``generate()`` twins can delegate to
+        their ``agenerate()`` implementation instead of duplicating the body.
+
+        Mirrors the running-loop handling in ``_structured_completion``: when no
+        event loop is running we use the shared async bridge; when a loop is
+        already running (FastAPI handler, Jupyter, async test) ``run_sync``
+        would raise, so we drive the coroutine on a dedicated worker thread with
+        its own loop instead.
+        """
+        try:
+            asyncio.get_running_loop()
+        except RuntimeError:
+            running_loop = False
+        else:
+            running_loop = True
+
+        if not running_loop:
+            return run_sync(coro)
+
+        result: List[Any] = []
+        error: List[BaseException] = []
+
+        def _runner() -> None:
+            try:
+                result.append(asyncio.run(coro))
+            except BaseException as e:  # noqa: BLE001 - re-raised on caller thread
+                error.append(e)
+
+        thread = threading.Thread(target=_runner, name="praisonai-sync-generate")
+        thread.start()
+        thread.join()
+        if error:
+            raise error[0]
+        return result[0]
+
     @staticmethod
     def get_available_tools() -> List[str]:
         """Return list of available tools for agent assignment."""
@@ -850,18 +904,8 @@ class AutoGenerator(BaseAutoGenerator):
             path = generator.generate()
             _rich_print(path)
         """
-        response = self._structured_completion(
-            response_model=_get_team_models()['TeamStructure'],
-            messages=[
-                {"role": "system", "content": "You are a helpful assistant designed to output complex team structures."},
-                {"role": "user", "content": self.get_user_content()}
-            ]
-        )
-        json_data = json.loads(response.model_dump_json())
-        self.convert_and_save(json_data, merge=merge)
-        full_path = os.path.abspath(self.agent_file)
-        return full_path
-    
+        return self._run_coro_sync(self._agenerate_impl(merge=merge, is_async=False))
+
     async def agenerate(self, merge=False):
         """
         Async version of generate() - generates a team structure for the specified topic.
@@ -880,12 +924,23 @@ class AutoGenerator(BaseAutoGenerator):
                 path = await gen.agenerate()
                 _rich_print(path)
         """
-        response = await self._astructured_completion(
+        return await self._agenerate_impl(merge=merge, is_async=True)
+
+    async def _agenerate_impl(self, merge=False, *, is_async: bool):
+        """Shared generation body for both sync and async entry points.
+
+        The only divergence is ``is_async``, which selects the async provider
+        clients (``agenerate()``) or the blocking sync clients driven on the
+        bridge (sync ``generate()``), preserving the historical sync provider
+        path while keeping merge/save/path-resolution logic in one place.
+        """
+        response = await self._structured_completion_async(
             response_model=_get_team_models()['TeamStructure'],
             messages=[
                 {"role": "system", "content": "You are a helpful assistant designed to output complex team structures."},
                 {"role": "user", "content": self.get_user_content()}
-            ]
+            ],
+            is_async=is_async,
         )
         json_data = json.loads(response.model_dump_json())
         self.convert_and_save(json_data, merge=merge)
@@ -1354,20 +1409,10 @@ Respond with:
         Returns:
             Path to the generated workflow file
         """
-        response = self._structured_completion(
-            response_model=_get_workflow_models()['WorkflowStructure'],
-            messages=[
-                {"role": "system", "content": "You are a helpful assistant that designs workflow structures."},
-                {"role": "user", "content": self._get_prompt(pattern)}
-            ]
+        return self._run_coro_sync(
+            self._agenerate_impl(pattern=pattern, merge=merge, is_async=False)
         )
-        
-        json_data = json.loads(response.model_dump_json())
-        
-        if merge and os.path.exists(self.workflow_file):
-            return self._save_workflow(self.merge_with_existing_workflow(json_data), pattern)
-        return self._save_workflow(json_data, pattern)
-    
+
     async def agenerate(self, pattern: str = "sequential", merge: bool = False) -> str:
         """
         Async version of generate() - Generate a workflow YAML file.
@@ -1380,12 +1425,25 @@ Respond with:
         Returns:
             Path to the generated workflow file
         """
-        response = await self._astructured_completion(
+        return await self._agenerate_impl(pattern=pattern, merge=merge, is_async=True)
+
+    async def _agenerate_impl(
+        self, pattern: str = "sequential", merge: bool = False, *, is_async: bool
+    ) -> str:
+        """Shared generation body for both sync and async entry points.
+
+        ``is_async`` selects the async provider clients (``agenerate()``) or the
+        blocking sync clients driven on the bridge (sync ``generate()``),
+        preserving the historical sync provider path while keeping merge/save
+        logic in one place.
+        """
+        response = await self._structured_completion_async(
             response_model=_get_workflow_models()['WorkflowStructure'],
             messages=[
                 {"role": "system", "content": "You are a helpful assistant that designs workflow structures."},
                 {"role": "user", "content": self._get_prompt(pattern)}
-            ]
+            ],
+            is_async=is_async,
         )
         
         json_data = json.loads(response.model_dump_json())
@@ -1708,18 +1766,12 @@ class JobWorkflowAutoGenerator(BaseAutoGenerator):
         Returns:
             Path to the generated workflow file
         """
-        prompt = self._get_prompt(include_judge, include_approve)
-        
-        response = self._structured_completion(
-            response_model=_get_job_workflow_models()['JobWorkflowStructure'],
-            messages=[
-                {"role": "system", "content": "You are a helpful assistant that designs job workflow structures with AI agent steps."},
-                {"role": "user", "content": prompt}
-            ]
+        return self._run_coro_sync(
+            self._agenerate_impl(
+                include_judge=include_judge, include_approve=include_approve, is_async=False
+            )
         )
-        
-        return self._save_workflow(response)
-    
+
     async def agenerate(self, include_judge: bool = True, include_approve: bool = False) -> str:
         """
         Async version of generate() - Generate a job workflow YAML file.
@@ -1731,14 +1783,29 @@ class JobWorkflowAutoGenerator(BaseAutoGenerator):
         Returns:
             Path to the generated workflow file
         """
+        return await self._agenerate_impl(
+            include_judge=include_judge, include_approve=include_approve, is_async=True
+        )
+
+    async def _agenerate_impl(
+        self, include_judge: bool = True, include_approve: bool = False, *, is_async: bool
+    ) -> str:
+        """Shared generation body for both sync and async entry points.
+
+        ``is_async`` selects the async provider clients (``agenerate()``) or the
+        blocking sync clients driven on the bridge (sync ``generate()``),
+        preserving the historical sync provider path while keeping prompt/save
+        logic in one place.
+        """
         prompt = self._get_prompt(include_judge, include_approve)
         
-        response = await self._astructured_completion(
+        response = await self._structured_completion_async(
             response_model=_get_job_workflow_models()['JobWorkflowStructure'],
             messages=[
                 {"role": "system", "content": "You are a helpful assistant that designs job workflow structures with AI agent steps."},
                 {"role": "user", "content": prompt}
-            ]
+            ],
+            is_async=is_async,
         )
         
         return self._save_workflow(response)