ayechat-dev 0.36.9.20260204011001__tar.gz → 0.36.9.20260204171331__tar.gz

{ayechat_dev-0.36.9.20260204011001/src/ayechat_dev.egg-info → ayechat_dev-0.36.9.20260204171331}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ayechat-dev
-Version: 0.36.9.20260204011001
+Version: 0.36.9.20260204171331
 Summary: Aye Chat: Terminal-first AI Code Generator
 Author-email: "Acrotron, Inc." <info@acrotron.com>
 License: MIT

{ayechat_dev-0.36.9.20260204011001 → ayechat_dev-0.36.9.20260204171331}/pyproject.toml

@@ -8,7 +8,7 @@ description = "Aye Chat: Terminal-first AI Code Generator"
 readme = "README.md"
 requires-python = ">=3.10, <3.14"
 dependencies = [ "typer>=0.20.0", "httpx>=0.28.1", "keyring>=25.7.0", "prompt-toolkit>=3.0.52", "pathspec>=0.12.1", "chromadb>=1.3.5", "rapidfuzz",]
-version = "0.36.9.20260204011001"
+version = "0.36.9.20260204171331"
 [[project.authors]]
 name = "Acrotron, Inc."
 email = "info@acrotron.com"

ayechat_dev-0.36.9.20260204171331/src/aye/plugins/databricks_model.py

@@ -0,0 +1,311 @@
+import os
+import json
+from typing import Dict, Any, Optional
+import httpx
+from pathlib import Path
+import traceback
+
+from rich import print as rprint
+
+from .plugin_base import Plugin
+from aye.model.config import SYSTEM_PROMPT, MODELS, DEFAULT_MAX_OUTPUT_TOKENS
+from aye.model.auth import get_user_config
+from aye.controller.util import is_truncated_json
+
+LLM_TIMEOUT = 600.0
+
+
+# Message shown when LLM response is truncated due to output token limits
+TRUNCATED_RESPONSE_MESSAGE = (
+    "It looks like my response was cut off because it exceeded the output limit. "
+    "This usually happens when you ask me to generate or modify many files at once.\n\n"
+    "**To fix this, please try:**\n"
+    "1. Break your request into smaller parts (e.g., one file at a time)\n"
+    "2. Use the `with` command to focus on specific files: `with file1.py, file2.py: your request`\n"
+    "3. Ask me to work on fewer files or smaller changes in each request\n\n"
+    "For example, instead of 'update all files to add logging', try:\n"
+    "  `with src/main.py: add logging to this file`"
+)
+
+
+def _get_model_config(model_id: str) -> Optional[Dict[str, Any]]:
+    """Get configuration for a specific model."""
+    for model in MODELS:
+        if model["id"] == model_id:
+            return model
+    return None
+
+
+def _extract_json_object(raw_response: str, prefer_last: bool = True, require_keys=None):
+    """
+    Best-effort extraction of a JSON object (dict) from a raw LLM response.
+
+    Handles common failure modes where the model returns:
+    - extra commentary before/after JSON
+    - multiple JSON objects (e.g., an invalid attempt + a corrected attempt)
+
+    Args:
+        raw_response: raw LLM response string
+        prefer_last: when multiple JSON objects exist, return the last parsed object
+        require_keys: optional iterable of keys; if provided, only consider objects
+                      that contain all of these keys
+
+    Returns:
+        dict or None
+    """
+    # 1) Direct JSON parse
+    try:
+        obj = json.loads(raw_response)
+        if isinstance(obj, dict):
+            if require_keys and not all(k in obj for k in require_keys):
+                return None
+            return obj
+    except Exception:
+        pass
+
+    text = str(raw_response)
+
+    # 2) Scan for balanced JSON object candidates (string/escape-aware)
+    candidates = []
+    depth = 0
+    start = None
+    in_str = False
+    escape = False
+
+    for i, ch in enumerate(text):
+        if in_str:
+            if escape:
+                escape = False
+            elif ch == '\\':
+                escape = True
+            elif ch == '"':
+                in_str = False
+            continue
+
+        # not in string
+        if ch == '"':
+            in_str = True
+            continue
+
+        if ch == '{':
+            if depth == 0:
+                start = i
+            depth += 1
+        elif ch == '}' and depth > 0:
+            depth -= 1
+            if depth == 0 and start is not None:
+                candidates.append(text[start : i + 1])
+                start = None
+
+    parsed = []
+    for cand in candidates:
+        try:
+            obj = json.loads(cand)
+            if not isinstance(obj, dict):
+                continue
+            if require_keys and not all(k in obj for k in require_keys):
+                continue
+            parsed.append(obj)
+        except Exception:
+            continue
+
+    if not parsed:
+        return None
+
+    return parsed[-1] if prefer_last else parsed[0]
+
+
+class DatabricksModelPlugin(Plugin):
+    name = "databricks_model"
+    version = "1.0.0"
+    premium = "free"
+
+    def __init__(self):
+        super().__init__()
+        self.chat_history: Dict[str, list] = {}
+        self.history_file: Optional[Path] = None
+
+    def init(self, cfg: Dict[str, Any]) -> None:
+        """Initialize the local model plugin."""
+        super().init(cfg)
+        if self.debug:
+            rprint(f"[bold yellow]Initializing {self.name} v{self.version}[/]")
+        
+    def _load_history(self) -> None:
+        """Load chat history from disk."""
+        if not self.history_file:
+            if self.verbose:
+                rprint("[yellow]History file path not set for local model. Skipping load.[/]")
+            self.chat_history = {}
+            return
+
+        if self.history_file.exists():
+            try:
+                data = json.loads(self.history_file.read_text(encoding="utf-8"))
+                self.chat_history = data.get("conversations", {})
+            except Exception as e:
+                if self.verbose:
+                    rprint(f"[yellow]Could not load chat history: {e}[/]")
+                self.chat_history = {}
+        else:
+            self.chat_history = {}
+
+    def _save_history(self) -> None:
+        """Save chat history to disk."""
+        if not self.history_file:
+            if self.verbose:
+                rprint("[yellow]History file path not set for local model. Skipping save.[/]")
+            return
+
+        try:
+            self.history_file.parent.mkdir(parents=True, exist_ok=True)
+            data = {"conversations": self.chat_history}
+            self.history_file.write_text(json.dumps(data, indent=2), encoding="utf-8")
+        except Exception as e:
+            if self.verbose:
+                rprint(f"[yellow]Could not save chat history: {e}[/]")
+
+    def _get_conversation_id(self, chat_id: Optional[int] = None) -> str:
+        """Get conversation ID for history tracking."""
+        return str(chat_id) if chat_id and chat_id > 0 else "default"
+
+    def _build_user_message(self, prompt: str, source_files: Dict[str, str]) -> str:
+        """Build the user message with optional source files appended."""
+        user_message = prompt
+        if source_files:
+            user_message += "\n\n--- Source files are below. ---\n"
+            for file_name, content in source_files.items():
+                user_message += f"\n** {file_name} **\n```\n{content}\n```\n"
+        return user_message
+
+    def _parse_llm_response(self, generated_text: str) -> Dict[str, Any]:
+        """Parse LLM response text and convert to expected format."""
+        try:
+            llm_response = json.loads(generated_text)
+        except json.JSONDecodeError:
+            # Check if this looks like a truncated response
+            #if is_truncated_json(generated_text):
+            #    return {
+            #        "summary": TRUNCATED_RESPONSE_MESSAGE,
+            #        "updated_files": []
+            #    }
+            
+            # Not truncated, just malformed - return as plain text
+            llm_response = {
+                "answer_summary": generated_text,
+                "source_files": []
+            }
+        
+        return {
+            "summary": llm_response.get("answer_summary", ""),
+            "updated_files": [
+                {
+                    "file_name": f.get("file_name"),
+                    "file_content": f.get("file_content")
+                }
+                for f in llm_response.get("source_files", [])
+            ]
+        }
+
+    def _create_error_response(self, error_msg: str) -> Dict[str, Any]:
+        """Create a standardized error response."""
+        if self.verbose:
+            rprint(f"[red]{error_msg}[/]")
+        return {
+            "summary": error_msg,
+            "updated_files": []
+        }
+
+    def _handle_databricks(self, prompt: str, source_files: Dict[str, str], chat_id: Optional[int] = None, system_prompt: Optional[str] = None, max_output_tokens: int = DEFAULT_MAX_OUTPUT_TOKENS) -> Optional[Dict[str, Any]]:
+        api_url = os.environ.get("AYE_DBX_API_URL")
+        api_key = os.environ.get("AYE_DBX_API_KEY")
+        model_name = os.environ.get("AYE_DBX_MODEL", "gpt-3.5-turbo")
+        
+        if not api_url or not api_key:
+            return None
+        
+        conv_id = self._get_conversation_id(chat_id)
+        if conv_id not in self.chat_history:
+            self.chat_history[conv_id] = []
+        
+        user_message = self._build_user_message(prompt, source_files)
+        effective_system_prompt = system_prompt if system_prompt else SYSTEM_PROMPT
+        
+        messages_json = [{"role": "system", "content": effective_system_prompt}] + self.chat_history[conv_id] + [{"role": "user", "content": user_message}]
+        messages = messages_json # json.dumps(messages_json)
+        if self.debug:
+            print(">>>>>>>>>>>>>>>>")
+            print(self.chat_history[conv_id])
+            print(">>>>>>>>>>>>>>>>")
+        
+        headers = {"Content-Type": "application/json", "Authorization": f"Bearer {api_key}"}
+        #payload = {"model": model_name, "messages": messages, "temperature": 0.7, "max_tokens": max_output_tokens, "response_format": {"type": "json_object"}}
+        payload = {"model": model_name, "messages": messages, "temperature": 0.7, "max_tokens": max_output_tokens}
+        
+        try:
+            with httpx.Client(timeout=LLM_TIMEOUT) as client:
+                response = client.post(api_url, json=payload, headers=headers)
+                if self.debug:
+                    print("-----------------")
+                    print(response.text)
+                    print("-----------------")
+                response.raise_for_status()
+                result = response.json()
+                if result.get("choices") and result["choices"][0].get("message"):
+                    raw_response = result["choices"][0]["message"]["content"]
+                    generated_json = _extract_json_object(raw_response)
+                    generated_text = json.dumps(generated_json)
+                    if self.debug:
+                        print("-----------------")
+                        print(response.text)
+                        print("-----------------")
+                        print(generated_text)
+                        print("-----------------")
+                    #generated_text = result["choices"][0]["message"]["content"][1]["text"]
+                    self.chat_history[conv_id].append({"role": "user", "content": user_message})
+                    self.chat_history[conv_id].append({"role": "assistant", "content": generated_text})
+                    self._save_history()
+                    #return json.dumps(generated_text)
+                    return self._parse_llm_response(generated_text)
+                return self._create_error_response("Failed to get a valid response from the Databricks API")
+        except httpx.HTTPStatusError as e:
+            traceback.print_exc()
+            error_msg = f"DBX API error: {e.response.status_code}"
+            try:
+                error_detail = e.response.json()
+                if "error" in error_detail:
+                    error_msg += f" - {error_detail['error'].get('message', str(error_detail['error']))}"
+            except: error_msg += f" - {e.response.text[:200]}"
+            return self._create_error_response(error_msg)
+        except Exception as e:
+            traceback.print_exc()
+            return self._create_error_response(f"Error calling Databricks API: {str(e)}")
+
+
+    def on_command(self, command_name: str, params: Dict[str, Any]) -> Optional[Dict[str, Any]]:
+        if command_name == "new_chat":
+            root = params.get("root")
+            history_file = Path(root) / ".aye" / "chat_history.json" if root else Path.cwd() / ".aye" / "chat_history.json"
+            history_file.unlink(missing_ok=True)
+            self.chat_history = {}
+            if self.verbose: rprint("[yellow]Local model chat history cleared.[/]")
+            return {"status": "local_history_cleared"}
+
+        if command_name == "local_model_invoke":
+            prompt = params.get("prompt", "").strip()
+            model_id = params.get("model_id", "")
+            source_files = params.get("source_files", {})
+            chat_id = params.get("chat_id")
+            root = params.get("root")
+            system_prompt = params.get("system_prompt")
+            max_output_tokens = params.get("max_output_tokens", DEFAULT_MAX_OUTPUT_TOKENS)
+
+            self.history_file = Path(root) / ".aye" / "chat_history.json" if root else Path.cwd() / ".aye" / "chat_history.json"
+            self._load_history()
+
+            result = self._handle_databricks(prompt, source_files, chat_id, system_prompt, max_output_tokens)
+            if result is not None: return result
+
+            return None
+
+        return None

{ayechat_dev-0.36.9.20260204011001 → ayechat_dev-0.36.9.20260204171331/src/ayechat_dev.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ayechat-dev
-Version: 0.36.9.20260204011001
+Version: 0.36.9.20260204171331
 Summary: Aye Chat: Terminal-first AI Code Generator
 Author-email: "Acrotron, Inc." <info@acrotron.com>
 License: MIT

{ayechat_dev-0.36.9.20260204011001 → ayechat_dev-0.36.9.20260204171331}/src/ayechat_dev.egg-info/SOURCES.txt

@@ -77,6 +77,7 @@ src/aye/plugins/__init__.py
 src/aye/plugins/at_file_completer.py
 src/aye/plugins/auto_detect_mask.py
 src/aye/plugins/completer.py
+src/aye/plugins/databricks_model.py
 src/aye/plugins/local_model.py
 src/aye/plugins/offline_llm.py
 src/aye/plugins/plugin_base.py