TalkPipe 0.4.1__tar.gz → 0.5.0__tar.gz

{talkpipe-0.4.1 → talkpipe-0.5.0}/.github/ISSUE_TEMPLATE/release-checklist.md

@@ -14,6 +14,7 @@ assignees: ''
 - [ ] Build docker image
 - [ ] Tag repository
 - [ ] Build whl files
+- [ ] Install whl file in new environment and test a script in chatterlang_server
 - [ ] Push to pypi
 - [ ] Push code to repo
 - [ ] Push tags to repo

{talkpipe-0.4.1 → talkpipe-0.5.0}/.vscode/settings.json

@@ -1,9 +1,4 @@
 {
-    "python.testing.pytestArgs": [
-        "tests",
-        "-m",
-        "not ddg"
-    ],
     "python.testing.unittestEnabled": false,
     "python.testing.pytestEnabled": true,
     "terminal.integrated.env.osx": {

talkpipe-0.5.0/CHANGELOG.md

@@ -0,0 +1,85 @@
+# Changelog
+
+## in development
+### New Segments and Sources
+ - lambda expression integration
+   - **lambda** - lets one write lambda expressions with limited, but common data manipulation segments.  This will
+  eliminate the need for most small, simple segments, even gt, lt, etc.
+   - **lambdaFilter** - filters based on a lambda expression.  Uses the same syntax as lambda.
+  - **readEmail** - source for reading email from an IMAP server, along with helper functions
+  - **sign** and **verify** for signing and verifying data.  Also added utility functions for creating keys and associated 
+   functions.
+ - **jsonReceiver** - both a source and an app for opening an endpoint that can receive json and pass it through a 
+ chatterlang script
+
+### Improvements
+ - Broad support for configuration and environment variables.  When a parameter to a segment or source starts with a dollar sign (e.g. $myval),
+ the system will look first in the current configuration for the key "myval" and then in the OS' environment variables for TALKPIPE_myval.
+ - Added a safe lambda expression compiler to data_manipulation.  Enables only basic data manipulation functionality
+ - Use can specify custom user-agent strings via a configuration file or environment variable for web page
+ downloader.
+ - Renamed threading.py to thread_ops.py to avoid name collision
+ - Renamed scriptendpoint.py to chatterlang_server.py for clarity
+ - Documentation improvements
+ - Moved code closing mongo connections to the end of the internal loop and out of the del code.  It was causing 
+error when being disposed as python was exiting.
+ - Changed n_iter to max_iter to resolve warning with tsne.
+ - Adjust n_neighbors in umap when larger than the number of points to resolve warning
+ - Added chatterlang_scipt to the package installer so the user can run talkpipe.app.runscript from the commandline
+
+## 0.4.2 
+### New Segments
+ - snippet - lets the script writer either load chatterlang code from a file or provide it as a parameter. 
+ - Abstracted the comparison filter used for gt and lt
+ - Added filters for gte (greater than or equal), lte (less than or equal), eq (equal) and neq (not equal)
+### Improvements
+ - Reconfigured the Docker containers so that there is a base container with the pre-requisites and one for talkpipe.
+ This reduces the time needed to build a talkpipe container if the base already exists.
+ - Retooled unit tests to eliminate the "online" fixture, replacing it with something that checks if ollama and mongo are
+ available locally and skipping tests that require them if they are not available.
+### Bug Fixes
+ - Fixed error in rss.py causing a crash if the URL was specified in the configuration file or environment variable
+ - Handled code where robots.txt response is returned compressed.  It now decompresses it rather than erroring out.
+
+
+
+## 0.4.1
+- Updated syntax so that pipelines in forks can contain more than one segment.  This was an oversight in the original design.
+- Added the mongoInsert and mongoSearch segments
+- Generalized LLM source so that new ones can be registered and used without modifying TalkPipe
+- Updated the Embedding API to match the chat api (with sources, etc)
+- Added ability to create and reduce matrices using UMAP and t-SNE
+- Added template segment for populating a string template
+- Update chatterlang so that a double-double quote ("") gets resolved to one double quote
+- Fixed bug in extract_property that would cause it to fail seeing some properties
+- Improved unit testing and coverage
+- Updated the guided generation classes to make it easier to do guided generations problems in talkpipe
+- Split the talkpipe.util module into a talkpipe.util subpackage with different modules for different types of utilities.
+- Updated the logo
+- Improved and refined the wording in the README
+- Move the documentation script into apps and added a script so you can generate documentation files when talkpipe is installed
+
+## 0.3.1
+- jupyter notebook widgets
+- a hash segment
+
+## 0.3.0
+- Basic web app for writing and executing ChatterLang scripts, along with endpoints for the underlying functionality.  Includes the ability to:
+  - write scripts in a text area and compile/execute them.
+  - interactive provide text input to the script if the script requires it.
+  - the ability to see logging statements emitted.
+  - run the server using a command "chatterlang_server" that is included in the whl file.
+  - view documentation for the include sources and segments.  Note that the documentation is rebuilt
+  everytime a whl file is generated.
+  - view and load various example scripts.
+installed.  This works on at least windows and linux
+- Added comment support for ChatterLang, anything from a hash mark (#) until the end of the line.
+- Added configureLogging segment for configuring logging levels and files from within a script.  This is probably temporary and will be removed before 0.3.0 as a better solution is written.
+- Consolidated the logic around configuring logging, migrating the various ways of specifying the configuration strings from runscript into the configure_logging method itself.
+
+## 0.2.2 (2025-02-21)
+- The rss segment can get the url to download from the rss_url configuration setting.
+- Updated OpenAI source so that it can support guided generation (and thus the llmScore segment)
+
+## 0.2.1 (2025-02-21)
+- Catch all ConnectionErrors when downloading the robots.txt file.

{talkpipe-0.4.1 → talkpipe-0.5.0}/Dockerfile

@@ -1,4 +1,4 @@
-FROM python:3.11
+FROM talkpipe:base
 
 WORKDIR /app
 
@@ -17,10 +17,8 @@ COPY .git/ .git/
 
 RUN pip install --upgrade pip
 RUN pip install -e .[dev]
-RUN pytest -m "not online" --log-cli-level=DEBUG
-# RUN pip install -e .
-
-# Install the basics of the scientific computing stack for analytics
-RUN pip install numpy pandas matplotlib scikit-learn scipy
+RUN pytest --log-cli-level=DEBUG
+RUN rm -fr .git
+RUN rm -fr tests
 
 CMD ["python", "-m", "talkpipe.app.runscript", "--load_module", "data/custom_module.py", "--env", "TALKPIPE_SCRIPT"]

{talkpipe-0.4.1 → talkpipe-0.5.0}/PKG-INFO

@@ -1,14 +1,14 @@
 Metadata-Version: 2.4
 Name: talkpipe
-Version: 0.4.1
+Version: 0.5.0
 Summary: Python internal and external DSL for writing generative AI analytics
 Author-email: Travis Bauer <tlbauer@sandia.gov>
 Maintainer-email: Travis Bauer <tlbauer@sandia.gov>
+License-Expression: Apache-2.0
 Project-URL: Homepage, https://github.com/sandialabs/talkpipe
 Keywords: ai
 Classifier: Development Status :: 3 - Alpha
 Classifier: Programming Language :: Python :: 3
-Classifier: License :: OSI Approved :: Apache Software License
 Classifier: Operating System :: OS Independent
 Requires-Python: >=3.11
 Description-Content-Type: text/markdown
@@ -31,6 +31,8 @@ Requires-Dist: ipywidgets
 Requires-Dist: pymongo
 Requires-Dist: umap-learn
 Requires-Dist: scikit-learn
+Requires-Dist: cryptography
+Requires-Dist: uvicorn
 Provides-Extra: dev
 Requires-Dist: pytest; extra == "dev"
 Requires-Dist: pytest-mock; extra == "dev"

talkpipe-0.5.0/base-image/Dockerfile

@@ -0,0 +1,19 @@
+FROM python:3.11
+
+# Set working directory
+WORKDIR /app
+
+# Install PDM
+RUN pip install pdm
+
+# Copy pyproject.toml into the container
+COPY pyproject.toml ./pyproject.toml
+
+# Run pdm lock, export deps to requirements.txt, and install
+RUN pdm lock && \
+    pdm export --without-hashes --dev -o requirements.txt && \
+    pip install -r requirements.txt && \
+    rm requirements.txt
+
+# Install the basics of the scientific computing stack for analytics
+RUN pip install numpy pandas matplotlib scikit-learn scipy

{talkpipe-0.4.1 → talkpipe-0.5.0}/pyproject.toml

@@ -14,10 +14,10 @@ description = "Python internal and external DSL for writing generative AI analyt
 readme = "README.md"
 requires-python = ">=3.11"
 keywords = ["ai"]
+license = "Apache-2.0"
 classifiers = [
     "Development Status :: 3 - Alpha",
     "Programming Language :: Python :: 3",
-    "License :: OSI Approved :: Apache Software License",
     "Operating System :: OS Independent",
 ]
 dependencies = [
@@ -38,7 +38,9 @@ dependencies = [
     'ipywidgets',
     'pymongo',
     'umap-learn',
-    'scikit-learn'
+    'scikit-learn',
+    'cryptography',
+    'uvicorn'
 ]
 dynamic = ["version"]
 
@@ -51,8 +53,11 @@ dev = [
 ]
 
 [project.scripts]
-chatterlang_server = "talkpipe.app.scriptendpoint:main"
+chatterlang_server = "talkpipe.app.chatterlang_server:main"
+chatterlang_script = "talkpipe.app.runscript:main"
 talkpipe_ref = "talkpipe.app.unit_documentation_analyzer:go"
+talkpipe_genkeys = "talkpipe.operations.crypto:generate_keys_cli"
+talkpipe_endpoint = "talkpipe.app.apiendpoint:go"
 
 [tool.setuptools.packages.find]
 where = ["src"]
@@ -66,4 +71,4 @@ where = ["src"]
 #use defaults
 
 [project.urls]
-Homepage = "https://github.com/sandialabs/talkpipe"
+Homepage = "https://github.com/sandialabs/talkpipe"

talkpipe-0.5.0/pytest.ini

@@ -0,0 +1,3 @@
+[pytest]
+log_level = DEBUG
+

{talkpipe-0.4.1 → talkpipe-0.5.0}/src/talkpipe/__init__.py

@@ -13,3 +13,5 @@ from talkpipe.app.chatcli import *
 from talkpipe.operations.filtering import *
 from talkpipe.operations.transforms import *
 from talkpipe.operations.matrices import *
+from talkpipe.operations.signatures import *
+from talkpipe.app.apiendpoint import *

talkpipe-0.5.0/src/talkpipe/app/apiendpoint.py

@@ -0,0 +1,407 @@
+"""
+FastAPI JSON Receiver Server
+Receives JSON data via HTTP and processes it with a configurable function
+"""
+from typing import Union
+import logging
+import argparse
+from fastapi.responses import FileResponse
+from fastapi.staticfiles import StaticFiles
+from fastapi import FastAPI, HTTPException, Depends, Header
+from fastapi.middleware.cors import CORSMiddleware
+from fastapi.responses import HTMLResponse
+from pydantic import BaseModel
+from typing import Optional, List, Any, Dict, Callable
+from datetime import datetime
+import uvicorn
+import json
+from pathlib import Path
+import threading
+from queue import Queue
+from talkpipe.pipe.core import AbstractSource
+from talkpipe.chatterlang import register_source
+from talkpipe.chatterlang import compile
+from talkpipe.util.config import get_config
+
+logger = logging.getLogger(__name__)
+
+# Models
+class DataResponse(BaseModel):
+    status: str
+    message: str
+    data: Dict[str, Any]
+    timestamp: datetime
+
+class DataHistory(BaseModel):
+    entries: List[dict]
+    count: int
+
+class JSONReceiver:
+    """JSON Data Receiver Server Class - supports multiple instances"""
+    
+    def __init__(
+        self,
+        host: str = "0.0.0.0",
+        port: int = 9999,
+        api_key: str = "your-secret-key-here",
+        require_auth: bool = False,
+        processor_func: Callable[[Dict[str, Any]], Any] = None,
+        title: str = "JSON Data Receiver",
+        history_length: int = 1000
+    ):
+        self.host = host
+        self.port = port
+        self.api_key = api_key
+        self.require_auth = require_auth
+        self.history = []
+        self.title = title
+        self.history_length = history_length
+        
+        # Set up processor function
+        self.processor_function = processor_func or self._default_print_processor
+        
+        # Create FastAPI app instance
+        self.app = FastAPI(title=f"{title} (Port {port})", version="1.0.0")
+
+        # Mount favicon.ico directly to root
+        @self.app.get("/favicon.ico")
+        async def favicon():
+            favicon_path = Path(__file__).parent / 'static' / 'favicon.ico'
+            if favicon_path.exists():
+                return FileResponse(favicon_path)
+            else:
+                raise HTTPException(status_code=404, detail="Favicon not found")
+
+        
+        # Configure middleware
+        self._setup_middleware()
+        
+        # Configure routes
+        self._setup_routes()
+        
+        # Server instance for stopping
+        self.server = None
+        self.server_thread = None
+    
+    def _setup_middleware(self):
+        """Configure CORS middleware"""
+        self.app.add_middleware(
+            CORSMiddleware,
+            allow_origins=["*"],
+            allow_credentials=True,
+            allow_methods=["*"],
+            allow_headers=["*"],
+        )
+    
+    def _setup_routes(self):
+        """Configure all API routes"""
+        
+        @self.app.get("/", response_class=HTMLResponse)
+        async def root():
+            return self._get_html_interface()
+        
+        @self.app.post("/process", response_model=DataResponse)
+        async def process_json(data: Dict[str, Any], api_key: str = Depends(self._verify_api_key)):
+            return await self._process_json(data)
+        
+        @self.app.get("/history", response_model=DataHistory)
+        async def get_history(limit: int = 50, api_key: str = Depends(self._verify_api_key)):
+            return self._get_history(limit)
+        
+        @self.app.delete("/history")
+        async def clear_history(api_key: str = Depends(self._verify_api_key)):
+            return self._clear_history()
+        
+        @self.app.get("/health")
+        async def health_check():
+            return {"status": "healthy", "timestamp": datetime.now(), "port": self.port}
+    
+    async def _verify_api_key(self, x_api_key: Optional[str] = Header(None)):
+        """Dependency for API key validation"""
+        if self.require_auth and x_api_key != self.api_key:
+            raise HTTPException(status_code=403, detail="Invalid API key")
+        return x_api_key
+    
+    def _default_print_processor(self, data: Dict[str, Any]) -> str:
+        """Default processor: print JSON data to stdout"""
+        try:
+            print(f"[Port {self.port}] Received JSON: {json.dumps(data, indent=2, default=str)}")
+            return "Data printed successfully"
+        except Exception as e:
+            logger.error(f"Error processing data: {e}")
+            raise HTTPException(status_code=500, detail=f"Processing error: {e}")
+    
+    def _save_to_history(self, data: Dict[str, Any]):
+        self.history.append(data)
+        while len(self.history) > self.history_length:
+            self.history.pop(0)
+    
+    async def _process_json(self, data: Dict[str, Any]) -> DataResponse:
+        """Process JSON data with the configured processor function"""
+        
+        # Log the request
+        logger.info(f"[{datetime.now().strftime('%H:%M:%S')}] Port {self.port}: Processing JSON data")
+        
+        try:
+            # Process the data
+            result = self.processor_function(data)
+            
+            # Save to history
+            self._save_to_history(data)
+            
+            # Determine message based on result
+            if isinstance(result, str):
+                message = result
+            else:
+                message = "Data processed successfully"
+            
+            return DataResponse(
+                status="success",
+                message=message,
+                data=data,
+                timestamp=datetime.now()
+            )
+            
+        except HTTPException:
+            # Re-raise HTTP exceptions from the processor
+            raise
+        except Exception as e:
+            logger.error(f"Error processing data: {e}")
+            raise HTTPException(status_code=500, detail=f"Processing failed: {str(e)}")
+    
+    def _get_history(self, limit: int) -> DataHistory:
+        """Get processing history"""
+        return DataHistory(entries=self.history, count=len(self.history))
+    
+    def _clear_history(self):
+        """Clear processing history"""
+        self.history.clear()
+        logger.info(f"Port {self.port}: History cleared")
+        return {"status": "success", "message": "History cleared", "timestamp": datetime.now()}
+    
+    def _get_html_interface(self) -> str:
+        """Get the HTML interface with port-specific URLs"""
+        return f"""
+        <!DOCTYPE html>
+        <html>
+        <head>
+            <title>{self.title} - Port {self.port}</title>
+            <link rel="icon" type="image/ico" href="/favicon.ico">
+            <style>
+                body {{ font-family: Arial, sans-serif; max-width: 800px; margin: 50px auto; padding: 20px; }}
+                input, button, textarea {{ font-size: 16px; padding: 10px; margin: 5px; }}
+                textarea {{ width: 500px; height: 150px; font-family: monospace; }}
+                .status {{ margin-top: 20px; padding: 10px; border-radius: 5px; }}
+                .success {{ background-color: #d4edda; color: #155724; }}
+                .error {{ background-color: #f8d7da; color: #721c24; }}
+                .history {{ margin-top: 30px; }}
+                .history-item {{ padding: 5px; border-bottom: 1px solid #ddd; }}
+                code {{ background-color: #f4f4f4; padding: 2px 5px; border-radius: 3px; }}
+                .port-info {{ background-color: #e3f2fd; padding: 10px; border-radius: 5px; margin-bottom: 20px; }}
+            </style>
+        </head>
+        <body>
+            <div class="port-info">
+                <h1>{self.title}</h1>
+                <p><strong>Server Port:</strong> {self.port}</p>
+                <p><strong>History Length:</strong> {len(self.history)}, max length={self.history_length}</p>
+            </div>
+            
+            <p>Send JSON data to process it on this machine:</p>
+            
+            <div>
+                <textarea id="jsonData" placeholder='{{"example": "data", "number": 42}}'></textarea><br>
+                <button onclick="sendData()">Send JSON Data</button>
+            </div>
+            
+            <div id="status"></div>
+            
+            <div class="history">
+                <h3>API Endpoints:</h3>
+                <ul>
+                    <li><code>POST /process</code> - Send JSON data for processing</li>
+                    <li><code>GET /history</code> - Get processing history</li>
+                    <li><code>DELETE /history</code> - Clear processing history</li>
+                    <li><code>GET /health</code> - Health check</li>
+                    <li><code>GET /docs</code> - API documentation</li>
+                </ul>
+                
+                <h3>Usage Examples:</h3>
+                <pre>
+# Using curl:
+curl -X POST http://localhost:{self.port}/process \\
+  -H "Content-Type: application/json" \\
+  -d '{{"message": "hello", "value": 123}}'
+
+# Using PowerShell:
+Invoke-RestMethod -Uri http://localhost:{self.port}/process \\
+  -Method POST -ContentType "application/json" \\
+  -Body '{{"key": "value", "number": 42}}'
+
+# Using Python:
+import requests
+response = requests.post('http://localhost:{self.port}/process', 
+  json={{'key': 'value', 'number': 42}})
+print(response.json())</pre>
+            </div>
+            
+            <script>
+                async function sendData() {{
+                    const jsonText = document.getElementById('jsonData').value;
+                    const status = document.getElementById('status');
+                    
+                    try {{
+                        const data = JSON.parse(jsonText);
+                        
+                        const response = await fetch('/process', {{
+                            method: 'POST',
+                            headers: {{'Content-Type': 'application/json'}},
+                            body: JSON.stringify(data)
+                        }});
+                        
+                        const result = await response.json();
+                        
+                        if (response.ok) {{
+                            status.className = 'status success';
+                            status.textContent = 'JSON data processed successfully on port {self.port}!';
+                        }} else {{
+                            status.className = 'status error';
+                            status.textContent = 'Error: ' + result.detail;
+                        }}
+                    }} catch (error) {{
+                        status.className = 'status error';
+                        status.textContent = 'Error: ' + error.message;
+                    }}
+                    
+                    setTimeout(() => status.textContent = '', 3000);
+                }}
+            </script>
+        </body>
+        </html>
+        """
+    
+    def set_processor_function(self, func: Callable[[Dict[str, Any]], Any]):
+        """Set the function used to process incoming JSON data"""
+        self.processor_function = func
+        logger.info(f"Port {self.port}: Processor function set to: {func.__name__}")
+    
+    def start(self, background: bool = False):
+        """Start the FastAPI server"""
+        logger.info(f"Starting JSON Data Receiver on http://{self.host}:{self.port}")
+        logger.info(f"API documentation: http://{self.host}:{self.port}/docs")
+        if self.require_auth:
+            logger.info(f"Port {self.port}: API key authentication: ENABLED")
+        
+        processor_name = self.processor_function.__name__ 
+        logger.info(f"Port {self.port}: Using processor function: {processor_name}")
+        
+        if background:
+            self.server_thread = threading.Thread(
+                target=lambda: uvicorn.run(self.app, host=self.host, port=self.port),
+                daemon=True
+            )
+            self.server_thread.start()
+            logger.info(f"Port {self.port}: Server started in background thread")
+        else:
+            uvicorn.run(self.app, host=self.host, port=self.port)
+    
+    def stop(self):
+        """Stop the server (only works if started in background)"""
+        if self.server_thread and self.server_thread.is_alive():
+            logger.info(f"Port {self.port}: Stopping server...")
+            # Note: uvicorn doesn't provide a clean stop method when run this way
+            # For production use, consider using uvicorn.Server with proper lifecycle management
+
+@register_source("jsonReceiver")
+class JSONReceiverSegment(AbstractSource):
+    """Segment for receiving JSON data via FastAPI"""
+    
+    def __init__(self, port: Union[int,str] = 9999, host: str = "0.0.0.0", 
+                 api_key: str = None, require_auth: bool = False):
+        super().__init__()
+        self.port = int(port)
+        self.host = host
+        self.api_key = api_key
+        self.require_auth = require_auth
+        self.queue = Queue(maxsize=1000)
+        
+        self.receiver = JSONReceiver(
+            host=host,
+            port=self.port,
+            api_key=api_key,
+            require_auth=require_auth,
+            title=f"JSON Receiver Segment (Port {port})",
+            processor_func=self.process_data
+        )
+        self.receiver.start(background=True)
+        logger.info(f"Finished initializing JSONReceiverSegment on port {port}")
+
+    def process_data(self, data: Dict[str, Any]) -> str:
+        """Process incoming JSON data and add it to the queue"""
+        try:
+            self.queue.put(data, block=True, timeout=60)
+            return f"Data received and queued: {data}"
+        except Exception as e:
+            logger.error(f"Error processing data: {e}")
+            return f"Error processing data: {str(e)}"
+        
+    def generate(self):
+        print("Starting JSONReceiverSegment generator...")
+        while True:
+            # Wait for data to be available in the queue
+            print("Waiting for data...")
+            # This will block until data is available
+            yield self.queue.get(block=True, timeout=None)
+
+def go():
+    
+    parser = argparse.ArgumentParser(description='FastAPI JSON Data Receiver')
+    parser.add_argument('-p', '--port', type=int, default=9999,
+                        help='Port to listen on (default: 9999)')
+    parser.add_argument('-o', '--host', default='0.0.0.0',
+                        help='Host to bind to (default: 0.0.0.0)')
+    parser.add_argument('--api-key', help='Set API key for authentication')
+    parser.add_argument('--require-auth', action='store_true',
+                        help='Require API key authentication')
+    parser.add_argument('--title', default='JSON Data Receiver',
+                        help='Title for the FastAPI application')
+    parser.add_argument('--script-var', default=None, 
+                        help='Configuration key containing script to run')
+    parser.add_argument('--script', default=None,
+                        help='Chatterlang script to run on received data')
+    
+    args = parser.parse_args()
+    
+
+    script = None
+    if args.script_var and args.script:
+        raise ValueError("Cannot specify both --script-var and --script. Use one or the other.")
+    if args.script_var:
+        script = get_config(args.script_var)
+        if not script:
+            raise ValueError(f"No script found for variable '{args.script_var}' in configuration.")
+    else:
+        script = args.script
+    
+    if script:
+        # Compile the Chatterlang script
+        compiled_script = compile(script)
+        compiled_script = compiled_script.asFunction(single_in=True, single_out=False)
+    else:
+        # Default processor function if no script is provided
+        compiled_script = None
+
+    receiver = JSONReceiver(
+        host=args.host,
+        port=args.port,
+        api_key=args.api_key,
+        require_auth=args.require_auth,
+        title=args.title,
+        processor_func=compiled_script,
+    )
+
+    # Start the server
+    receiver.start(background=False)
+
+if __name__ == "__main__":
+    go()

talkpipe-0.4.1/src/talkpipe/app/scriptendpoint.py → talkpipe-0.5.0/src/talkpipe/app/chatterlang_server.py

@@ -1011,7 +1011,7 @@ def main():
 
     print(f"Starting TalkPipe server at http://{args.host}:{args.port}")
     uvicorn.run(
-        "talkpipe.app.scriptendpoint:app",
+        "talkpipe.app.chatterlang_server:app",
         host=args.host,
         port=args.port,
         reload=args.reload