chore: make cprint write to stderr (#2250)

Also do sys.exit(1) in case of errors

llama_stack/distribution/build.py

@@ -6,6 +6,7 @@
 
 import importlib.resources
 import logging
+import sys
 from pathlib import Path
 
 from pydantic import BaseModel
@@ -95,10 +96,11 @@ def print_pip_install_help(config: BuildConfig):
 
     cprint(
         f"Please install needed dependencies using the following commands:\n\nuv pip install {' '.join(normal_deps)}",
-        "yellow",
+        color="yellow",
+        file=sys.stderr,
     )
     for special_dep in special_deps:
-        cprint(f"uv pip install {special_dep}", "yellow")
+        cprint(f"uv pip install {special_dep}", color="yellow", file=sys.stderr)
     print()
 
 

llama_stack/distribution/client.py

@@ -6,6 +6,7 @@
 
 import inspect
 import json
+import sys
 from collections.abc import AsyncIterator
 from enum import Enum
 from typing import Any, Union, get_args, get_origin
@@ -96,13 +97,13 @@ def create_api_client_class(protocol) -> type:
                             try:
                                 data = json.loads(data)
                                 if "error" in data:
-                                    cprint(data, "red")
+                                    cprint(data, color="red", file=sys.stderr)
                                     continue
 
                                 yield parse_obj_as(return_type, data)
                             except Exception as e:
-                                print(f"Error with parsing or validation: {e}")
-                                print(data)
+                                cprint(f"Error with parsing or validation: {e}", color="red", file=sys.stderr)
+                                cprint(data, color="red", file=sys.stderr)
 
         def httpx_request_params(self, method_name: str, *args, **kwargs) -> dict:
             webmethod, sig = self.routes[method_name]

llama_stack/distribution/library_client.py

@@ -9,6 +9,7 @@ import inspect
 import json
 import logging
 import os
+import sys
 from concurrent.futures import ThreadPoolExecutor
 from enum import Enum
 from pathlib import Path
@@ -210,10 +211,11 @@ class AsyncLlamaStackAsLibraryClient(AsyncLlamaStackClient):
             self.endpoint_impls = None
             self.impls = await construct_stack(self.config, self.custom_provider_registry)
         except ModuleNotFoundError as _e:
-            cprint(_e.msg, "red")
+            cprint(_e.msg, color="red", file=sys.stderr)
             cprint(
                 "Using llama-stack as a library requires installing dependencies depending on the template (providers) you choose.\n",
-                "yellow",
+                color="yellow",
+                file=sys.stderr,
             )
             if self.config_path_or_template_name.endswith(".yaml"):
                 # Convert Provider objects to their types
@@ -234,7 +236,13 @@ class AsyncLlamaStackAsLibraryClient(AsyncLlamaStackClient):
                 cprint(
                     f"Please run:\n\n{prefix}llama stack build --template {self.config_path_or_template_name} --image-type venv\n\n",
                     "yellow",
+                    file=sys.stderr,
                 )
+            cprint(
+                "Please check your internet connection and try again.",
+                "red",
+                file=sys.stderr,
+            )
             raise _e
 
         if Api.telemetry in self.impls:

llama_stack/distribution/utils/exec.py

@@ -8,6 +8,7 @@ import logging
 import os
 import signal
 import subprocess
+import sys
 
 from termcolor import cprint
 
@@ -33,6 +34,7 @@ def formulate_run_args(image_type, image_name, config, template_name) -> list:
             cprint(
                 "No current conda environment detected, please specify a conda environment name with --image-name",
                 color="red",
+                file=sys.stderr,
             )
             return
 
@@ -49,12 +51,13 @@ def formulate_run_args(image_type, image_name, config, template_name) -> list:
                     return envpath
             return None
 
-        print(f"Using conda environment: {env_name}")
+        cprint(f"Using conda environment: {env_name}", color="green", file=sys.stderr)
         conda_prefix = get_conda_prefix(env_name)
         if not conda_prefix:
             cprint(
                 f"Conda environment {env_name} does not exist.",
                 color="red",
+                file=sys.stderr,
             )
             return
 
@@ -63,6 +66,7 @@ def formulate_run_args(image_type, image_name, config, template_name) -> list:
             cprint(
                 f"Build file {build_file} does not exist.\n\nPlease run `llama stack build` or specify the correct conda environment name with --image-name",
                 color="red",
+                file=sys.stderr,
             )
             return
     else:
@@ -73,9 +77,10 @@ def formulate_run_args(image_type, image_name, config, template_name) -> list:
             cprint(
                 "No current virtual environment detected, please specify a virtual environment name with --image-name",
                 color="red",
+                file=sys.stderr,
             )
             return
-        print(f"Using virtual environment: {env_name}")
+        cprint(f"Using virtual environment: {env_name}", file=sys.stderr)
 
     script = importlib.resources.files("llama_stack") / "distribution/start_stack.sh"
     run_args = [