Streaming response fix & docs

README.md

@@ -23,8 +23,7 @@ Free [plan](https://github.com/features/copilot#pricing) includes up to 2,000 co
 
 ### Auth Command Help
 
-```
-
+```bash
 $ llm github_copilot auth --help
 Usage: llm github_copilot auth [OPTIONS] COMMAND [ARGS]...
 
@@ -54,6 +53,7 @@ When you run the login command, the plugin will:
 Example login output:
 
 ```
+$ llm github_copilot auth login --help
 Usage: llm github_copilot auth login [OPTIONS]
 
   Authenticate with GitHub Copilot to generate a new access token.
@@ -80,11 +80,10 @@ llm github_copilot auth login --force
 ```
 
 You can also perform the login process and display the obtained access token and API key without saving them to the LLM keystore or the API key file. This is useful if you want to manually manage these tokens or use them in environment variables:
+
 ```bash
-llm github_copilot auth login --show-only
-```
-This will output something like:
-```
+$ llm github_copilot auth login --show-only
+
 GitHub Copilot: ✓ Authenticated
           User: your_github_username
    AccessToken: Valid
@@ -96,14 +95,11 @@ You can set GH_COPILOT_TOKEN or GITHUB_COPILOT_TOKEN to the token value above.
 Note: These tokens have NOT been saved to the LLM keystore or API key file.
 ```
 
-### Authentication Status
-```
-
 ### Authentication Status
 
 You can check your authentication status with:
 
-```
+```bash
 $ llm github_copilot auth status
 GitHub Copilot: ✓ Authenticated
        API Key: Valid, expires 2025-07-15 10:30:00
@@ -112,14 +108,15 @@ GitHub Copilot: ✓ Authenticated
 
 For more detailed information, use the verbose flag:
 
-```
+```bash
 $ llm github_copilot auth status --verbose
 GitHub Copilot: ✓ Authenticated
           User: testuser
    AccessToken: Valid, via keystore github_copilot_access_token
    AccessToken: ghu_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
        API Key: Valid, expires 2025-07-15 10:30:00
        API key: abcdef1234567890abcdef1234567890
+
 ```
 
 ## Usage

llm_github_copilot.py

@@ -41,8 +41,12 @@ def _fetch_models_data(authenticator: "GitHubCopilotAuthenticator") -> dict:
         response.raise_for_status()
         return response.json()
     except httpx.HTTPStatusError as e:
+        # Ensure the response body is read before accessing .text,
+        # in case of chunked error responses.
+        e.response.read()
+        error_body_text = e.response.text
         print(
-            f"Error fetching models data (HTTP {e.response.status_code}): {e.response.text}"
+            f"Error fetching models data (HTTP {e.response.status_code}): {error_body_text}"
         )
         raise
     except httpx.RequestError as e:
@@ -743,7 +747,10 @@ def _non_streaming_request(
             yield api_response.text
 
         except httpx.HTTPStatusError as e:
-            error_text = f"HTTP error {e.response.status_code}: {e.response.text}"
+            # Ensure the response body is read before accessing .text
+            e.response.read()
+            error_body_text = e.response.text
+            error_text = f"HTTP error {e.response.status_code}: {error_body_text}"
             print(error_text)
 
             yield error_text
@@ -970,7 +977,11 @@ def _handle_streaming_request(
                                     yield data
 
         except httpx.HTTPStatusError as e:
-            error_msg = f"HTTP error {e.response.status_code}: {e.response.text}"
+            # Read the response body before trying to access .text
+            # This is crucial for streaming responses that error out
+            e.response.read()
+            error_body_text = e.response.text
+            error_msg = f"HTTP error {e.response.status_code}: {error_body_text}"
             print(error_msg)
             # Print more detailed error information
             print(f"Request headers: {headers}")