ngpt 2.14.0__tar.gz → 2.15.0__tar.gz

{ngpt-2.14.0 → ngpt-2.15.0}/.gitignore

@@ -11,3 +11,5 @@ wheels/
 
 # Environment variables
 .env
+# Reference files
+ref-*

{ngpt-2.14.0 → ngpt-2.15.0}/CONTRIBUTING.md

@@ -30,8 +30,10 @@ Thank you for your interest in contributing to NGPT! This document provides guid
     - `formatters.py` - Output formatting logic
     - `interactive.py` - Interactive mode implementation
     - `modes/` - Different operational modes
+      - `__init__.py` - Modes initialization
       - `chat.py` - Chat mode functionality
       - `code.py` - Code generation mode
+      - `gitcommsg.py` - Git commit message generation
       - `shell.py` - Shell command generation
       - `text.py` - Text generation mode
       - `rewrite.py` - Text rewriting mode

{ngpt-2.14.0 → ngpt-2.15.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ngpt
-Version: 2.14.0
+Version: 2.15.0
 Summary: A lightweight Python CLI and library for interacting with OpenAI-compatible APIs, supporting both official and self-hosted LLM endpoints.
 Project-URL: Homepage, https://github.com/nazdridoy/ngpt
 Project-URL: Repository, https://github.com/nazdridoy/ngpt
@@ -154,6 +154,13 @@ ngpt --log "Tell me about quantum computing"
 
 # Process text from stdin using the {} placeholder
 cat README.md | ngpt --stdin "Summarize this document: {}"
+
+# Use different model providers by specifying the provider name
+ngpt --provider Groq "Explain quantum computing"
+
+# Compare outputs from different providers
+ngpt --provider OpenAI "Explain quantum physics" > openai_response.txt
+ngpt --provider Ollama "Explain quantum physics" > ollama_response.txt
 ```
 
 For more examples and detailed usage, visit the [CLI Usage Guide](https://nazdridoy.github.io/ngpt/usage/cli_usage.html).
@@ -162,7 +169,7 @@ For more examples and detailed usage, visit the [CLI Usage Guide](https://nazdri
 
 - ✅ **Versatile**: Use as a CLI tool, Python library, or CLI framework for building custom tools
 - 🪶 **Lightweight**: Minimal dependencies with everything you need included
-- 🔄 **API Flexibility**: Works with OpenAI, Ollama, Groq, and any compatible endpoint
+- 🔄 **API Flexibility**: Works with OpenAI, Ollama, Groq, Claude, and any compatible endpoint
 - 💬 **Interactive Chat**: Continuous conversation with memory in modern UI
 - 📊 **Streaming Responses**: Real-time output for better user experience
 - 🔍 **Web Search**: Integrated with compatible API endpoints
@@ -179,6 +186,8 @@ For more examples and detailed usage, visit the [CLI Usage Guide](https://nazdri
 - 📃 **Conversation Logging**: Save your conversations to text files for later reference
 - 🧰 **CLI Components**: Reusable components for building custom AI-powered command-line tools
 - 🔌 **Modular Architecture**: Well-structured codebase with clean separation of concerns
+- 🔄 **Provider Switching**: Easily switch between different LLM providers with a single parameter
+- 🚀 **Performance Optimized**: Fast response times and minimal resource usage
 
 See the [Feature Overview](https://nazdridoy.github.io/ngpt/overview.html) for more details.
 
@@ -327,6 +336,16 @@ print(result.stdout)
 # Returns only code without markdown or explanations
 code = client.generate_code("function that converts Celsius to Fahrenheit")
 print(code)
+
+# Compare responses from different providers
+openai_config = load_config(config_index=0)  # OpenAI
+groq_config = load_config(config_index=1)    # Groq
+
+openai_client = NGPTClient(**openai_config)
+groq_client = NGPTClient(**groq_config)
+
+openai_response = openai_client.chat("Explain quantum computing")
+groq_response = groq_client.chat("Explain quantum computing")
 ```
 
 For advanced usage patterns and integrations, check out the [Advanced Examples](https://nazdridoy.github.io/ngpt/examples/advanced.html).
@@ -392,6 +411,7 @@ You can configure nGPT using the following options:
 | `-t, --text` | Open interactive multiline editor for complex prompts with syntax highlighting |
 | `--stdin` | Read from stdin and use content with prompt. Use {} in prompt as placeholder for stdin content |
 | `--rewrite` | Rewrite text to improve quality while preserving original tone and meaning |
+| `--gitcommsg` | Generate AI-powered git commit messages from staged changes or diff files |
 
 #### Global Options
 

{ngpt-2.14.0 → ngpt-2.15.0}/README.md

@@ -118,6 +118,13 @@ ngpt --log "Tell me about quantum computing"
 
 # Process text from stdin using the {} placeholder
 cat README.md | ngpt --stdin "Summarize this document: {}"
+
+# Use different model providers by specifying the provider name
+ngpt --provider Groq "Explain quantum computing"
+
+# Compare outputs from different providers
+ngpt --provider OpenAI "Explain quantum physics" > openai_response.txt
+ngpt --provider Ollama "Explain quantum physics" > ollama_response.txt
 ```
 
 For more examples and detailed usage, visit the [CLI Usage Guide](https://nazdridoy.github.io/ngpt/usage/cli_usage.html).
@@ -126,7 +133,7 @@ For more examples and detailed usage, visit the [CLI Usage Guide](https://nazdri
 
 - ✅ **Versatile**: Use as a CLI tool, Python library, or CLI framework for building custom tools
 - 🪶 **Lightweight**: Minimal dependencies with everything you need included
-- 🔄 **API Flexibility**: Works with OpenAI, Ollama, Groq, and any compatible endpoint
+- 🔄 **API Flexibility**: Works with OpenAI, Ollama, Groq, Claude, and any compatible endpoint
 - 💬 **Interactive Chat**: Continuous conversation with memory in modern UI
 - 📊 **Streaming Responses**: Real-time output for better user experience
 - 🔍 **Web Search**: Integrated with compatible API endpoints
@@ -143,6 +150,8 @@ For more examples and detailed usage, visit the [CLI Usage Guide](https://nazdri
 - 📃 **Conversation Logging**: Save your conversations to text files for later reference
 - 🧰 **CLI Components**: Reusable components for building custom AI-powered command-line tools
 - 🔌 **Modular Architecture**: Well-structured codebase with clean separation of concerns
+- 🔄 **Provider Switching**: Easily switch between different LLM providers with a single parameter
+- 🚀 **Performance Optimized**: Fast response times and minimal resource usage
 
 See the [Feature Overview](https://nazdridoy.github.io/ngpt/overview.html) for more details.
 
@@ -291,6 +300,16 @@ print(result.stdout)
 # Returns only code without markdown or explanations
 code = client.generate_code("function that converts Celsius to Fahrenheit")
 print(code)
+
+# Compare responses from different providers
+openai_config = load_config(config_index=0)  # OpenAI
+groq_config = load_config(config_index=1)    # Groq
+
+openai_client = NGPTClient(**openai_config)
+groq_client = NGPTClient(**groq_config)
+
+openai_response = openai_client.chat("Explain quantum computing")
+groq_response = groq_client.chat("Explain quantum computing")
 ```
 
 For advanced usage patterns and integrations, check out the [Advanced Examples](https://nazdridoy.github.io/ngpt/examples/advanced.html).
@@ -356,6 +375,7 @@ You can configure nGPT using the following options:
 | `-t, --text` | Open interactive multiline editor for complex prompts with syntax highlighting |
 | `--stdin` | Read from stdin and use content with prompt. Use {} in prompt as placeholder for stdin content |
 | `--rewrite` | Rewrite text to improve quality while preserving original tone and meaning |
+| `--gitcommsg` | Generate AI-powered git commit messages from staged changes or diff files |
 
 #### Global Options
 

{ngpt-2.14.0 → ngpt-2.15.0}/docs/CONTRIBUTING.md

@@ -35,8 +35,10 @@ Thank you for your interest in contributing to NGPT! This document provides guid
     - `formatters.py` - Output formatting logic
     - `interactive.py` - Interactive mode implementation
     - `modes/` - Different operational modes
+      - `__init__.py` - Modes initialization
       - `chat.py` - Chat mode functionality
       - `code.py` - Code generation mode
+      - `gitcommsg.py` - Git commit message generation
       - `shell.py` - Shell command generation
       - `text.py` - Text generation mode
       - `rewrite.py` - Text rewriting mode

{ngpt-2.14.0 → ngpt-2.15.0}/docs/README.md

@@ -14,10 +14,13 @@ Welcome to the documentation for nGPT, a lightweight Python CLI and library for
   - [CLI Usage](usage/cli_usage.md)
   - [Library Usage](usage/library_usage.md)
   - [CLI Framework](usage/cli_framework.md)
+  - [CLI Configuration](usage/cli_config.md)
+  - [Git Commit Messages](usage/gitcommsg.md)
 - [API Reference](api/README.md)
   - [NGPTClient](api/client.md)
   - [Configuration](api/config.md)
   - [CLI Components](api/cli.md)
+  - [CLI Configuration](api/cli_config.md)
 - [Examples](examples/README.md)
   - [Basic Examples](examples/basic.md)
   - [Advanced Examples](examples/advanced.md)

{ngpt-2.14.0 → ngpt-2.15.0}/docs/api/README.md

@@ -10,6 +10,7 @@ nGPT's API consists of several main components:
 2. **Configuration Utilities**: Functions for managing configuration files and settings
 3. **CLI Components**: Modular CLI utilities that can be reused in custom applications
 4. **CLI Configuration Utilities**: Functions for managing CLI-specific configurations
+5. **Logging Utilities**: Tools for logging conversations and errors
 
 ## Table of Contents
 
@@ -26,14 +27,22 @@ nGPT's API consists of several main components:
   - [Editing Configurations](config.md#editing-configurations)
   - [Removing Configurations](config.md#removing-configurations)
   - [Configuration Paths](config.md#configuration-paths)
+  - [Default Configuration](config.md#default-configuration)
 
 - [CLI Components](cli.md) - CLI functionality
   - [Module Structure](cli.md#module-structure)
   - [Interactive Chat Module](cli.md#interactive-chat-module)
   - [Formatters Module](cli.md#formatters-module)
   - [Renderers Module](cli.md#renderers-module)
+  - [UI Components](cli.md#ui-components)
   - [CLI Configuration Utilities](cli.md#cli-configuration-utilities)
   - [Operation Modes](cli.md#operation-modes)
+    - [Chat Mode](cli.md#chat-mode)
+    - [Code Mode](cli.md#code-mode)
+    - [Shell Mode](cli.md#shell-mode)
+    - [Text Mode](cli.md#text-mode)
+    - [Rewrite Mode](cli.md#rewrite-mode)
+    - [Git Commit Message Mode](cli.md#git-commit-message-mode)
 
 - [CLI Configuration](cli_config.md) - Persistent CLI settings
   - [Loading Configuration](cli_config.md#load_cli_config)
@@ -42,12 +51,20 @@ nGPT's API consists of several main components:
   - [Removing Options](cli_config.md#unset_cli_config_option)
   - [Applying Configuration](cli_config.md#apply_cli_config)
   - [Available Options](cli_config.md#available-cli-configuration-options)
+  - [Listing Options](cli_config.md#list_cli_config_options)
+  - [Configuration Paths](cli_config.md#configuration-paths)
+
+- [Logging](logging.md) - Logging utilities
+  - [Creating Loggers](logging.md#create_logger)
+  - [Logger Class](logging.md#logger-class)
+  - [Log Levels](logging.md#log-levels)
+  - [Log Formatting](logging.md#log-formatting)
 
 ## Quick Reference
 
 ```python
 # Import core components
-from ngpt import NGPTClient, load_config
+from ngpt import NGPTClient, load_config, __version__
 
 # Import configuration utilities
 from ngpt.utils.config import (
@@ -55,7 +72,9 @@ from ngpt.utils.config import (
     get_config_path,
     get_config_dir,
     add_config_entry,
-    remove_config_entry
+    remove_config_entry,
+    DEFAULT_CONFIG,
+    DEFAULT_CONFIG_ENTRY
 )
 
 # Import CLI configuration utilities
@@ -64,22 +83,32 @@ from ngpt.utils.cli_config import (
     set_cli_config_option,
     get_cli_config_option,
     unset_cli_config_option,
-    apply_cli_config
+    apply_cli_config,
+    list_cli_config_options,
+    CLI_CONFIG_OPTIONS,
+    get_cli_config_dir,
+    get_cli_config_path
 )
 
+# Import logging utilities
+from ngpt.utils.log import create_logger, Logger
+
 # Import CLI module components
 from ngpt.cli.interactive import interactive_chat_session
 from ngpt.cli.formatters import prettify_markdown, ColoredHelpFormatter, COLORS
 from ngpt.cli.renderers import prettify_streaming_markdown, has_markdown_renderer
+from ngpt.cli.ui import create_progress_bar, create_spinner
 
 # Import operation modes
 from ngpt.cli.modes.chat import chat_mode
 from ngpt.cli.modes.code import code_mode
 from ngpt.cli.modes.shell import shell_mode
 from ngpt.cli.modes.text import text_mode
+from ngpt.cli.modes.rewrite import rewrite_mode
+from ngpt.cli.modes.gitcommsg import git_commit_message_mode
 
-# Import version information
-from ngpt import __version__
+# Import main CLI entry point
+from ngpt.cli import main
 ```
 
-For complete documentation on using these components, see the linked reference pages. 
+For complete documentation on using these components, see the linked reference pages. For examples of integrating nGPT into your applications, see the [Examples](../examples/) section. 

{ngpt-2.14.0 → ngpt-2.15.0}/docs/api/cli.md

@@ -22,6 +22,7 @@ The CLI components for nGPT are organized into a modular structure under the `ng
     - `ngpt.cli.modes.shell`: Shell command generation mode functionality
     - `ngpt.cli.modes.text`: Text generation mode functionality
     - `ngpt.cli.modes.rewrite`: Text rewriting mode functionality
+    - `ngpt.cli.modes.gitcommsg`: Git commit message generation functionality
 
 ## Interactive Chat Module
 
@@ -675,6 +676,80 @@ else:
     print("This example would open a multiline editor in a real terminal")
 ```
 
+### Git Commit Message Mode
+
+```python
+from ngpt.cli.modes.gitcommsg import gitcommsg_mode
+
+def gitcommsg_mode(client, args, logger=None)
+```
+
+Executes a git commit message generation operation to create commit messages based on staged changes in a git repository.
+
+**Parameters:**
+- `client` (NGPTClient): The initialized client for the operation
+- `args` (namespace): Parsed command-line arguments including:
+  - `diff_file` (str, optional): Path to diff file to use instead of git diff --staged
+  - `message_context` (str, optional): Additional context or instructions for commit message generation
+  - `temperature` (float): Temperature setting (0.0-1.0) 
+  - `max_tokens` (int, optional): Maximum tokens to generate
+  - `chunk_size` (int): Number of lines per chunk for large diffs
+  - `max_lines` (int): Maximum lines in the generated commit message
+  - `recursive` (bool): Whether to use recursive chunking for large diffs
+  - `deep_recursive` (bool): Whether to use deeper recursive analysis
+  - `web_search` (bool): Whether to enable web search
+- `logger` (object, optional): Logger instance
+
+**Features:**
+- **Git Diff Analysis**: Analyzes staged git changes to create relevant commit messages
+- **Conventional Commit Format**: Follows standard commit message format (type(scope): message)
+- **Chunking Strategy**: Handles large diffs by splitting into manageable chunks
+- **Recursive Analysis**: Optional deep analysis of complex changes
+- **Context Directives**: Supports filtering and focusing on specific file types/components
+- **Technical Detail Extraction**: Extracts function names, line numbers, and specific changes
+
+**Example:**
+```python
+from ngpt import NGPTClient, load_config
+from ngpt.cli.modes.gitcommsg import gitcommsg_mode
+import argparse
+
+client = NGPTClient(**load_config())
+
+# Create args namespace with required parameters
+args = argparse.Namespace()
+args.diff_file = None  # Use git staged changes
+args.message_context = "type:feat focus on authentication"
+args.temperature = 0.4
+args.max_tokens = None
+args.chunk_size = 200
+args.max_lines = 20
+args.recursive = True
+args.deep_recursive = False
+args.web_search = False
+
+gitcommsg_mode(
+    client=client,
+    args=args
+)
+```
+
+**Context Directive Examples:**
+- `type:feat` - Set commit type to feature
+- `focus on authentication` - Focus only on authentication-related changes
+- `ignore tests` - Exclude test changes from the commit message
+- `javascript` - Focus only on JavaScript file changes
+
+**Output Example:**
+```
+feat(auth): implement JWT authentication and user session management
+
+- Add generateToken function in auth/jwt.js
+- Create validateJWTMiddleware in middleware/auth.js
+- Add token refresh endpoint in routes/auth.js
+- Update user model to store refresh tokens
+```
+
 ## Reference Tables
 
 ### Markdown Renderers

{ngpt-2.14.0 → ngpt-2.15.0}/docs/api/cli_config.md

@@ -13,7 +13,7 @@ nGPT provides a set of utilities for managing CLI-specific configuration setting
 ```python
 from ngpt.utils.cli_config import load_cli_config
 
-def load_cli_config():
+def load_cli_config() -> Dict[str, Any]:
 ```
 
 Loads the CLI configuration from the configuration file.
@@ -35,14 +35,14 @@ print(f"Current CLI configuration: {cli_config}")
 ```python
 from ngpt.utils.cli_config import set_cli_config_option
 
-def set_cli_config_option(option, value):
+def set_cli_config_option(option: str, value: Any) -> Tuple[bool, str]:
 ```
 
 Sets a CLI configuration option.
 
 **Parameters:**
 - `option` (str): The name of the option to set
-- `value` (str): The value to set for the option
+- `value` (Any): The value to set for the option
 
 **Returns:**
 - tuple: (success, message) where success is a boolean indicating whether the operation was successful and message is a string explaining the result
@@ -65,7 +65,7 @@ print(message)
 ```python
 from ngpt.utils.cli_config import get_cli_config_option
 
-def get_cli_config_option(option=None):
+def get_cli_config_option(option: str = None) -> Tuple[bool, Union[str, Dict[str, Any]]]:
 ```
 
 Gets the value of a CLI configuration option, or all options if none is specified.
@@ -99,7 +99,7 @@ if success:
 ```python
 from ngpt.utils.cli_config import unset_cli_config_option
 
-def unset_cli_config_option(option):
+def unset_cli_config_option(option: str) -> Tuple[bool, str]:
 ```
 
 Removes a CLI configuration option.
@@ -124,18 +124,17 @@ print(message)
 ```python
 from ngpt.utils.cli_config import apply_cli_config
 
-def apply_cli_config(args, options=None, context="all"):
+def apply_cli_config(args: Any, mode: str) -> Any:
 ```
 
-Applies CLI configuration options to the provided argument namespace.
+Applies CLI configuration options to the provided argument namespace based on the current mode, respecting context and not overriding explicit args.
 
 **Parameters:**
-- `args` (namespace): The argument namespace (from argparse)
-- `options` (list, optional): List of option names to apply (applies all if None)
-- `context` (str): The context for applying the configuration (e.g., "all", "chat", "code", "shell", "text")
+- `args` (Any): The argument namespace (from argparse)
+- `mode` (str): The current mode ('interactive', 'shell', 'code', 'text', 'gitcommsg', or 'all' for default)
 
 **Returns:**
-- None: Modifies the args namespace in-place
+- Any: The updated argument namespace
 
 **Example:**
 ```python
@@ -150,13 +149,115 @@ parser.add_argument("--markdown-format", action="store_true")
 args = parser.parse_args()
 
 # Apply CLI configuration for code generation context
-apply_cli_config(args, context="code")
+apply_cli_config(args, mode="code")
 
 # Use the updated arguments
 print(f"Temperature: {args.temperature}")
 print(f"Language: {args.language}")
 ```
 
+### `list_cli_config_options`
+
+```python
+from ngpt.utils.cli_config import list_cli_config_options
+
+def list_cli_config_options() -> List[str]:
+```
+
+Lists all available CLI configuration options.
+
+**Returns:**
+- list: A sorted list of option names
+
+**Example:**
+```python
+from ngpt.utils.cli_config import list_cli_config_options
+
+# Get all available options
+options = list_cli_config_options()
+print("Available configuration options:")
+for option in options:
+    print(f"- {option}")
+```
+
+### `get_cli_config_dir`
+
+```python
+from ngpt.utils.cli_config import get_cli_config_dir
+
+def get_cli_config_dir() -> Path:
+```
+
+Gets the directory where the CLI configuration is stored.
+
+**Returns:**
+- Path: The path to the CLI configuration directory
+
+**Example:**
+```python
+from ngpt.utils.cli_config import get_cli_config_dir
+
+# Get the CLI config directory
+config_dir = get_cli_config_dir()
+print(f"CLI configuration directory: {config_dir}")
+```
+
+### `get_cli_config_path`
+
+```python
+from ngpt.utils.cli_config import get_cli_config_path
+
+def get_cli_config_path() -> Path:
+```
+
+Gets the path to the CLI configuration file.
+
+**Returns:**
+- Path: The path to the CLI configuration file
+
+**Example:**
+```python
+from ngpt.utils.cli_config import get_cli_config_path
+
+# Get the CLI config file path
+config_path = get_cli_config_path()
+print(f"CLI configuration file: {config_path}")
+```
+
+### `save_cli_config`
+
+```python
+from ngpt.utils.cli_config import save_cli_config
+
+def save_cli_config(config: Dict[str, Any]) -> bool:
+```
+
+Saves CLI configuration to the config file.
+
+**Parameters:**
+- `config` (Dict[str, Any]): The configuration dictionary to save
+
+**Returns:**
+- bool: True if the operation was successful, False otherwise
+
+**Example:**
+```python
+from ngpt.utils.cli_config import load_cli_config, save_cli_config
+
+# Load existing config
+config = load_cli_config()
+
+# Modify config
+config['temperature'] = 0.9
+
+# Save the updated config
+success = save_cli_config(config)
+if success:
+    print("Configuration saved successfully")
+else:
+    print("Failed to save configuration")
+```
+
 ## Available CLI Configuration Options
 
 The following options are available for configuration:
@@ -164,18 +265,25 @@ The following options are available for configuration:
 | Option | Type | Context | Description |
 |--------|------|---------|-------------|
 | `temperature` | float | all | Controls randomness in the response (0.0-1.0) |
-| `top-p` | float | all | Controls diversity via nucleus sampling (0.0-1.0) |
+| `top_p` | float | all | Controls diversity via nucleus sampling (0.0-1.0) |
 | `no-stream` | bool | all | Disables streaming responses |
-| `max-tokens` | int | all | Maximum number of tokens to generate |
-| `web-search` | bool | all | Enables web search capability |
 | `prettify` | bool | all | Enables markdown prettification |
+| `stream-prettify` | bool | all | Enables streaming prettification |
+| `max_tokens` | int | all | Maximum number of tokens to generate |
+| `web-search` | bool | all | Enables web search capability |
 | `renderer` | string | all | Markdown renderer to use ('auto', 'rich', 'glow') |
 | `language` | string | code | Programming language for code generation |
-| `execute` | bool | shell | Whether to execute generated shell commands |
-| `markdown-format` | bool | all | Format responses with markdown |
 | `provider` | string | all | Default provider to use |
 | `config-index` | int | all | Default configuration index to use |
-| `model` | string | all | Default model to use |
+| `log` | string | all | Path to log file |
+| `preprompt` | string | all | Custom preprompt to use |
+| `message-context` | string | gitcommsg | Context for git commit message generation |
+| `recursive-chunk` | bool | gitcommsg | Enable recursive chunking for large diffs |
+| `diff` | string | gitcommsg | Path to diff file for git commit message generation |
+| `chunk-size` | int | gitcommsg | Maximum number of lines per chunk |
+| `analyses-chunk-size` | int | gitcommsg | Maximum number of lines per chunk for analyses |
+| `max-msg-lines` | int | gitcommsg | Maximum number of lines in commit message |
+| `max-recursion-depth` | int | gitcommsg | Maximum recursion depth for message condensing |
 
 ## Configuration File Location
 
@@ -194,7 +302,8 @@ from ngpt.utils.cli_config import (
     load_cli_config,
     set_cli_config_option,
     get_cli_config_option,
-    apply_cli_config
+    apply_cli_config,
+    unset_cli_config_option
 )
 from ngpt.cli.formatters import ColoredHelpFormatter, COLORS
 
@@ -215,7 +324,7 @@ def main():
     args = parser.parse_args()
     
     # Apply existing CLI configuration (for code generation context)
-    apply_cli_config(args, context="code")
+    apply_cli_config(args, mode="code")
     
     # If user wants to save current settings
     if args.save_config: