devcommit 0.1.5.0__tar.gz → 0.1.5.1__tar.gz

{devcommit-0.1.5.0 → devcommit-0.1.5.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: devcommit
-Version: 0.1.5.0
+Version: 0.1.5.1
 Summary: AI-powered git commit message generator
 License: GNU GENERAL PUBLIC LICENSE
                                 Version 3, 29 June 2007
@@ -711,6 +711,7 @@ A command-line AI tool for autocommits.
 
 - 🤖 **Multi-AI Provider Support** - Choose from Gemini, Groq, OpenAI, Claude, Ollama, or custom APIs
 - 🚀 Automatic commit generation using AI
+- 📝 **Changelog Generation** - Automatically generate markdown changelogs from your changes
 - 📁 Directory-based commits - create separate commits for each root directory
 - 🎯 Interactive mode to choose between global or directory-based commits
 - 📄 **Commit specific files or folders** - Stage and commit only selected files/directories
@@ -1115,6 +1116,47 @@ Selecting "Regenerate commit messages" will:
 
 This works for all commit modes (global, directory, and per-file commits).
 
+### Changelog Generation
+
+DevCommit can automatically generate markdown changelog files from your changes using AI.
+
+**Usage:**
+
+```bash
+# Generate changelog after committing
+devcommit --changelog
+
+# Generate changelog before staging (recommended)
+devcommit --stageAll --changelog
+
+# Short form
+devcommit -s -c
+
+# With specific files
+devcommit --stageAll --changelog --files src/
+```
+
+**How it works:**
+
+- **With `--stageAll`**: Changelog is generated from unstaged changes **before** staging
+- **Without `--stageAll`**: Changelog is generated from the last commit **after** committing
+- Changelogs are saved as markdown files with datetime-based names (e.g., `2026-01-28_00-55-30.md`)
+- Default directory: `changelogs/` (configurable via `CHANGELOG_DIR` in `.dcommit`)
+- Uses Keep a Changelog format with AI-generated content
+
+**Example workflow:**
+
+```bash
+# Make changes to your code
+# ...
+
+# Stage all changes and generate changelog before committing
+devcommit --stageAll --changelog
+
+# The changelog file is created in changelogs/ directory
+# Then changes are staged and committed
+```
+
 ### Additional Options
 
 - `--excludeFiles` or `-e`: Exclude specific files from the diff
@@ -1124,6 +1166,7 @@ This works for all commit modes (global, directory, and per-file commits).
 - `--directory` or `-d`: Force directory-based commits
 - `--files` or `-f`: Stage and commit specific files or folders (can specify multiple)
 - `--push` or `-p`: Push commits to remote after committing
+- `--changelog` or `-c`: Generate changelog file from changes
 
 ### Examples
 
@@ -1157,6 +1200,15 @@ devcommit -s -f src/core src/modules/account/ --directory
 
 # Stage and commit, then push
 devcommit -s -f src/core src/modules/account/ -p
+
+# Generate changelog before staging and committing
+devcommit --stageAll --changelog
+
+# Generate changelog after committing
+devcommit --changelog
+
+# Generate changelog with specific files
+devcommit -s -c -f src/
 ```
 
 ## AI Provider Support
@@ -1188,28 +1240,19 @@ devcommit
 ```bash
 export AI_PROVIDER=openrouter
 export OPENROUTER_API_KEY='your-openrouter-api-key'
-# Optional: specify model (default: mistralai/devstral-2512:free)
-export OPENROUTER_MODEL='mistralai/devstral-2512:free'
+# Optional: specify model (default: meta-llama/llama-3.3-70b-instruct:free)
+export OPENROUTER_MODEL='meta-llama/llama-3.3-70b-instruct:free'
 devcommit
 ```
 
 **Popular free models on OpenRouter (add `:free` suffix):**
 
-**Coding & Development:**
-- `mistralai/devstral-2512:free` - Mistral's state-of-the-art coding model (123B params)
-- `kwaipilot/kat-coder-pro:free` - Advanced agentic coding model (73.4% SWE-Bench solve rate)
-- `qwen/qwen3-coder:free` - Qwen's MoE code generation model (480B params, 35B active)
-- `deepseek/deepseek-r1-0528:free` - DeepSeek R1 reasoning model (671B params, open-source)
-
-**Reasoning & General Purpose:**
-- `xiaomi/mimo-v2-flash:free` - Mixture-of-Experts model (309B params, top open-source on SWE-bench)
-- `tngtech/deepseek-r1t2-chimera:free` - DeepSeek R1T2 reasoning model (671B params)
-- `tngtech/deepseek-r1t-chimera:free` - DeepSeek R1T reasoning model
-- `tngtech/tng-r1t-chimera:free` - Creative storytelling and character interaction model
-
-**Lightweight & Fast:**
-- `z-ai/glm-4.5-air:free` - GLM 4.5 Air lightweight variant (MoE architecture)
-- `nvidia/nemotron-3-nano-30b-a3b:free` - NVIDIA's efficient small model (30B params, 3B active)
+**Recommended Models:**
+- `meta-llama/llama-3.3-70b-instruct:free` - Llama 3.3 70B Instruct (Powerful & General Purpose)
+- `google/gemma-3-27b-it:free` - Google Gemma 3 27B Instruct (Efficient & Capable)
+- `openai/gpt-oss-120b:free` - OpenAI GPT-OSS 120B (Large & Experimental)
+- `tngtech/deepseek-r1t-chimera:free` - DeepSeek R1T Chimera (Strong Reasoning)
+- `qwen/qwen3-next-80b-a3b-instruct:free` - Qwen3 Next 80B (Advanced Instruction Following)
 
 **Important Notes:**
 - **Logging Requirements:** Some free models may log your prompts and responses for model improvement purposes. This means:
@@ -1274,7 +1317,7 @@ All configuration can be set via **environment variables** or **`.dcommit` file*
 | Variable | Description | Default |
 |----------|-------------|---------|
 | `OPENROUTER_API_KEY` | OpenRouter API key ([Get it here](https://openrouter.ai/keys)) | - |
-| `OPENROUTER_MODEL` | Model name (add `:free` suffix for free models) | `mistralai/devstral-2512:free` |
+| `OPENROUTER_MODEL` | Model name (add `:free` suffix for free models) | `meta-llama/llama-3.3-70b-instruct:free` |
 
 **Anthropic:**
 | Variable | Description | Default |
@@ -1305,6 +1348,7 @@ All configuration can be set via **environment variables** or **`.dcommit` file*
 | `COMMIT_MODE`   | Default commit strategy              | `auto`                                                 | `auto`, `directory`, `global`, `related` |
 | `EXCLUDE_FILES` | Files to exclude from diff           | `package-lock.json, pnpm-lock.yaml, yarn.lock, *.lock` | Comma-separated file patterns            |
 | `MAX_TOKENS`    | Maximum tokens for AI response       | `8192`                                                 | Any positive integer                     |
+| `CHANGELOG_DIR` | Directory for changelog files        | `changelogs`                                           | Any directory path                       |
 
 ### Configuration Priority
 

{devcommit-0.1.5.0 → devcommit-0.1.5.1}/README.md

@@ -6,6 +6,7 @@ A command-line AI tool for autocommits.
 
 - 🤖 **Multi-AI Provider Support** - Choose from Gemini, Groq, OpenAI, Claude, Ollama, or custom APIs
 - 🚀 Automatic commit generation using AI
+- 📝 **Changelog Generation** - Automatically generate markdown changelogs from your changes
 - 📁 Directory-based commits - create separate commits for each root directory
 - 🎯 Interactive mode to choose between global or directory-based commits
 - 📄 **Commit specific files or folders** - Stage and commit only selected files/directories
@@ -410,6 +411,47 @@ Selecting "Regenerate commit messages" will:
 
 This works for all commit modes (global, directory, and per-file commits).
 
+### Changelog Generation
+
+DevCommit can automatically generate markdown changelog files from your changes using AI.
+
+**Usage:**
+
+```bash
+# Generate changelog after committing
+devcommit --changelog
+
+# Generate changelog before staging (recommended)
+devcommit --stageAll --changelog
+
+# Short form
+devcommit -s -c
+
+# With specific files
+devcommit --stageAll --changelog --files src/
+```
+
+**How it works:**
+
+- **With `--stageAll`**: Changelog is generated from unstaged changes **before** staging
+- **Without `--stageAll`**: Changelog is generated from the last commit **after** committing
+- Changelogs are saved as markdown files with datetime-based names (e.g., `2026-01-28_00-55-30.md`)
+- Default directory: `changelogs/` (configurable via `CHANGELOG_DIR` in `.dcommit`)
+- Uses Keep a Changelog format with AI-generated content
+
+**Example workflow:**
+
+```bash
+# Make changes to your code
+# ...
+
+# Stage all changes and generate changelog before committing
+devcommit --stageAll --changelog
+
+# The changelog file is created in changelogs/ directory
+# Then changes are staged and committed
+```
+
 ### Additional Options
 
 - `--excludeFiles` or `-e`: Exclude specific files from the diff
@@ -419,6 +461,7 @@ This works for all commit modes (global, directory, and per-file commits).
 - `--directory` or `-d`: Force directory-based commits
 - `--files` or `-f`: Stage and commit specific files or folders (can specify multiple)
 - `--push` or `-p`: Push commits to remote after committing
+- `--changelog` or `-c`: Generate changelog file from changes
 
 ### Examples
 
@@ -452,6 +495,15 @@ devcommit -s -f src/core src/modules/account/ --directory
 
 # Stage and commit, then push
 devcommit -s -f src/core src/modules/account/ -p
+
+# Generate changelog before staging and committing
+devcommit --stageAll --changelog
+
+# Generate changelog after committing
+devcommit --changelog
+
+# Generate changelog with specific files
+devcommit -s -c -f src/
 ```
 
 ## AI Provider Support
@@ -483,28 +535,19 @@ devcommit
 ```bash
 export AI_PROVIDER=openrouter
 export OPENROUTER_API_KEY='your-openrouter-api-key'
-# Optional: specify model (default: mistralai/devstral-2512:free)
-export OPENROUTER_MODEL='mistralai/devstral-2512:free'
+# Optional: specify model (default: meta-llama/llama-3.3-70b-instruct:free)
+export OPENROUTER_MODEL='meta-llama/llama-3.3-70b-instruct:free'
 devcommit
 ```
 
 **Popular free models on OpenRouter (add `:free` suffix):**
 
-**Coding & Development:**
-- `mistralai/devstral-2512:free` - Mistral's state-of-the-art coding model (123B params)
-- `kwaipilot/kat-coder-pro:free` - Advanced agentic coding model (73.4% SWE-Bench solve rate)
-- `qwen/qwen3-coder:free` - Qwen's MoE code generation model (480B params, 35B active)
-- `deepseek/deepseek-r1-0528:free` - DeepSeek R1 reasoning model (671B params, open-source)
-
-**Reasoning & General Purpose:**
-- `xiaomi/mimo-v2-flash:free` - Mixture-of-Experts model (309B params, top open-source on SWE-bench)
-- `tngtech/deepseek-r1t2-chimera:free` - DeepSeek R1T2 reasoning model (671B params)
-- `tngtech/deepseek-r1t-chimera:free` - DeepSeek R1T reasoning model
-- `tngtech/tng-r1t-chimera:free` - Creative storytelling and character interaction model
-
-**Lightweight & Fast:**
-- `z-ai/glm-4.5-air:free` - GLM 4.5 Air lightweight variant (MoE architecture)
-- `nvidia/nemotron-3-nano-30b-a3b:free` - NVIDIA's efficient small model (30B params, 3B active)
+**Recommended Models:**
+- `meta-llama/llama-3.3-70b-instruct:free` - Llama 3.3 70B Instruct (Powerful & General Purpose)
+- `google/gemma-3-27b-it:free` - Google Gemma 3 27B Instruct (Efficient & Capable)
+- `openai/gpt-oss-120b:free` - OpenAI GPT-OSS 120B (Large & Experimental)
+- `tngtech/deepseek-r1t-chimera:free` - DeepSeek R1T Chimera (Strong Reasoning)
+- `qwen/qwen3-next-80b-a3b-instruct:free` - Qwen3 Next 80B (Advanced Instruction Following)
 
 **Important Notes:**
 - **Logging Requirements:** Some free models may log your prompts and responses for model improvement purposes. This means:
@@ -569,7 +612,7 @@ All configuration can be set via **environment variables** or **`.dcommit` file*
 | Variable | Description | Default |
 |----------|-------------|---------|
 | `OPENROUTER_API_KEY` | OpenRouter API key ([Get it here](https://openrouter.ai/keys)) | - |
-| `OPENROUTER_MODEL` | Model name (add `:free` suffix for free models) | `mistralai/devstral-2512:free` |
+| `OPENROUTER_MODEL` | Model name (add `:free` suffix for free models) | `meta-llama/llama-3.3-70b-instruct:free` |
 
 **Anthropic:**
 | Variable | Description | Default |
@@ -600,6 +643,7 @@ All configuration can be set via **environment variables** or **`.dcommit` file*
 | `COMMIT_MODE`   | Default commit strategy              | `auto`                                                 | `auto`, `directory`, `global`, `related` |
 | `EXCLUDE_FILES` | Files to exclude from diff           | `package-lock.json, pnpm-lock.yaml, yarn.lock, *.lock` | Comma-separated file patterns            |
 | `MAX_TOKENS`    | Maximum tokens for AI response       | `8192`                                                 | Any positive integer                     |
+| `CHANGELOG_DIR` | Directory for changelog files        | `changelogs`                                           | Any directory path                       |
 
 ### Configuration Priority
 

{devcommit-0.1.5.0 → devcommit-0.1.5.1}/devcommit/app/ai_providers.py

@@ -8,7 +8,7 @@ from typing import Optional
 
 # Suppress stderr for all AI imports
 _stderr = sys.stderr
-_devnull = open(os.devnull, 'w')
+_devnull = open(os.devnull, "w")
 sys.stderr = _devnull
 
 try:
@@ -37,31 +37,37 @@ _devnull.close()
 
 class AIProvider(ABC):
     """Base class for AI providers"""
-    
+
     @abstractmethod
-    def generate_commit_message(self, diff: str, prompt: str, max_tokens: int) -> str:
+    def generate_commit_message(
+        self, diff: str, prompt: str, max_tokens: int
+    ) -> str:
         """Generate commit message from diff"""
         pass
 
 
 class GeminiProvider(AIProvider):
     """Google Gemini AI provider"""
-    
+
     def __init__(self, api_key: str, model: str = "gemini-2.0-flash-exp"):
         if not genai:
-            raise ImportError("google-generativeai not installed. Run: pip install google-generativeai")
-        
+            raise ImportError(
+                "google-generativeai not installed. Run: pip install google-generativeai"
+            )
+
         # Suppress stderr during configuration
         _stderr = sys.stderr
-        _devnull = open(os.devnull, 'w')
+        _devnull = open(os.devnull, "w")
         sys.stderr = _devnull
         genai.configure(api_key=api_key)
         sys.stderr = _stderr
         _devnull.close()
-        
+
         self.model_name = model
-        
-    def generate_commit_message(self, diff: str, prompt: str, max_tokens: int) -> str:
+
+    def generate_commit_message(
+        self, diff: str, prompt: str, max_tokens: int
+    ) -> str:
         generation_config = {
             "response_mime_type": "text/plain",
             "max_output_tokens": max_tokens,
@@ -69,12 +75,12 @@ class GeminiProvider(AIProvider):
             "top_p": 0.9,
             "temperature": 0.7,
         }
-        
+
         model = genai.GenerativeModel(
             generation_config=generation_config,
             model_name=self.model_name,
         )
-        
+
         chat_session = model.start_chat(
             history=[
                 {
@@ -83,15 +89,15 @@ class GeminiProvider(AIProvider):
                 },
             ]
         )
-        
+
         # Suppress stderr during API call
         _stderr = sys.stderr
-        _devnull = open(os.devnull, 'w')
+        _devnull = open(os.devnull, "w")
         sys.stderr = _devnull
         response = chat_session.send_message(diff)
         sys.stderr = _stderr
         _devnull.close()
-        
+
         if response and hasattr(response, "text"):
             return response.text.strip()
         return "No valid commit message generated."
@@ -99,232 +105,258 @@ class GeminiProvider(AIProvider):
 
 class OpenAIProvider(AIProvider):
     """OpenAI GPT provider"""
-    
+
     def __init__(self, api_key: str, model: str = "gpt-4o-mini"):
         if not openai:
             raise ImportError("openai not installed. Run: pip install openai")
         self.client = openai.OpenAI(api_key=api_key)
         self.model = model
-        
-    def generate_commit_message(self, diff: str, prompt: str, max_tokens: int) -> str:
+
+    def generate_commit_message(
+        self, diff: str, prompt: str, max_tokens: int
+    ) -> str:
         response = self.client.chat.completions.create(
             model=self.model,
             messages=[
                 {"role": "system", "content": prompt},
-                {"role": "user", "content": diff}
+                {"role": "user", "content": diff},
             ],
             max_tokens=max_tokens,
-            temperature=0.7
+            temperature=0.7,
         )
         return response.choices[0].message.content.strip()
 
 
 class GroqProvider(AIProvider):
     """Groq AI provider (OpenAI-compatible)"""
-    
+
     def __init__(self, api_key: str, model: str = "llama-3.3-70b-versatile"):
         if not openai:
             raise ImportError("openai not installed. Run: pip install openai")
         self.client = openai.OpenAI(
-            api_key=api_key,
-            base_url="https://api.groq.com/openai/v1"
+            api_key=api_key, base_url="https://api.groq.com/openai/v1"
         )
         self.model = model
-        
-    def generate_commit_message(self, diff: str, prompt: str, max_tokens: int) -> str:
+
+    def generate_commit_message(
+        self, diff: str, prompt: str, max_tokens: int
+    ) -> str:
         response = self.client.chat.completions.create(
             model=self.model,
             messages=[
                 {"role": "system", "content": prompt},
-                {"role": "user", "content": diff}
+                {"role": "user", "content": diff},
             ],
             max_tokens=max_tokens,
-            temperature=0.7
+            temperature=0.7,
         )
         return response.choices[0].message.content.strip()
 
 
 class OpenRouterProvider(AIProvider):
     """OpenRouter.ai provider (OpenAI-compatible, access to multiple models)"""
-    
-    def __init__(self, api_key: str, model: str = "mistralai/devstral-2512:free"):
+
+    def __init__(
+        self,
+        api_key: str,
+        model: str = "meta-llama/llama-3.3-70b-instruct:free",
+    ):
         if not openai:
             raise ImportError("openai not installed. Run: pip install openai")
         self.client = openai.OpenAI(
             api_key=api_key,
-            base_url="https://openrouter.ai/api/v1"
+            base_url="https://openrouter.ai/api/v1",
+            default_headers={
+                "HTTP-Referer": "https://github.com/opsguild/DevCommit",
+                "X-Title": "DevCommit",
+            },
         )
         self.model = model
-        
-    def generate_commit_message(self, diff: str, prompt: str, max_tokens: int) -> str:
+
+    def generate_commit_message(
+        self, diff: str, prompt: str, max_tokens: int
+    ) -> str:
         response = self.client.chat.completions.create(
             model=self.model,
             messages=[
                 {"role": "system", "content": prompt},
-                {"role": "user", "content": diff}
+                {"role": "user", "content": diff},
             ],
             max_tokens=max_tokens,
-            temperature=0.7
+            temperature=0.7,
+            extra_body={"transforms": ["middle-out"]},
+            extra_headers={
+                "HTTP-Referer": "https://github.com/hordunlarmy/DevCommit",
+                "X-Title": "DevCommit",
+            },
         )
         return response.choices[0].message.content.strip()
 
 
 class AnthropicProvider(AIProvider):
     """Anthropic Claude provider"""
-    
+
     def __init__(self, api_key: str, model: str = "claude-3-haiku-20240307"):
         if not anthropic:
-            raise ImportError("anthropic not installed. Run: pip install anthropic")
+            raise ImportError(
+                "anthropic not installed. Run: pip install anthropic"
+            )
         self.client = anthropic.Anthropic(api_key=api_key)
         self.model = model
-        
-    def generate_commit_message(self, diff: str, prompt: str, max_tokens: int) -> str:
+
+    def generate_commit_message(
+        self, diff: str, prompt: str, max_tokens: int
+    ) -> str:
         message = self.client.messages.create(
             model=self.model,
             max_tokens=max_tokens,
             system=prompt,
-            messages=[
-                {"role": "user", "content": diff}
-            ]
+            messages=[{"role": "user", "content": diff}],
         )
         return message.content[0].text.strip()
 
 
 class OllamaProvider(AIProvider):
     """Ollama local model provider"""
-    
-    def __init__(self, base_url: str = "http://localhost:11434", model: str = "llama3"):
+
+    def __init__(
+        self, base_url: str = "http://localhost:11434", model: str = "llama3"
+    ):
         if not requests:
-            raise ImportError("requests not installed. Run: pip install requests")
-        self.base_url = base_url.rstrip('/')
+            raise ImportError(
+                "requests not installed. Run: pip install requests"
+            )
+        self.base_url = base_url.rstrip("/")
         self.model = model
-        
-    def generate_commit_message(self, diff: str, prompt: str, max_tokens: int) -> str:
+
+    def generate_commit_message(
+        self, diff: str, prompt: str, max_tokens: int
+    ) -> str:
         url = f"{self.base_url}/api/generate"
         data = {
             "model": self.model,
             "prompt": f"{prompt}\n\n{diff}",
             "stream": False,
-            "options": {
-                "temperature": 0.7,
-                "num_predict": max_tokens
-            }
+            "options": {"temperature": 0.7, "num_predict": max_tokens},
         }
-        
+
         response = requests.post(url, json=data, timeout=60)
         response.raise_for_status()
         result = response.json()["response"].strip()
-        
+
         # Return raw result - normalization is done centrally in gemini_ai.py
         return result
 
 
 class CustomProvider(AIProvider):
     """Custom OpenAI-compatible API provider"""
-    
-    def __init__(self, api_url: str, api_key: Optional[str] = None, model: str = "default"):
+
+    def __init__(
+        self,
+        api_url: str,
+        api_key: Optional[str] = None,
+        model: str = "default",
+    ):
         if not openai:
             raise ImportError("openai not installed. Run: pip install openai")
-        
+
         # Extract base URL (remove /chat/completions if present)
-        base_url = api_url.replace('/chat/completions', '').replace('/v1/chat/completions', '')
-        if not base_url.endswith('/v1'):
-            base_url = base_url.rstrip('/') + '/v1'
-            
+        base_url = api_url.replace("/chat/completions", "").replace(
+            "/v1/chat/completions", ""
+        )
+        if not base_url.endswith("/v1"):
+            base_url = base_url.rstrip("/") + "/v1"
+
         self.client = openai.OpenAI(
-            api_key=api_key or "dummy-key",
-            base_url=base_url
+            api_key=api_key or "dummy-key", base_url=base_url
         )
         self.model = model
-        
-    def generate_commit_message(self, diff: str, prompt: str, max_tokens: int) -> str:
+
+    def generate_commit_message(
+        self, diff: str, prompt: str, max_tokens: int
+    ) -> str:
         response = self.client.chat.completions.create(
             model=self.model,
             messages=[
                 {"role": "system", "content": prompt},
-                {"role": "user", "content": diff}
+                {"role": "user", "content": diff},
             ],
             max_tokens=max_tokens,
-            temperature=0.7
+            temperature=0.7,
         )
         return response.choices[0].message.content.strip()
 
 
 def get_ai_provider(config) -> AIProvider:
     """Factory function to get the appropriate AI provider based on config"""
-    
+
     provider_name = config("AI_PROVIDER", default="gemini").lower()
-    
+
     if provider_name == "gemini":
         api_key = config("GEMINI_API_KEY", default=None)
         if not api_key:
             raise ValueError("GEMINI_API_KEY not set")
         # Support legacy MODEL_NAME for backward compatibility
-        model = config("GEMINI_MODEL", default=None) or config("MODEL_NAME", default="gemini-2.0-flash-exp")
+        model = config("GEMINI_MODEL", default=None) or config(
+            "MODEL_NAME", default="gemini-2.0-flash-exp"
+        )
         return GeminiProvider(api_key, model)
-    
+
     elif provider_name == "openai":
         api_key = config("OPENAI_API_KEY", default=None)
         if not api_key:
             raise ValueError("OPENAI_API_KEY not set")
-        model = (
-            config("OPENAI_MODEL", default=None)
-            or config("MODEL_NAME", default="gpt-4o-mini")
+        model = config("OPENAI_MODEL", default=None) or config(
+            "MODEL_NAME", default="gpt-4o-mini"
         )
         return OpenAIProvider(api_key, model)
-    
+
     elif provider_name == "groq":
         api_key = config("GROQ_API_KEY", default=None)
         if not api_key:
             raise ValueError("GROQ_API_KEY not set")
-        model = (
-            config("GROQ_MODEL", default=None)
-            or config("MODEL_NAME", default="llama-3.3-70b-versatile")
+        model = config("GROQ_MODEL", default=None) or config(
+            "MODEL_NAME", default="llama-3.3-70b-versatile"
         )
         return GroqProvider(api_key, model)
-    
+
     elif provider_name == "openrouter":
         api_key = config("OPENROUTER_API_KEY", default=None)
         if not api_key:
             raise ValueError("OPENROUTER_API_KEY not set")
-        model = (
-            config("OPENROUTER_MODEL", default=None)
-            or config("MODEL_NAME", default="mistralai/devstral-2512:free")
+        model = config("OPENROUTER_MODEL", default=None) or config(
+            "MODEL_NAME", default="meta-llama/llama-3.3-70b-instruct:free"
         )
         return OpenRouterProvider(api_key, model)
-    
+
     elif provider_name == "anthropic":
         api_key = config("ANTHROPIC_API_KEY", default=None)
         if not api_key:
             raise ValueError("ANTHROPIC_API_KEY not set")
-        model = (
-            config("ANTHROPIC_MODEL", default=None)
-            or config("MODEL_NAME", default="claude-3-haiku-20240307")
+        model = config("ANTHROPIC_MODEL", default=None) or config(
+            "MODEL_NAME", default="claude-3-haiku-20240307"
         )
         return AnthropicProvider(api_key, model)
-    
+
     elif provider_name == "ollama":
         base_url = config("OLLAMA_BASE_URL", default="http://localhost:11434")
-        model = (
-            config("OLLAMA_MODEL", default=None)
-            or config("MODEL_NAME", default="llama3")
+        model = config("OLLAMA_MODEL", default=None) or config(
+            "MODEL_NAME", default="llama3"
         )
         return OllamaProvider(base_url, model)
-    
+
     elif provider_name == "custom":
         api_url = config("CUSTOM_API_URL", default=None)
         if not api_url:
             raise ValueError("CUSTOM_API_URL not set for custom provider")
         api_key = config("CUSTOM_API_KEY", default=None)
-        model = (
-            config("CUSTOM_MODEL", default=None)
-            or config("MODEL_NAME", default="default")
+        model = config("CUSTOM_MODEL", default=None) or config(
+            "MODEL_NAME", default="default"
         )
         return CustomProvider(api_url, api_key, model)
-    
+
     else:
         raise ValueError(
             f"Unknown AI provider: {provider_name}. "
             f"Supported: gemini, openai, groq, openrouter, anthropic, ollama, custom"
         )
-

devcommit-0.1.5.1/devcommit/app/changelog.py

@@ -0,0 +1,89 @@
+#!/usr/bin/env python
+"""Generate changelog files from git diffs using AI"""
+
+import os
+from datetime import datetime
+from devcommit.utils.logger import config
+from devcommit.app.ai_providers import get_ai_provider
+
+
+def generate_changelog_prompt() -> str:
+    """Generate the prompt for changelog creation"""
+    return """You are a changelog generator. Analyze the git diff and create a structured changelog in Keep a Changelog format.
+
+Follow these guidelines:
+1. Use markdown format with clear sections
+2. Categorize changes into: Added, Changed, Fixed, Removed, Deprecated, Security
+3. Write clear, user-friendly descriptions (not implementation details)
+4. Group related changes together
+5. Focus on what changed from a user/developer perspective
+6. Be concise but informative
+
+Format:
+# Changelog
+
+## [Unreleased]
+
+### Added
+- List new features
+
+### Changed
+- List changes to existing functionality
+
+### Fixed
+- List bug fixes
+
+### Removed
+- List removed features
+
+Only include sections that have changes. Do not add empty sections."""
+
+
+def generate_changelog(diff: str) -> str:
+    """Generate changelog content from git diff using AI.
+    
+    Args:
+        diff: Git diff string
+        
+    Returns:
+        Formatted markdown changelog content
+    """
+    prompt = generate_changelog_prompt()
+    
+    # Get AI provider from config
+    provider = get_ai_provider(config)
+    
+    # Generate changelog using AI
+    max_tokens = config("MAX_TOKENS", default=8192, cast=int)
+    changelog_content = provider.generate_commit_message(diff, prompt, max_tokens)
+    
+    return changelog_content
+
+
+def save_changelog(content: str, directory: str = None) -> str:
+    """Save changelog content to a file.
+    
+    Args:
+        content: Changelog markdown content
+        directory: Directory to save changelog (default from config)
+        
+    Returns:
+        Path to the saved changelog file
+    """
+    # Get directory from config if not provided
+    if directory is None:
+        directory = config("CHANGELOG_DIR", default="changelogs")
+    
+    # Create directory if it doesn't exist
+    os.makedirs(directory, exist_ok=True)
+    
+    # Generate filename with current datetime
+    now = datetime.now()
+    filename = now.strftime("%Y-%m-%d_%H-%M-%S.md")
+    filepath = os.path.join(directory, filename)
+    
+    # Write content to file
+    with open(filepath, 'w', encoding='utf-8') as f:
+        f.write(content)
+    
+    return filepath

{devcommit-0.1.5.0 → devcommit-0.1.5.1}/devcommit/app/gemini_ai.py

@@ -16,6 +16,7 @@ warnings.filterwarnings('ignore', message='.*ALTS.*')
 warnings.filterwarnings('ignore', category=UserWarning)
 
 from devcommit.utils.logger import Logger, config
+from devcommit.utils.git import KnownError
 from .ai_providers import get_ai_provider
 from .prompt import generate_prompt
 
@@ -83,8 +84,16 @@ def generateCommitMessage(diff: str) -> str:
         return normalized_response
 
     except Exception as e:
-        logger.error(f"Error generating commit message: {e}")
-        return f"Error generating commit message: {str(e)}"
+        error_msg = str(e)
+        logger.error(f"Error generating commit message: {error_msg}")
+        
+        # Raise KnownError with user-friendly message that includes error details
+        # This prevents error messages from being shown as commit options
+        # while still informing the user what went wrong
+        raise KnownError(
+            f"Failed to generate commit message: {error_msg}. "
+            f"Please check your API configuration and try again."
+        )
     finally:
         # Restore stderr and close devnull
         sys.stderr = _stderr

{devcommit-0.1.5.0 → devcommit-0.1.5.1}/devcommit/main.py

@@ -10,6 +10,7 @@ from InquirerPy import get_style, inquirer
 from rich.console import Console
 
 from devcommit.app.gemini_ai import generateCommitMessage
+from devcommit.app.changelog import generate_changelog, save_changelog
 from devcommit.utils.git import (KnownError, assert_git_repo,
                                  get_detected_message, get_diff_for_files,
                                  get_files_from_paths, get_staged_diff,
@@ -34,6 +35,36 @@ def has_commits() -> bool:
     return result.returncode == 0
 
 
+def sanitize_commit_messages(messages):
+    """Filter out error messages from commit message list.
+    
+    Args:
+        messages: String or list of commit messages
+        
+    Returns:
+        List of valid commit messages with error messages filtered out
+    """
+    # Convert string to list if needed
+    if isinstance(messages, str):
+        if not messages or messages.strip() == "":
+            return []
+        messages = messages.split("|")
+    
+    # Filter out error messages and empty strings
+    valid_messages = []
+    for msg in messages:
+        if not msg:
+            continue
+        msg = msg.strip()
+        # Skip if it's an error message
+        if msg.startswith("Error generating commit message:"):
+            continue
+        if msg and msg.strip():
+            valid_messages.append(msg)
+    
+    return valid_messages
+
+
 # Main function
 def main(flags: CommitFlag = None):
     if flags is None:
@@ -99,6 +130,42 @@ def main(flags: CommitFlag = None):
             except Exception as e:
                 raise KnownError(f"Failed to get files from paths: {str(e)}")
 
+
+        # Generate changelog before staging if --changelog and --stageAll are both set
+        if flags["stageAll"] and flags["changelog"]:
+            console.print()
+            console.print("[bold cyan]📝 Generating changelog...[/bold cyan]")
+            
+            with console.status(
+                "[magenta]🤖 AI generating changelog from changes...[/magenta]",
+                spinner="dots",
+                spinner_style="magenta"
+            ):
+                # Get diff of unstaged changes
+                if push_files_list:
+                    # Get diff for specific files
+                    diff = get_diff_for_files(push_files_list, flags["excludeFiles"])
+                else:
+                    # Get diff for all changes
+                    import subprocess
+                    diff = subprocess.run(
+                        ["git", "diff"],
+                        stdout=subprocess.PIPE,
+                        text=True,
+                    ).stdout
+                
+                if diff:
+                    try:
+                        changelog_content = generate_changelog(diff)
+                        changelog_path = save_changelog(changelog_content)
+                        console.print(f"[bold green]✅ Changelog saved to:[/bold green] [cyan]{changelog_path}[/cyan]")
+                    except Exception as e:
+                        logger.error(f"Failed to generate changelog: {e}")
+                        console.print(f"[bold yellow]⚠️  Failed to generate changelog: {e}[/bold yellow]")
+                else:
+                    console.print("[bold yellow]⚠️  No changes to generate changelog from[/bold yellow]")
+            console.print()
+
         if flags["stageAll"]:
             if push_files_list:
                 # Stage specific files/folders only
@@ -258,6 +325,36 @@ def main(flags: CommitFlag = None):
         elif flags.get("push", False) and not commit_made:
             console.print("\n[bold yellow]⚠️  No commits were made, skipping push[/bold yellow]\n")
         
+        # Generate changelog after commit if --changelog is used without --stageAll
+        if flags.get("changelog", False) and not flags.get("stageAll", False) and commit_made:
+            console.print()
+            console.print("[bold cyan]📝 Generating changelog from committed changes...[/bold cyan]")
+            
+            with console.status(
+                "[magenta]🤖 AI generating changelog...[/magenta]",
+                spinner="dots",
+                spinner_style="magenta"
+            ):
+                # Get diff from last commit
+                import subprocess
+                diff = subprocess.run(
+                    ["git", "diff", "HEAD~1", "HEAD"],
+                    stdout=subprocess.PIPE,
+                    text=True,
+                ).stdout
+                
+                if diff:
+                    try:
+                        changelog_content = generate_changelog(diff)
+                        changelog_path = save_changelog(changelog_content)
+                        console.print(f"[bold green]✅ Changelog saved to:[/bold green] [cyan]{changelog_path}[/cyan]")
+                    except Exception as e:
+                        logger.error(f"Failed to generate changelog: {e}")
+                        console.print(f"[bold yellow]⚠️  Failed to generate changelog: {e}[/bold yellow]")
+                else:
+                    console.print("[bold yellow]⚠️  No changes to generate changelog from[/bold yellow]")
+            console.print()
+        
         # Print stylish completion message only if commits were made
         if commit_made:
             console.print()
@@ -273,7 +370,8 @@ def main(flags: CommitFlag = None):
         return
     except KnownError as error:
         logger.error(str(error))
-        console.print(f"\n[bold red]❌ Error:[/bold red] [red]{error}[/red]\n")
+        # Don't print here - Rich status context already displays the error
+        pass
     except subprocess.CalledProcessError as error:
         logger.error(str(error))
         console.print(f"\n[bold red]❌ Git command failed:[/bold red] [red]{error}[/red]\n")
@@ -322,6 +420,9 @@ def analyze_changes(console, files=None):
     """
     import sys
     
+    # Store any exception to re-raise after status context exits
+    caught_exception = None
+    
     with console.status(
         "[magenta]🤖 AI analyzing changes...[/magenta]",
         spinner="dots",
@@ -350,17 +451,28 @@ def analyze_changes(console, files=None):
         
         try:
             commit_message = generateCommitMessage(diff)
+        except KnownError as e:
+            # Catch KnownError to prevent Rich status from printing it
+            caught_exception = e
+            commit_message = None
         finally:
             sys.stderr = _stderr
             _devnull.close()
         
-        if isinstance(commit_message, str):
-            commit_message = commit_message.split("|")
-
-        if not commit_message:
-            raise KnownError("No commit messages were generated. Try again.")
+        # If we caught an exception, we'll re-raise it after the status context exits
+        if caught_exception:
+            pass  # Will be raised below
+        else:
+            commit_message = sanitize_commit_messages(commit_message)
 
-        return commit_message
+            if not commit_message:
+                raise KnownError("No commit messages were generated. Try again.")
+    
+    # Re-raise the exception outside the status context to avoid duplicate printing
+    if caught_exception:
+        raise caught_exception
+    
+    return commit_message
 
 
 def prompt_commit_message(console, commit_message, regenerate_callback=None):
@@ -703,8 +815,7 @@ def process_per_directory_commits(console, staged, flags):
                 sys.stderr = _stderr
                 _devnull.close()
             
-            if isinstance(commit_message, str):
-                commit_message = commit_message.split("|")
+            commit_message = sanitize_commit_messages(commit_message)
             
             if not commit_message:
                 console.print(f"\n[bold yellow]⚠️  No commit message generated for {directory}, skipping[/bold yellow]\n")
@@ -722,9 +833,7 @@ def process_per_directory_commits(console, staged, flags):
                 sys.stderr = _devnull
                 try:
                     msg = generateCommitMessage(diff)
-                    if isinstance(msg, str):
-                        msg = msg.split("|")
-                    return msg
+                    return sanitize_commit_messages(msg)
                 finally:
                     sys.stderr = _stderr
                     _devnull.close()
@@ -865,8 +974,7 @@ def process_per_file_commits(console, staged, flags):
                 sys.stderr = _stderr
                 _devnull.close()
             
-            if isinstance(commit_message, str):
-                commit_message = commit_message.split("|")
+            commit_message = sanitize_commit_messages(commit_message)
             
             if not commit_message:
                 console.print(f"\n[bold yellow]⚠️  No commit message generated for {file}, skipping[/bold yellow]\n")
@@ -884,9 +992,7 @@ def process_per_file_commits(console, staged, flags):
                 sys.stderr = _devnull
                 try:
                     msg = generateCommitMessage(diff)
-                    if isinstance(msg, str):
-                        msg = msg.split("|")
-                    return msg
+                    return sanitize_commit_messages(msg)
                 finally:
                     sys.stderr = _stderr
                     _devnull.close()
@@ -997,8 +1103,7 @@ def process_per_directory_commits_from_paths(console, staged, flags, original_pa
                 sys.stderr = _stderr
                 _devnull.close()
             
-            if isinstance(commit_message, str):
-                commit_message = commit_message.split("|")
+            commit_message = sanitize_commit_messages(commit_message)
             
             if not commit_message:
                 console.print(f"\n[bold yellow]⚠️  No commit message generated for {path}, skipping[/bold yellow]\n")
@@ -1016,9 +1121,7 @@ def process_per_directory_commits_from_paths(console, staged, flags, original_pa
                 sys.stderr = _devnull
                 try:
                     msg = generateCommitMessage(diff)
-                    if isinstance(msg, str):
-                        msg = msg.split("|")
-                    return msg
+                    return sanitize_commit_messages(msg)
                 finally:
                     sys.stderr = _stderr
                     _devnull.close()
@@ -1140,7 +1243,7 @@ def process_per_related_commits(console, staged, flags):
             # Fallback mode detected - AI parsing failed, so commit_messages are empty
             # Generate commit messages for all groups (one call per group, but done upfront)
             # This is still extra calls, but at least they're done before user interaction
-            from devcommit.app.gemini_ai import generateCommitMessage
+            # Note: generateCommitMessage is already imported at the top of the file
             
             console.print("[dim]⚠️  AI grouping response had no commit messages. Generating now...[/dim]")
             with console.status(
@@ -1160,9 +1263,7 @@ def process_per_related_commits(console, staged, flags):
                         diff = get_diff_for_files(group_files, flags["excludeFiles"])
                         if diff:
                             commit_msgs = generateCommitMessage(diff)
-                            if isinstance(commit_msgs, str):
-                                commit_msgs = commit_msgs.split("|")
-                            group_data['commit_messages'] = [msg.strip() for msg in commit_msgs if msg and msg.strip()]
+                            group_data['commit_messages'] = sanitize_commit_messages(commit_msgs)
                 finally:
                     sys.stderr = _stderr
                     _devnull.close()
@@ -1345,8 +1446,7 @@ def process_per_related_commits(console, staged, flags):
                     sys.stderr = _stderr
                     _devnull.close()
                 
-                if isinstance(commit_message, str):
-                    commit_message = commit_message.split("|")
+                commit_message = sanitize_commit_messages(commit_message)
                 
                 if not commit_message:
                     console.print(f"\n[bold yellow]⚠️  No commit message generated for {group_name}, skipping[/bold yellow]\n")
@@ -1364,9 +1464,7 @@ def process_per_related_commits(console, staged, flags):
                 sys.stderr = _devnull
                 try:
                     msg = generateCommitMessage(diff)
-                    if isinstance(msg, str):
-                        msg = msg.split("|")
-                    return msg
+                    return sanitize_commit_messages(msg)
                 finally:
                     sys.stderr = _stderr
                     _devnull.close()

{devcommit-0.1.5.0 → devcommit-0.1.5.1}/devcommit/utils/parser.py

@@ -11,6 +11,7 @@ class CommitFlag(TypedDict):
     directory: bool
     files: List[str]
     push: bool
+    changelog: bool
     rawArgv: List[str]
 
 
@@ -52,6 +53,10 @@ def parse_arguments() -> CommitFlag:
         "--push", "-p", action="store_true",
         help="Push commits to remote after committing"
     )
+    parser.add_argument(
+        "--changelog", "-c", action="store_true",
+        help="Generate changelog file from changes"
+    )
     parser.add_argument(
         "rawArgv", nargs="*", help="Additional arguments for git commit"
     )
@@ -66,5 +71,6 @@ def parse_arguments() -> CommitFlag:
         directory=args.directory,
         files=args.files or [],
         push=args.push,
+        changelog=args.changelog,
         rawArgv=args.rawArgv,
     )

{devcommit-0.1.5.0 → devcommit-0.1.5.1}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "devcommit"
-version = "0.1.5.0"
+version = "0.1.5.1"
 description = "AI-powered git commit message generator"
 readme = "README.md"
 license = {file = "COPYING"}