Merge pull request #20 from loryanstrant/feature/v2.0.0-image-generation

🎨 Release v2.0.0: Add DALL-E image generation support

Author: loryanstrant

README.md

@@ -14,7 +14,9 @@ A Home Assistant custom integration that facilitates AI tasks using Azure AI ser
 - Easy configuration through Home Assistant UI
 - Secure API key management  
 - **User-configurable AI models for chat responses** (GPT-3.5, GPT-4, GPT-4o, etc.) - type in any model name
+- **🎨 Image generation with DALL-E support** - generate images from text prompts using DALL-E 2/3
 - **Image and video analysis with attachment support** - analyze camera streams and uploaded images
+- **Flexible entity configuration** - create chat-only, image-only, or combined entities
 - **Reconfiguration support** - change models without re-entering credentials
 - **Multiple entry support** - use different API endpoints and keys for different purposes
 - Compatible with Azure OpenAI and other Azure AI services
@@ -52,9 +54,10 @@ Or replace steps 1-6 with this:
 2. Search for "Azure AI Tasks"
 3. Enter your Azure AI endpoint URL (make sure you use the Azure OpenAI URL - i.e. https://USE-YOUR-RESOURCE-URL.openai.azure.com)
 4. Enter your API key
-5. **Enter your preferred chat model** (gpt-35-turbo, gpt-4, gpt-4o, etc.)
-6. Give your integration a name
-7. Click Submit
+5. **Enter your preferred chat model** (gpt-35-turbo, gpt-4, gpt-4o, etc.) - leave empty for image-only entities
+6. **Enter your preferred image model** (dall-e-2, dall-e-3, etc.) - leave empty for chat-only entities  
+7. Give your integration a name
+8. Click Submit
 
 <img width="383" height="478" alt="image" src="https://github.com/user-attachments/assets/8932c51a-8fcb-42bc-9e22-ead143c610d7" />
 
@@ -68,9 +71,14 @@ Or replace steps 1-6 with this:
 To change AI models without re-entering credentials:
 1. Go to your Azure AI Tasks integration
 2. Click "Configure" 
-3. Enter a different model as needed
+3. Enter different chat/image models as needed (use placeholder text `[None - leave empty to disable chat]` or `[None - leave empty to disable images]` to clear fields)
 4. Save changes
 
+**Note**: You can create specialized entities by leaving one model type empty:
+- **Chat-only entities**: Configure chat model, leave image model empty
+- **Image-only entities**: Configure image model, leave chat model empty  
+- **Combined entities**: Configure both models
+
 <img width="1072" height="700" alt="image" src="https://github.com/user-attachments/assets/598b8c28-7663-4507-be63-22413cac4b9d" />
 
 
@@ -92,6 +100,34 @@ data:
 ```
 ![HA Azure AI Task example](https://github.com/user-attachments/assets/592ec039-20ea-436f-a6f0-caf88bef9b56)
 
+### 🎨 Image Generation
+Example service calls for generating images with DALL-E:
+
+**Basic Image Generation:**
+```yaml
+action: ai_image.generate_image
+data:
+  prompt: "A futuristic smart home with holographic displays and AI assistants"
+  entity_id: ai_image.azure_ai_tasks_dall_e_3
+```
+
+**Advanced Image Generation with Parameters:**
+```yaml
+action: ai_image.generate_image
+data:
+  prompt: "A cozy living room during sunset with warm lighting"
+  entity_id: ai_image.azure_ai_tasks_dall_e_3
+  size: "1024x1024"
+  quality: "hd"
+  style: "vivid"
+```
+
+**Supported DALL-E Parameters:**
+- **size**: Image dimensions (DALL-E 2: 256x256, 512x512, 1024x1024; DALL-E 3: 1024x1024, 1024x1792, 1792x1024)
+- **quality**: Image quality for DALL-E 3 (standard, hd)  
+- **style**: Image style for DALL-E 3 (natural, vivid)
+- **n**: Number of images to generate (1-10 for DALL-E 2, 1 for DALL-E 3)
+
 
 
 ### Image/Video Analysis with Attachments
@@ -135,11 +171,16 @@ data:
 
 ### Available Models
 
-You can enter any model name that your Azure AI deployment supports.
+**Chat Models**: You can enter any chat model name that your Azure AI deployment supports:
+- gpt-35-turbo, gpt-4, gpt-4o, gpt-4-turbo, etc.
+
+**Image Models**: Supported image generation models:
+- **dall-e-2**: Classic DALL-E model with multiple size options
+- **dall-e-3**: Latest DALL-E model with enhanced quality and style controls
 
 ## Requirements
 
-- Home Assistant 2024.1 or later
+- **Home Assistant 2025.10.0 or later** (required for AI Task and AI Image services)
 - Azure AI service with API access
 - Valid Azure AI endpoint and API key
 

custom_components/azure_ai_tasks/__init__.py

@@ -1,30 +1,99 @@
 """The Azure AI Tasks integration."""
 from __future__ import annotations
 
+import logging
+
 from homeassistant.config_entries import ConfigEntry
-from homeassistant.const import Platform
+from homeassistant.const import Platform, __version__ as ha_version
 from homeassistant.core import HomeAssistant
+from homeassistant.exceptions import ConfigEntryNotReady
 
 from .const import DOMAIN
 
 PLATFORMS: list[Platform] = [Platform.AI_TASK]
 
+_LOGGER = logging.getLogger(__name__)
+
+# Minimum Home Assistant version required
+MIN_HA_VERSION = "2025.10.0"
+
+
+def _check_ha_version() -> None:
+    """Check if Home Assistant version meets minimum requirements."""
+    from packaging import version
+    
+    try:
+        current_version = version.parse(ha_version.split(".dev")[0])  # Remove .dev suffix if present
+        min_version = version.parse(MIN_HA_VERSION)
+        
+        if current_version < min_version:
+            raise ConfigEntryNotReady(
+                f"Home Assistant {MIN_HA_VERSION} or newer is required. "
+                f"Current version: {ha_version}"
+            )
+    except Exception as err:
+        _LOGGER.warning(
+            "Unable to verify Home Assistant version compatibility: %s. "
+            "Integration may not work correctly if running on older versions.",
+            err
+        )
+
+
+async def async_migrate_entry(hass: HomeAssistant, config_entry: ConfigEntry) -> bool:
+    """Migrate old entry to new version."""
+    _LOGGER.info(f"Migrating Azure AI Tasks config entry {config_entry.entry_id} from version {config_entry.version} to version 2")
+    
+    if config_entry.version == 1:
+        new_data = dict(config_entry.data)
+        new_options = dict(config_entry.options)
+        migrated = False
+        
+        # Remove deprecated gpt-35-turbo from both data and options
+        if new_data.get("chat_model") == "gpt-35-turbo":
+            new_data["chat_model"] = ""
+            migrated = True
+            _LOGGER.info("Removed deprecated gpt-35-turbo from data.chat_model")
+            
+        if new_options.get("chat_model") == "gpt-35-turbo":
+            new_options["chat_model"] = ""  
+            migrated = True
+            _LOGGER.info("Removed deprecated gpt-35-turbo from options.chat_model")
+        
+        # Update the config entry
+        hass.config_entries.async_update_entry(
+            config_entry,
+            data=new_data,
+            options=new_options,
+            version=2
+        )
+        
+        if migrated:
+            _LOGGER.info(f"Successfully migrated config entry {config_entry.entry_id}, cleaned deprecated model")
+        else:
+            _LOGGER.info(f"Migrated config entry {config_entry.entry_id} to version 2, no deprecated models found")
+            
+    return True
+
 
 async def async_setup_entry(hass: HomeAssistant, entry: ConfigEntry) -> bool:
     """Set up Azure AI Tasks from a config entry."""
+    # Check Home Assistant version compatibility
+    _check_ha_version()
+    
+    # Set up the integration
     hass.data.setdefault(DOMAIN, {})
     hass.data[DOMAIN][entry.entry_id] = entry.data
-
-    # Set up update listener for options changes
-    entry.async_on_unload(entry.add_update_listener(async_update_options))
-
-    await hass.config_entries.async_forward_entry_setups(entry, PLATFORMS)
-
+    
+    # Forward entry setup to AI task platform
+    await hass.config_entries.async_forward_entry_setups(entry, ["ai_task"])
+    
     return True
 
 
 async def async_update_options(hass: HomeAssistant, entry: ConfigEntry) -> None:
     """Update options."""
+    _LOGGER.info("Azure AI Tasks options updated for entry %s", entry.entry_id)
+    _LOGGER.info("New options: %s", entry.options)
     # Reload the integration when options change
     await hass.config_entries.async_reload(entry.entry_id)