Fixed issue with Gemini API

lib/gpt_providers/image_to_text_gen/gemini_image_details.py

@@ -1,79 +0,0 @@
-"""
-"""
-import os
-import logging
-from pathlib import Path
-
-import google.generativeai as genai
-logging.basicConfig(level=logging.INFO, format='%(asctime)s-%(levelname)s-%(module)s-%(lineno)d-%(message)s')
-from dotenv import load_dotenv
-load_dotenv(Path('../../.env'))
-
-from tenacity import (
-    retry,
-    stop_after_attempt,
-    wait_random_exponential,
-) # for exponential backoff
-
-
-@retry(wait=wait_random_exponential(min=1, max=60), stop=stop_after_attempt(6))
-def gemini_get_img_info(prompt, img_path):
-    """ Get image details from arxiv papers. """
-    logging.info(f"Get image details from Gemini Pro.")
-    try:
-        genai.configure(api_key=os.getenv("GEMINI_API_KEY"))
-    except Exception as e:
-        logging.error(f"Could not load gemini API key: {e}")
-        raise e
-
-    # Set up the model
-    generation_config = {
-        "temperature": 0.9,
-        "top_p": 1,
-        "top_k": 1,
-        "max_output_tokens": 1096,
-    }
-
-    safety_settings = [{
-        "category": "HARM_CATEGORY_HARASSMENT",
-        "threshold": "BLOCK_NONE"
-    },
-    {
-        "category": "HARM_CATEGORY_HATE_SPEECH",
-        "threshold": "BLOCK_NONE"
-    },
-    {
-        "category": "HARM_CATEGORY_SEXUALLY_EXPLICIT",
-        "threshold": "BLOCK_NONE"
-    },
-    {
-        "category": "HARM_CATEGORY_DANGEROUS_CONTENT",
-        "threshold": "BLOCK_NONE"
-    },]
-
-    try:
-        model = genai.GenerativeModel(model_name="gemini-pro-vision",
-            generation_config=generation_config,
-            safety_settings=safety_settings)
-    except Exception as e:
-        logging.error(f"Could not create GenerativeModel: {e}")
-        raise e
-
-    # Validate that an image is present
-    if not (img := Path(img_path)).exists():
-        raise FileNotFoundError(f"Could not find image: {img}")
-
-    image_parts = [{
-        "mime_type": "image/png",
-        "data": Path(img_path).read_bytes()
-    },]
-
-    prompt_parts = [f"{prompt}", image_parts[0],]
-
-    try:
-        response = model.generate_content(prompt_parts)
-        return response.text
-    except Exception as e:
-        logging.error(f"Gemini is blocking this request: {response.prompt_feedback.block_reason}")
-        logging.error(f"Gemini Vision, Failed to give image Details: {e}\n{response.prompt_feedback}")
-        raise e

lib/gpt_providers/image_to_text_gen/openai_vision_image_details.py

@@ -0,0 +1,79 @@
+"""
+This module provides functionality to analyze images using OpenAI's Vision API. 
+It encodes an image to a base64 string and sends a request to the OpenAI API 
+to interpret the contents of the image, returning a textual description.
+"""
+
+import requests
+import sys
+import re
+import base64
+
+def analyze_and_extract_details_from_image(image_path, api_key):
+    """
+    Analyzes an image using OpenAI's Vision API and extracts Alt Text, Description, Title, and Caption.
+
+    Args:
+        image_path (str): Path to the image file.
+        api_key (str): Your OpenAI API key.
+
+    Returns:
+        dict: Extracted details including Alt Text, Description, Title, and Caption.
+    """
+    def encode_image(path):
+        """ Encodes an image to a base64 string. """
+        with open(path, "rb", encoding="utf-8") as image_file:
+            return base64.b64encode(image_file.read()).decode('utf-8')
+
+    base64_image = encode_image(image_path)
+
+    headers = {
+        "Content-Type": "application/json",
+        "Authorization": f"Bearer {api_key}"
+    }
+
+    payload = {
+        "model": "gpt-4-vision-preview",
+        "messages": [
+            {
+                "role": "user",
+                "content": [
+                    {
+                        "type": "text",
+                        "text": "The given image is used in blog content. Analyze the given image and suggest alternative(alt) test, description, title, caption."
+                    },
+                    {
+                        "type": "image_url",
+                        "image_url": {
+                            "url": f"data:image/jpeg;base64,{base64_image}"
+                        }
+                    }
+                ]
+            }
+        ],
+        "max_tokens": 300
+    }
+
+    try:
+        response = requests.post("https://api.openai.com/v1/chat/completions", headers=headers, json=payload)
+        response.raise_for_status()
+
+        assistant_message = response.json()['choices'][0]['message']['content']
+
+        # Extracting details using regular expressions
+        alt_text_match = re.search(r'Alt Text: "(.*?)"', assistant_message)
+        description_match = re.search(r'Description: (.*?)\n\n', assistant_message)
+        title_match = re.search(r'Title: "(.*?)"', assistant_message)
+        caption_match = re.search(r'Caption: "(.*?)"', assistant_message)
+
+        return {
+            'alt_text': alt_text_match.group(1) if alt_text_match else None,
+            'description': description_match.group(1) if description_match else None,
+            'title': title_match.group(1) if title_match else None,
+            'caption': caption_match.group(1) if caption_match else None
+        }
+
+    except requests.RequestException as e:
+        sys.exit(f"Error: Failed to communicate with OpenAI API. Error: {e}")
+    except Exception as e:
+        sys.exit(f"Error occurred: {e}")