Added extraction of labels from image

.env.example

@@ -1 +1,2 @@
-OPENAI_KEY_PERSONAL=<openai_key>
+OPENAI_KEY_PERSONAL=<openai_key>
+GOOGLE_APPLICATION_CREDENTIALS=<gcp_key>

.gitignore

@@ -1,4 +1,5 @@
 venv
 .vscode
 logs
-.env
+.env
+google-vision.json

code_tasks/custom.py

@@ -0,0 +1,22 @@
+from concurrent.futures import ThreadPoolExecutor
+from typing import Dict, List
+
+from google.cloud import vision
+
+from utils.io import print_system
+
+
+def get_labels_from_image(url: str) -> List[str]:
+    client = vision.ImageAnnotatorClient()
+    image = vision.Image()
+    image.source.image_uri = url  # type: ignore
+
+    response = client.label_detection(image=image)  # type: ignore
+    return [label.description for label in response.label_annotations]
+
+
+def get_labels_from_images(urls) -> Dict:
+    print_system("Generating labels for images...")
+    with ThreadPoolExecutor() as executor:
+        labels = list(executor.map(get_labels_from_image, urls))
+    return {urls[i]: labels[i] for i, url in enumerate(labels)}

code_tasks/images_in_url.py

@@ -0,0 +1,42 @@
+"""
+ChatGPT-4 prompt: write a python function that given an url returns all images in the website
+"""
+
+import requests
+from bs4 import BeautifulSoup
+
+def get_images_from_url(url):
+    # Send a GET request to the URL
+    response = requests.get(url)
+
+    # If the GET request is successful, the status code will be 200
+    if response.status_code == 200:
+        # Get the content of the response
+        page_content = response.content
+
+        # Create a BeautifulSoup object and specify the parser
+        soup = BeautifulSoup(page_content, 'html.parser')
+
+        # Find all image tags
+        images = soup.find_all('img')
+
+        # Create a list to store the URLs of the images
+        image_urls = []
+
+        # For each image tag, get the URL of the image
+        for image in images:
+            # If the tag has the 'src' attribute
+            if image.has_attr('src'):
+                image_url = image['src']
+                
+                # If URL is relative, convert it to absolute
+                if image_url.startswith('/'):
+                    image_url = url + image_url
+
+                image_urls.append(image_url)
+
+        return image_urls
+
+    # If the GET request is not successful, return None
+    else:
+        return None

control_flow/main.py

@@ -1,13 +1,18 @@
 from ai import image
 from ai_tasks.best_headlines import get_headlines
 from ai_tasks.image_prompt import generate_prompt
-from code_tasks.url_text import get_text_from_url
+from code_tasks.custom import get_labels_from_images
+from code_tasks.images_in_url import get_images_from_url
+from code_tasks.text_in_url import get_text_from_url
 from utils.io import print_assistant, print_system, user_input
 
 
 def run():
     url = user_input("URL: ")
     text = get_text_from_url(url)
+    images = get_images_from_url(url)
+    image_labels = get_labels_from_images(images)
+
     headlines = get_headlines(text)
     print_assistant(headlines)
     prompt = generate_prompt(text)