WIP - Streamlit UI, Porting CLI

lib/ai_writers/keywords_to_blog_streamlit.py

@@ -20,6 +20,7 @@ from .blog_from_google_serp import write_blog_google_serp, blog_with_research
 from ..ai_web_researcher.you_web_reseacher import get_rag_results, search_ydc_index
 from ..blog_metadata.get_blog_metadata import blog_metadata
 from ..blog_postprocessing.save_blog_to_file import save_blog_to_file
+from ..gpt_providers.text_to_image_generation.main_generate_image_from_prompt import generate_image
 
 
 def write_blog_from_keywords(search_keywords, url=None):
@@ -80,22 +81,29 @@ def write_blog_from_keywords(search_keywords, url=None):
             logger.error(f"Failed to do Tavily AI research: {err}")
 
         status.update(label="🙎 Generating - Title, Meta Description, Tags, Categories for the content.")
-        blog_title, blog_meta_desc, blog_tags, blog_categories = blog_metadata(blog_markdown_str, 
+        try:
-                search_keywords, example_blog_titles)
+            blog_title, blog_meta_desc, blog_tags, blog_categories = blog_metadata(blog_markdown_str)
-   
+        except Exception as err:
-        generated_image_filepath = None
+            st.error(f"Failed to get blog metadata: {err}")
+
+        try:
+            generated_image_filepath = generate_image(f"{blog_title} + ' ' + {blog_meta_desc}")
+        except Exception as err:
+            st.error(f"Failed in Image generation: {err}")
 
         saved_blog_to_file = save_blog_to_file(blog_markdown_str, blog_title, blog_meta_desc, 
                             blog_tags, blog_categories, generated_image_filepath)
         status.update(label=f"Saved the content in this file: {saved_blog_to_file}")
         blog_frontmatter = dedent(f"""
         \n---------------------------------------------------------------------
-        title: {blog_title.strip()}\n
+        title: {blog_title}\n
-        categories: [{blog_categories.strip()}]\n
+        categories: [{blog_categories}]\n
-        tags: [{blog_tags.strip()}]\n
+        tags: [{blog_tags}]\n
-        Meta description: {blog_meta_desc.replace(":", "-").strip()}\n
+        Meta description: {blog_meta_desc.replace(":", "-")}\n
         ---------------------------------------------------------------------\n
         """)
         logger.info(f"\n\n --------- Finished writing Blog for : {search_keywords} -------------- \n")
-        st.markdown(f"{blog_frontmatter}\n\n{blog_markdown_str}")
+        st.markdown(f"{blog_frontmatter"}
+        st.image(generated_image_filepath)
+        st.markdown(f"{blog_markdown_str}")
         status.update(label=f"Finished, Review & Use your Original Content Below: {saved_blog_to_file}")

lib/blog_metadata/get_blog_metadata.py

@@ -1,26 +1,58 @@
 import sys
-
+import configparser
-from .get_blog_meta_desc import generate_blog_description
+import json
-from .get_tags import get_blog_tags
-from .get_blog_category import get_blog_categories
-from .get_blog_title import generate_blog_title
 
 from loguru import logger
 logger.remove()
 logger.add(sys.stdout,
         colorize=True,
         format="<level>{level}</level>|<green>{file}:{line}:{function}</green>| {message}"
-    )
+    )   
+
+from ..gpt_providers.text_generation.main_text_generation import llm_text_gen
 
 
-def blog_metadata(blog_content, search_keywords=None, blog_titles=None):
+def blog_metadata(blog_content):
     """ Common function to get blog metadata """
-    blog_title = generate_blog_title(blog_content, search_keywords, blog_titles)
+    logger.info(f"Generating Content MetaData\n")
-    blog_meta_desc = generate_blog_description(blog_content)
-    logger.info(f"The blog meta description is: {blog_meta_desc}\n")
-    blog_tags = get_blog_tags(blog_content)
-    logger.info(f"Blog tags for generated content: {blog_tags}")
-    blog_categories = get_blog_categories(blog_content)
-    logger.info(f"Generated blog categories: {blog_categories}\n")
 
-    return(blog_title, blog_meta_desc, blog_tags, blog_categories)
+    blog_metadata_prompt = """
+		As an expert SEO and content writer, I will provide you with blog content.
+		
+		1. Suggest only 2 blog categories which are most relevant to the provided blog content, by identifying the main topic. 
+        Also consider the target audience and the blog's category taxonomy. Only reply with comma-separated values.
+		2. Compose a compelling meta description for the given blog content, adhering to SEO best practices. 
+        Keep it between 150-160 characters. Provide a glimpse of the content's value to entice readers. 
+        Respond with only one of your best efforts and do not include your explanations.
+		3. Write 1 blog title following SEO best practices. Please keep the title concise, not exceeding 60 words. 
+        Respond with only 1 title and no explanations. Negative Keywords: Unveiling, unleash, power of. Don't use such words in your title.
+		4. Suggest only 2 relevant and specific blog tags for the given blog content. Only reply with comma-separated values.
+		
+		The blog content is: '{blog_article}'
+		
+		Please provide the result in the following JSON format:
+		
+		{
+		  "title": "Your generated blog title",
+		  "meta_description": "Your generated meta description",
+		  "tags": ["tag1", "tag2"],
+		  "categories": ["category1", "category2"]
+		}
+		"""
+    try:
+        response = llm_text_gen(blog_metadata_prompt)
+        """ Cleans the response by removing ``` and 'json' strings """
+        result_json = response.replace("```", "").replace("json", "").strip()
+        # Convert the cleaned response to JSON
+        result_json = json.loads(result_json)
+    except Exception as err:
+        logger.error(f"Failed to get response from LLM: {err}")
+        raise err
+
+    # Extract the data from the JSON response
+    blog_title = result_json.get("title")
+    blog_meta_desc = result_json.get("meta_description")
+    blog_tags = result_json.get("tags")
+    blog_categories = result_json.get("categories")
+
+    return blog_title, blog_meta_desc, blog_tags, blog_categories

lib/gpt_providers/text_to_image_generation/gen_stabl_diff_img.py

@@ -45,7 +45,7 @@ def generate_stable_diffusion_image(prompt):
         raise Exception("Non-200 response: " + str(response.text))
     
     data = response.json()
-    save_generated_image(data)
+    img_path = save_generated_image(data)
 
     for i, image in enumerate(data["artifacts"]):
         # Decode base64 image data
@@ -53,4 +53,6 @@ def generate_stable_diffusion_image(prompt):
         # Open image using PIL
         img = Image.open(BytesIO(img_data))
         # Display the image
-        img.show()  
+        img.show()
+
+    return img_path

lib/gpt_providers/text_to_image_generation/main_generate_image_from_prompt.py

@@ -26,7 +26,7 @@ from .gen_stabl_diff_img import generate_stable_diffusion_image
 from ..text_generation.main_text_generation import llm_text_gen
 
 
-def generate_image(user_prompt, image_engine):
+def generate_image(user_prompt):
     """
     The generation API endpoint creates an image based on a text prompt.
 
@@ -42,12 +42,15 @@ def generate_image(user_prompt, image_engine):
     Must be one of "url" or "b64_json". Defaults to "url".
     --> user (str): A unique identifier representing your end-user, which will help OpenAI to monitor and detect abuse.
     """
+    # FIXME: Need to remove default value to match sidebar input.
+    image_engine = 'Stability-AI'
+
     try:
         img_prompt = generate_img_prompt(user_prompt)
         if 'Dalle3' in image_engine:
             logger.info(f"Calling Dalle3 text-to-image with prompt: {img_prompt}")
             image_stored_at = generate_dalle3_images(img_prompt)
-        elif 'Stability-Stable-Diffusion' in image_engine:
+        elif 'Stability-AI' in image_engine:
             logger.info(f"Calling Stable diffusion text-to-image with prompt: \n{img_prompt}")
             print("\n\n")
             image_stored_at = generate_stable_diffusion_image(img_prompt)