WIP - Streamlit UI, firecrawl - V0.5

2024-06-13 17:43:57 +05:30
parent 8dfcb1f536
commit 128b6f3878
6 changed files with 196 additions and 40 deletions
--- a/alwrity.py
+++ b/alwrity.py
@@ -307,6 +307,7 @@ def main():
        Welcome to Alwrity!
    </div>
    """, unsafe_allow_html=True)
+    
    # Export the paths and file names. Dont want alwrity to be chatty and prompt for inputs.
    os.environ["SEARCH_SAVE_FILE"] = os.path.join(os.getcwd(), "lib", "workspace", "alwrity_web_research",
                                                  f"web_research_report_{datetime.now().strftime('%Y-%m-%d_%H-%M-%S')}")
@@ -375,7 +376,7 @@ def write_blog():
        "AI Copywriter",
        "Quit"
    ]
-    choice = st.selectbox("**Select a content creation type:**", options, index=0, format_func=lambda x: f"📝 {x}")
+    choice = st.selectbox("**👇Select a content creation type:**", options, index=0, format_func=lambda x: f"📝 {x}")

    if choice == "AI Blog Writer":
        blog_from_keyword()
--- a/lib/ai_writers/image_ai_writer.py
+++ b/lib/ai_writers/image_ai_writer.py
@@ -0,0 +1,112 @@
+import sys
+import os
+
+from textwrap import dedent
+from PIL import Image
+import json
+from pathlib import Path
+from datetime import datetime
+import streamlit as st
+
+from dotenv import load_dotenv
+load_dotenv(Path('../../.env'))
+from loguru import logger
+logger.remove()
+logger.add(sys.stdout,
+        colorize=True,
+        format="<level>{level}</level>|<green>{file}:{line}:{function}</green>| {message}"
+    )
+
+from ..ai_web_researcher.firecrawl_web_crawler import scrape_url
+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
+from ..gpt_providers.text_generation.main_text_generation import llm_text_gen
+
+import google.generativeai as genai
+
+
+def blog_from_image(prompt, uploaded_img):
+    """
+    This function will take a blog Topic to first generate sections for it
+    and then generate content for each section.
+    """
+    # Use to store the blog in a string, to save in a *.md file.
+    blog_markdown_str = None
+    logger.info(f"Researching and Writing Blog on {uploaded_img} and {prompt}")
+    # FIXME: Implement support for Openai.
+    if not os.getenv("GEMINI_API_KEY"):
+        st.error("Only Gemini supported, Open Issue ticket on github for Openai, others.")
+        st.stop()
+
+    with st.status("Started Writing from Image..", expanded=True) as status:
+        st.empty()
+        status.update(label=f"Researching and Writing Blog on given Image")
+        try:
+            blog_markdown_str = write_blog_from_image(prompt, uploaded_img)
+        except Exception as err:
+            st.error(f"Failed to write blog from Image - Error: {err}")
+            logger.error(f"Failed to write blog from image: {err}")
+            st.stop()
+        status.update(label="Successfully wrote blog from image.", expanded=False, state="complete")
+
+        try:
+            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)
+        except Exception as err:
+            st.error(f"Failed to get blog metadata: {err}")
+
+        try:
+            status.update(label="🙎 Generating Image for the new blog.")
+            generated_image_filepath = generate_image(f"{blog_title} + ' ' + {blog_meta_desc}")
+        except Exception as err:
+            st.warning(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}")
+        logger.info(f"\n\n --------- Finished writing Blog -------------- \n")
+        st.image(generated_image_filepath, caption=blog_title)
+        st.markdown(f"{blog_markdown_str}")
+        status.update(label=f"Finished, Review & Use your Original Content Below: {saved_blog_to_file}", state="complete")
+
+        # Clean up the temporary file after processing (optional)
+        os.remove(uploaded_img)
+
+
+def write_blog_from_image(prompt, uploaded_img):
+    """Combine the given online research and GPT blog content"""
+    try:
+        config_path = Path(os.environ["ALWRITY_CONFIG"])
+        with open(config_path, 'r', encoding='utf-8') as file:
+            config = json.load(file)
+    except Exception as err:
+        logger.error(f"Error: Failed to read values from config: {err}")
+        exit(1)
+
+    blog_characteristics = config['Blog Content Characteristics']
+    
+    if not prompt:
+        prompt = f"""
+            As expert Creative Content writer, analyse the given image carefully.
+            I want you to write a detailed {blog_characteristics['Blog Type']} blog post including 5 FAQs.
+        
+            Below are the guidelines to follow:
+            1). You must respond in {blog_characteristics['Blog Language']} language.
+            2). Tone and Brand Alignment: Adjust your tone, voice, personality for {blog_characteristics['Blog Tone']} audience.
+            3). Make sure your response content length is of {blog_characteristics['Blog Length']} words.
+        """
+    logger.info("Generating blog and FAQs from Google web search results.")
+    
+    try:
+        #response = llm_text_gen(prompt)
+        genai.configure(api_key=os.getenv('GEMINI_API_KEY'))
+        version = 'models/gemini-1.5-flash'
+        model = genai.GenerativeModel(version)
+        model_info = genai.get_model(version)
+        print(f'{version} - input limit: {model_info.input_token_limit}, output limit: {model_info.output_token_limit}')
+        response = model.generate_content([prompt, Image.open(uploaded_img)])
+        return response.text
+    except Exception as err:
+        logger.error(f"Exit: Failed to get response from LLM: {err}")
+        exit(1)
--- a/lib/ai_writers/web_url_ai_writer.py
+++ b/lib/ai_writers/web_url_ai_writer.py
@@ -71,19 +71,10 @@ def blog_from_url(weburl):
        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}\n
-        categories: [{blog_categories}]\n
-        tags: [{blog_tags}]\n
-        Meta description: {blog_meta_desc.replace(":", "-")}\n
-        ---------------------------------------------------------------------\n
-        """)
        logger.info(f"\n\n --------- Finished writing Blog for : {weburl} -------------- \n")
-        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}")
+        status.update(label=f"Finished, Review & Use your Original Content Below: {saved_blog_to_file}", state="complete")
        

 def write_blog_from_weburl(scraped_website):
--- a/lib/gpt_providers/text_generation/gemini_pro_text.py
+++ b/lib/gpt_providers/text_generation/gemini_pro_text.py
@@ -29,7 +29,7 @@ def gemini_text_response(prompt, temperature, top_p, n, max_tokens):
    except Exception as err:
        logger.error(f"Failed to configure Gemini: {err}")
    logger.info(f"Temp: {temperature}, MaxTokens: {max_tokens}, TopP: {top_p}, N: {n}")
-    # Set up the model
+    # Set up AI model config
    generation_config = {
        "temperature": temperature,
        "top_p": top_p,
--- a/lib/utils/alwrity_utils.py
+++ b/lib/utils/alwrity_utils.py
@@ -1,6 +1,7 @@
 import os
 import re
 import streamlit as st
+import tempfile
 from pathlib import Path
 import configparser
 from datetime import datetime
@@ -21,6 +22,7 @@ from lib.ai_writers.twitter_ai_writer import tweet_writer
 from lib.ai_writers.insta_ai_writer import insta_writer
 from lib.ai_writers.youtube_ai_writer import write_yt_title, write_yt_description, write_yt_script
 from lib.ai_writers.web_url_ai_writer import blog_from_url
+from lib.ai_writers.image_ai_writer import blog_from_image
 from lib.ai_writers.ai_story_writer import ai_story_generator
 from lib.ai_writers.ai_essay_writer import ai_essay_generator
 from lib.gpt_providers.text_to_image_generation.main_generate_image_from_prompt import generate_image
@@ -61,11 +63,15 @@ def process_input(input_text, uploaded_file):
            st.write("PDF file uploaded. Add your PDF processing logic here.")
        elif uploaded_file.type in ["application/vnd.openxmlformats-officedocument.wordprocessingml.document", "application/msword"]:
            st.write("Word document uploaded. Add your DOCX processing logic here.")
+        
        elif uploaded_file.type.startswith("image/"):
            st.image(uploaded_file)
+            return("image_file")
+
        elif uploaded_file.type.startswith("audio/"):
            st.audio(uploaded_file)
            return("audio_file")
+        
        elif uploaded_file.type.startswith("video/"):
            st.video(uploaded_file)
    
@@ -75,7 +81,7 @@ def blog_from_keyword():
    st.title("Blog Content Writer")
    col1, col2 = st.columns([2, 1.5])
    with col1:
-        user_input = st.text_area('**Enter Keywords/Title/YouTube Link/Web URLs**',
+        user_input = st.text_area('**👇Enter Keywords/Title/YouTube Link/Web URLs**',
                                  help='Provide keywords, titles, YouTube links, or web URLs to generate content.',
                                  placeholder="""Write Blog From:
        - Keywords/Blog Title: Provide keywords to web research & write blog.
@@ -84,11 +90,19 @@ def blog_from_keyword():
        - Web URLs: Provide web URL to write similar blog on.""")

    with col2:
-        uploaded_file = st.file_uploader("**Attach files (Audio, Video, Image, Document)**",
+        uploaded_file = st.file_uploader("**👇Attach files (Audio, Video, Image, Document)**",
                                         type=["txt", "pdf", "docx", "jpg", "jpeg", "png", "mp3", "wav", "mp4", "mkv", "avi"],
                                         help='Attach files such as audio, video, images, or documents.')

-    content_type = st.radio("Select content type:", ["Normal-length content", "Long-form content", "Experimental - AI Agents team"])
+    temp_file_path = None
+    if uploaded_file is not None:
+        # Save the uploaded file to a temporary file
+        with tempfile.NamedTemporaryFile(delete=False, suffix=uploaded_file.name) as temp_file:
+            temp_file.write(uploaded_file.read())
+            temp_file_path = temp_file.name
+
+
+    content_type = st.radio("**👇Select content type:**", ["Normal-length content", "Long-form content", "Experimental - AI Agents team"])
    if st.button("Write Blog"):
        # Clear the previous results from the screen
        st.empty()
@@ -126,6 +140,8 @@ def blog_from_keyword():
            generate_audio_blog(user_input)
        elif 'web_url' in input_type:
            blog_from_url(user_input)
+        elif 'image_file' in input_type:
+            blog_from_image(user_input, temp_file_path)


 def ai_agents_team():
--- a/lib/workspace/alwrity_ui_styling.css
+++ b/lib/workspace/alwrity_ui_styling.css
@@ -3,15 +3,13 @@ body {
    background: #f0f4f8;
    background-image: linear-gradient(to bottom right, #d0e1f9, #e1ebf9);
    font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
-    margin: 0;  /* Remove any default margin */
-    padding: 0;  /* Remove any default padding */
+    margin: 0;
+    padding: 0;
 }
+
 .block-container {
-    padding-top: 2.5rem;
-    padding-bottom: 3rem;
-    padding-left: 5rem;
-    padding-right: 5rem;
-                }
+    padding: 2.5rem 5rem 3rem 5rem;
+}

 /* Main header styling */
 .main-header {
@@ -21,7 +19,7 @@ body {
    margin-bottom: 20px;
    text-align: center;
    text-shadow: 2px 2px 4px rgba(0, 0, 0, 0.2);
-    padding-top: 10px; /* Add padding to the top to move it up */
+    padding-top: 10px;
 }

 /* Sub-header styling */
@@ -36,11 +34,11 @@ body {

 /* Navigation tabs styling */
 .stTabs [role="tab"] {
-    font-size: 3.75em;  /* Increased font size */
+    font-size: 1.25em;
    font-weight: bold;
    color: white;
    background: #1565C0;
-    padding: 16px 24px;  /* Increased padding */
+    padding: 16px 24px;
    margin: 5px;
    border-radius: 8px;
    border: 2px solid #ddd;
@@ -57,11 +55,9 @@ body {
    border: 2px solid #1565C0;
 }

-
-
 /* Sidebar header styling */
 .sidebar-header {
-    font-size: 2.2em;
+    font-size: 1.5em;
    font-weight: bold;
    color: #333;
    margin-bottom: 20px;
@@ -70,8 +66,14 @@ body {
 /* Sidebar option styling */
 .sidebar-option {
    margin-bottom: 10px;
-    font-size: 2.2em;
+    font-size: 1.1em;
    color: #1565C0;
+    cursor: pointer;
+    transition: color 0.3s ease;
+}
+
+.sidebar-option:hover {
+    color: #1976D2;
 }

 /* Content section styling */
@@ -80,11 +82,11 @@ body {
    margin-bottom: 30px;
    border-radius: 10px;
    box-shadow: 2px 2px 10px rgba(0, 0, 0, 0.1);
-    background-color: #f0f4f8;  /* Change background to match body */
+    background-color: #ffffff;
 }

 /* Input fields styling */
-input[type="text"] {
+input[type="text"], input[type="file"], select {
    width: 100%;
    padding: 12px;
    margin: 8px 0;
@@ -92,11 +94,18 @@ input[type="text"] {
    border: 2px solid #ddd;
    border-radius: 4px;
    background-color: #f0f8ff;
+    font-size: 16px;
+    transition: border-color 0.3s ease, box-shadow 0.3s ease;
+}
+
+input[type="text"]:focus, input[type="file"]:focus, select:focus {
+    border-color: #1565C0;
+    box-shadow: 0 0 5px rgba(21, 101, 192, 0.5);
 }

 /* Custom button styling */
 div.stButton > button:first-child {
-    background: #1565C0;  /* Match tab color */
+    background: #1565C0;
    color: white;
    border: none;
    padding: 12px 24px;
@@ -113,7 +122,7 @@ div.stButton > button:first-child {
 }

 div.stButton > button:hover:first-child {
-    background-color: #1976A2;  /* Match tab hover color */
+    background-color: #1976A2;
    box-shadow: 2px 2px 8px rgba(0, 0, 0, 0.3);
 }

@@ -166,20 +175,47 @@ div.row-widget.stRadio > div[role="radiogroup"] > label[data-baseweb="radio"] {
    margin-bottom: 10px;
    color: #333;
 }
+
+/* Audio player styling */
 audio::-webkit-media-controls-panel,
 audio::-webkit-media-controls-enclosure {
-    background-color:#532b5a;}
+    background-color: #532b5a;
+}

 audio::-webkit-media-controls-time-remaining-display,
 audio::-webkit-media-controls-current-time-display {
    color: white;
-    text-shadow: none; 
-
+    text-shadow: none;
 }

 audio::-webkit-media-controls-timeline {
-  background-color: #532b5a;
-  border-radius: 25px;
-  margin-left: 10px;
-  margin-right: 10px;
+    background-color: #532b5a;
+    border-radius: 25px;
+    margin-left: 10px;
+    margin-right: 10px;
 }
+
+/* Select input styling */
+.stSelectbox > div[data-baseweb="select"] > div {
+    padding: 12px;
+    margin: 8px 0;
+    border: 2px solid #ddd;
+    border-radius: 4px;
+    background-color: #f0f8ff;
+    font-size: 16px;
+    transition: border-color 0.3s ease, box-shadow 0.3s ease;
+    appearance: none;
+    background-repeat: no-repeat;
+    background-position: right 10px center;
+    background-size: 12px;
+}
+
+select:focus {
+    border-color: #1565C0;
+    box-shadow: 0 0 5px rgba(21, 101, 192, 0.5);
+}
+
+select option {
+    padding: 10px;
+}
+