kunthawat/ALwrity
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Blog writer enhancements & fixes

Browse Source
This commit is contained in:
ajaysi
2025-04-29 08:55:47 +05:30
parent ef462f05f2
commit 9db20db0d1
45 changed files with 3000 additions and 3290 deletions
Download Patch File Download Diff File Expand all files Collapse all files

257
lib/utils/alwrity_utils.py
View File

@@ -1,272 +1,25 @@
import re
import os
import PyPDF2
import tiktoken
import openai
import streamlit as st
import tempfile
from loguru import logger
from lib.ai_web_researcher.gpt_online_researcher import gpt_web_researcher
from lib.ai_writers.keywords_to_blog_streamlit import write_blog_from_keywords
from lib.ai_writers.speech_to_blog.main_audio_to_blog import generate_audio_blog
from lib.ai_writers.long_form_ai_writer import long_form_generator
from lib.ai_writers.ai_news_article_writer import ai_news_generation
#from lib.ai_writers.ai_agents_crew_writer import ai_agents_writers
from lib.ai_writers.ai_financial_writer import write_basic_ta_report
from lib.ai_writers.ai_facebook_writer.facebook_ai_writer import facebook_main_menu
from lib.ai_writers.linkedin_writer.linkedin_ai_writer import linkedin_main_menu
from lib.ai_writers.twitter_writers.twitter_dashboard import run_dashboard
from lib.ai_writers.insta_ai_writer import insta_writer
from lib.ai_writers.youtube_writers.youtube_ai_writer import youtube_main_menu
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_essay_writer import ai_essay_generator
from lib.gpt_providers.text_to_image_generation.main_generate_image_from_prompt import generate_image
from lib.utils.voice_processing import record_voice
#from lib.content_planning_calender.content_planning_agents_alwrity_crew import ai_agents_content_planner
from lib.gpt_providers.text_generation.main_text_generation import llm_text_gen
def is_youtube_link(text):
if text is not None:
youtube_regex = re.compile(r'(https?://)?(www\.)?(youtube|youtu|youtube-nocookie)\.(com|be)/(watch\?v=|embed/|v/|.+\?v=)?([^&=%\?]{11})')
return youtube_regex.match(text)
def is_web_link(text):
if text is not None:
web_regex = re.compile(r'(https?://)?(www\.)?[-a-zA-Z0-9@:%._\+~#=]{1,256}\.[a-zA-Z0-9()]{1,6}\b([-a-zA-Z0-9()@:%_\+.~#?&//=]*)')
return web_regex.match(text)
def process_input(input_text, uploaded_file):
if input_text and is_youtube_link(input_text):
if input_text.startswith("https://www.youtube.com/") or input_text.startswith("http://www.youtube.com/"):
return "youtube_url"
else:
st.error("Invalid YouTube URL. Please enter a valid URL.")
return None
elif input_text and is_web_link(input_text):
return "web_url"
elif input_text:
return "keywords"
if uploaded_file is not None:
file_details = {"filename": uploaded_file.name, "filetype": uploaded_file.type}
st.write(file_details)
if uploaded_file.type.startswith("text/"):
content = uploaded_file.read().decode("utf-8")
st.text(content)
elif uploaded_file.type == "application/pdf":
return "PDF_file"
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)
return "video_file"
return None
def blog_from_keyword():
""" Input blog keywords, research and write a factual blog."""
st.header("Blog Content Writer")
col1, col2, col3 = st.columns([2, 1.5, 0.5])
with col1:
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.
- Attach file: Attach Text, Audio, Video, Image file to blog on.
- YouTube Link: Provide a YouTube video link to convert into blog.
- Web URLs: Provide web URL to write similar blog on.
- Provide Local folder location with your documents to use for content creation.""")
with col2:
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.')
with col3:
audio_input = record_voice()
if audio_input:
st.info(audio_input)
# Validate the provided folder path
#st.info("🚨 Currently supported file formats are: PDF, plain text, CSV, Excel, Markdown, PowerPoint, and Word documents.")
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"])
# Add an expandable section for advanced writing options
with st.expander("Advanced Writing Options", expanded=False):
# Option 1: Select content type
content_type = st.radio("**👇 Select content type:**",
["Normal-length content", "Long-form content", "Experimental - AI Agents team"])
# Option 2: Checkbox for 'Create SEO tags' (Checked by default)
create_seo_tags = st.checkbox('Create SEO tags', value=True,
help='Generate json-ld schema, Twitter, and Facebook tags.')
# Option 3: Checkbox for 'Generate Social Media content' (Unchecked by default)
generate_social_media = st.checkbox('Generate Social Media content', value=False,
help="Write Facebook, Instagram posts & tweets for generated blog. Needed for marketing your blogs.")
# Option 4: Checkbox for 'Do Content Analysis & Critique' (Unchecked by default)
content_analysis = st.checkbox('Do Content Analysis & Critique', value=False,
help="Blog Proof reading, Critique generated blog. Provide actionable changes & Editing options.")
# Display a message at the bottom for user guidance
st.info("🚨 Make sure to personalize content from the sidebar. Important.")
if st.button("Write Blog"):
# Clear the previous results from the screen
st.empty()
if user_input == "":
user_input = None
if not uploaded_file and not user_input and not audio_input:
st.error("🤬🤬 Either Enter/Type/Attach, can't read your mind.(yet..)")
st.stop()
else:
input_type = process_input(user_input, uploaded_file)
if input_type == "keywords":
if user_input and len(user_input.split()) >= 2:
if content_type == "Normal-length content":
try:
short_blog = write_blog_from_keywords(user_input)
st.markdown(short_blog)
except Exception as err:
st.error(f"🚫 Failed to write blog on {user_input}, Error: {err}")
elif content_type == "Long-form content":
try:
long_form_generator(user_input)
st.success(f"Successfully wrote long-form blog on: {user_input}")
except Exception as err:
st.error(f"🚫 Failed to write blog on {user_input}, Error: {err}")
elif content_type == "Experimental - AI Agents team":
try:
ai_agents_writers(user_input)
st.success(f"Successfully wrote content with AI agents on: {user_input}")
except Exception as err:
st.error(f"🚫 Failed to Write content with AI agents: {err}")
else:
st.error('🚫 Blog keywords should be at least two words long. Please try again.')
elif input_type == "youtube_url" or input_type == "audio_file":
if not generate_audio_blog(user_input):
st.stop()
elif input_type == "web_url":
blog_from_url(user_input)
elif input_type == "image_file":
blog_from_image(user_input, temp_file_path)
elif input_type == "PDF_file":
pdf_reader = PyPDF2.PdfReader(uploaded_file)
text = ""
combined_result = ""
# Create a placeholder for the progress bar
progress_bar = st.progress(0)
# Loop through each page with a progress bar
for page_num, page in enumerate(pdf_reader.pages):
text += page.extract_text()
# Replace newlines with spaces
text = text.replace("\n", " ")
# Use regex to add a space between words that are combined
text = re.sub(r"(\w)([A-Z])", r"\1 \2", text)
results = blog_from_pdf(text)
# Update the progress bar
progress_bar.progress((page_num + 1) / len(pdf_reader.pages))
combined_result += str(results[-1])
# Clear progress bar at the end
progress_bar.empty()
st.markdown(combined_result)
def blog_from_pdf(pdf_text):
""" Load in a long PDF and pull the text out. Create a prompt to be used to extract key bits of information.
Chunk up our document and process each chunk to pull any answers out. Combine them at the end.
This simple approach will then be extended to three more difficult questions.
"""
# FixME:
document = '<document>'
template_prompt=f'''Extract key pieces of information from the given document.
When you extract a key piece of information, include the closest page number.
Ex: Extracted Information (Page number)
\n\nDocument: \"\"\"<document>\"\"\"\n\n'''
# Initialise tokenizer
tokenizer = tiktoken.get_encoding("cl100k_base")
results = []
chunks = create_chunks(pdf_text, 1000, tokenizer)
text_chunks = [tokenizer.decode(chunk) for chunk in chunks]
for chunk in text_chunks:
results.append(extract_chunk(chunk, template_prompt))
#zipped = list(zip(*groups))
#zipped = [x for y in zipped for x in y if "Not specified" not in x and "__" not in x]
return results
# Split a text into smaller chunks of size n, preferably ending at the end of a sentence
def create_chunks(text, n, tokenizer):
tokens = tokenizer.encode(text)
"""Yield successive n-sized chunks from text."""
i = 0
while i < len(tokens):
# Find the nearest end of sentence within a range of 0.5 * n and 1.5 * n tokens
j = min(i + int(1.5 * n), len(tokens))
while j > i + int(0.5 * n):
# Decode the tokens and check for full stop or newline
chunk = tokenizer.decode(tokens[i:j])
if chunk.endswith(".") or chunk.endswith("\n"):
break
j -= 1
# If no end of sentence found, use n tokens as the chunk size
if j == i + int(0.5 * n):
j = min(i + n, len(tokens))
yield tokens[i:j]
i = j
def extract_chunk(document, template_prompt):
""" Chunking for large documents, exceed context window"""
client = openai.OpenAI(api_key=os.environ.get("OPENAI_API_KEY"))
prompt = template_prompt.replace('<document>', document)
try:
response = llm_text_gen(prompt)
return response
except Exception as err:
logger.error(f"Exit: Failed to get response from LLM: {err}")
exit(1)
def ai_agents_team():
# Define options for AI Content Teams
st.title("🐲 Your AI Agents Teams")
@@ -316,9 +69,9 @@ def content_agents():
if content_keywords and len(content_keywords.split()) >= 2:
with st.spinner("Generating Content..."):
try:
calendar_content = ai_agents_writers(content_keywords)
st.success(f"Successfully generated content for: {content_keywords}")
st.markdown(calendar_content)
#calendar_content = ai_agents_writers(content_keywords)
st.success(f"🚫 Not implemented yet: {content_keywords}")
#st.markdown(calendar_content)
except Exception as err:
st.error(f"🚫 Failed to generate content with AI Agents: {err}")
else:
@@ -474,4 +227,4 @@ def ai_social_writer():
elif "instagram" in selected_platform:
insta_writer()
elif "youtube" in selected_platform:
youtube_main_menu()
youtube_main_menu()

50
lib/utils/content_generators.py
View File

@@ -1,57 +1,11 @@
import streamlit as st
from lib.utils.alwrity_utils import (
blog_from_keyword, ai_agents_team, essay_writer, ai_news_writer,
ai_finance_ta_writer
)
from lib.alwrity_ui.similar_analysis import competitor_analysis
from lib.alwrity_ui.keyword_web_researcher import do_web_research
from lib.ai_writers.ai_story_writer.story_writer import story_input_section
from lib.ai_writers.ai_product_description_writer import write_ai_prod_desc
from lib.ai_writers.ai_copywriter.copywriter_dashboard import copywriter_dashboard
from lib.ai_writers.linkedin_writer import LinkedInAIWriter
#from lib.content_planning_calender.content_planning_agents_alwrity_crew import ai_agents_content_planner
def ai_writers():
options = [
"AI Blog Writer",
"Story Writer",
"Essay writer",
"Write News reports",
"Write Financial TA report",
"AI Product Description Writer",
"AI Copywriter",
"LinkedIn AI Writer",
"Quit"
]
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()
elif choice == "Story Writer":
story_input_section()
elif choice == "Essay writer":
essay_writer()
elif choice == "Write News reports":
ai_news_writer()
elif choice == "Write Financial TA report":
ai_finance_ta_writer()
elif choice == "AI Product Description Writer":
write_ai_prod_desc()
elif choice == "AI Copywriter":
# Initialize the copywriter dashboard
copywriter_dashboard()
elif choice == "LinkedIn AI Writer":
# Initialize the LinkedIn AI Writer
linkedin_writer = LinkedInAIWriter()
linkedin_writer.run()
elif choice == "Quit":
st.info("Thank you for using Alwrity. Goodbye!")
st.stop()
def content_planning_tools():
# Add custom CSS for compact layout
# A custom CSS for compact layout
st.markdown("""
<style>
/* Reduce top padding of main container */

75
lib/utils/ui_setup.py
View File

@@ -1,10 +1,11 @@
import os
import streamlit as st
from lib.utils.file_processor import load_image
from lib.utils.content_generators import content_planning_tools, ai_writers
from lib.utils.content_generators import content_planning_tools
from lib.utils.alwrity_utils import ai_social_writer
from lib.utils.seo_tools import ai_seo_tools
from lib.utils.settings_page import render_settings_page
from loguru import logger
# Import social media writer functions
from lib.ai_writers.ai_facebook_writer.facebook_ai_writer import facebook_main_menu
@@ -12,6 +13,7 @@ from lib.ai_writers.linkedin_writer.linkedin_ai_writer import linkedin_main_menu
from lib.ai_writers.twitter_writers import run_dashboard
from lib.ai_writers.insta_ai_writer import insta_writer
from lib.ai_writers.youtube_writers.youtube_ai_writer import youtube_main_menu
from lib.ai_writers.ai_writer_dashboard import get_ai_writers, list_ai_writers
def setup_ui():
@@ -295,22 +297,26 @@ def setup_ui():
def setup_alwrity_ui():
"""Sets up the main navigation in the sidebar."""
logger.info("Setting up ALwrity UI")
# Initialize session state for active tab if not exists
if 'active_tab' not in st.session_state:
st.session_state.active_tab = "Content Planning"
logger.info(f"Initialized active_tab to: {st.session_state.active_tab}")
# Initialize session state for active sub-tab if not exists
if 'active_sub_tab' not in st.session_state:
st.session_state.active_sub_tab = None
logger.info("Initialized active_sub_tab to None")
# Define the navigation items with their icons and functions
nav_items = {
"AI Writers": ("📝", get_ai_writers),
"Content Planning": ("📅", content_planning_tools),
"AI Writers": ("📝", ai_writers),
"Agents Teams": ("🤝", lambda: st.subheader("Agents Teams - Coming Soon!")),
"AI SEO Tools": ("🔍", ai_seo_tools),
"AI Social Tools": ("📱", None), # Set to None as we'll handle this separately
"Ask Alwrity": ("💬", lambda: (
"Agents Teams(TBD)": ("🤝", lambda: st.subheader("Agents Teams - Coming Soon!")),
"Ask Alwrity(TBD)": ("💬", lambda: (
st.subheader("Chat with your Data, Chat with any Data.. COMING SOON !"),
st.markdown("Create a collection by uploading files (PDF, MD, CSV, etc), or crawl a data source (Websites, more sources coming soon."),
st.markdown("One can ask/chat, summarize and do semantic search over the uploaded data")
@@ -318,6 +324,8 @@ def setup_alwrity_ui():
"ALwrity Settings": ("⚙️", render_settings_page)
}
logger.info(f"Defined {len(nav_items)} navigation items")
# Define sub-menu items for AI Social Tools
social_tools_submenu = {
"Facebook": ("📘", lambda: facebook_main_menu()),
@@ -326,6 +334,8 @@ def setup_alwrity_ui():
"Instagram": ("📸", lambda: insta_writer()),
"YouTube": ("🎥", lambda: youtube_main_menu())
}
logger.info(f"Defined {len(social_tools_submenu)} social tools submenu items")
# Create sidebar navigation
st.sidebar.markdown("### ALwrity Options")
@@ -342,6 +352,7 @@ def setup_alwrity_ui():
st.session_state.active_tab = name
# Reset sub-tab when main tab changes
st.session_state.active_sub_tab = None
logger.info(f"Selected main tab: {name}")
# If AI Social Tools is active, show the sub-menu
if st.session_state.active_tab == "AI Social Tools":
@@ -367,6 +378,7 @@ def setup_alwrity_ui():
if st.sidebar.button(f"{sub_icon} {sub_name}", key=button_key,
help=f"Navigate to {sub_name}", use_container_width=True):
st.session_state.active_sub_tab = sub_name
logger.info(f"Selected social tool: {sub_name}")
# Close the div
st.sidebar.markdown('</div>', unsafe_allow_html=True)
@@ -379,6 +391,7 @@ def setup_alwrity_ui():
st.session_state.active_tab = name
# Reset sub-tab when main tab changes
st.session_state.active_sub_tab = None
logger.info(f"Selected main tab: {name}")
st.sidebar.markdown('</div>', unsafe_allow_html=True)
@@ -427,13 +440,47 @@ def setup_alwrity_ui():
# Call the function directly without any title
social_tools_submenu[st.session_state.active_sub_tab][1]()
else:
st.markdown("""
<style>
.main .block-container {
padding-top: 0.25rem !important;
padding-bottom: 0;
}
</style>
""", unsafe_allow_html=True)
st.title(f"{nav_items[st.session_state.active_tab][0]} {st.session_state.active_tab}")
nav_items[st.session_state.active_tab][1]()
# Check if we're in the AI Writers section and handle writer selection
if st.session_state.active_tab == "AI Writers":
# Get the writer parameter from the URL using st.query_params
writer = st.query_params.get("writer")
logger.info(f"Current writer from query params: {writer}")
if writer:
# Get the list of writers without rendering the dashboard
writers = list_ai_writers()
logger.info(f"Found {len(writers)} writers")
writer_found = False
for w in writers:
logger.info(f"Checking writer: {w['name']} with path: {w['path']}")
if w["path"] == writer:
writer_found = True
logger.info(f"Found matching writer: {w['name']}, executing function")
# Clear any existing content
st.empty()
# Execute the writer function
w["function"]()
break
if not writer_found:
logger.error(f"No writer found with path: {writer}")
st.error(f"No writer found with path: {writer}")
else:
# If no writer selected, show the dashboard
logger.info("No writer selected, showing dashboard")
get_ai_writers()
else:
# For all other tabs, show the title
st.markdown("""
<style>
.main .block-container {
padding-top: 0.25rem !important;
padding-bottom: 0;
}
</style>
""", unsafe_allow_html=True)
st.title(f"{nav_items[st.session_state.active_tab][0]} {st.session_state.active_tab}")
nav_items[st.session_state.active_tab][1]()
logger.info("Finished setting up ALwrity UI")