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

Feature: AI SEO - Generate rich snippet from url

Browse Source
This commit is contained in:
ajaysi
2024-07-12 19:05:20 +05:30
parent 7abc396633
commit e06c4ffae3
5 changed files with 153 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

128
lib/ai_seo_tools/seo_structured_data.py Normal file
View File

@@ -0,0 +1,128 @@
import streamlit as st
import json
from datetime import date
from ..ai_web_researcher.firecrawl_web_crawler import scrape_url
from ..gpt_providers.text_generation.main_text_generation import llm_text_gen
# Define a dictionary for schema types
schema_types = {
"Article": {
"fields": ["Headline", "Author", "Date Published", "Keywords"],
"schema_type": "Article",
},
"Product": {
"fields": ["Name", "Description", "Price", "Brand", "Image URL"],
"schema_type": "Product",
},
"Recipe": {
"fields": ["Name", "Ingredients", "Cooking Time", "Serving Size", "Image URL"],
"schema_type": "Recipe",
},
"Event": {
"fields": ["Name", "Start Date", "End Date", "Location", "Description"],
"schema_type": "Event",
},
"LocalBusiness": {
"fields": ["Name", "Address", "Phone Number", "Opening Hours", "Image URL"],
"schema_type": "LocalBusiness",
},
# ... (add more schema types as needed)
}
def generate_json_data(content_type, details, url):
"""Generates structured data (JSON-LD) based on user input."""
try:
scraped_text = scrape_url(url)
#logger.info(scraped_text)
except Exception as err:
st.error(f"Failed to scrape web page from url-{weburl} - Error: {err}")
st.stop()
schema = schema_types.get(content_type)
if not schema:
st.error(f"Invalid content type: {content_type}")
return
data = {
"@context": "https://schema.org",
"@type": schema["schema_type"],
}
for field in schema["fields"]:
value = details.get(field)
if isinstance(value, date):
value = value.isoformat()
if value:
data[field] = value
else:
data[field] = "N/A" # Use placeholder values if input is missing
if url:
data['url'] = url
llm_structured_data = get_llm_structured_data(content_type, data, scraped_text)
return llm_structured_data
def get_llm_structured_data(content_type, data, scraped_text):
""" Function to get structured data from LLM """
prompt = f"""Given the following information:
HTML Content: <<<HTML>>> {scraped_text} <<<END_HTML>>>
Content Type: <<<CONTENT_TYPE>>> {content_type} <<<END_CONTENT_TYPE>>>
Additional Relevant Data: <<<ADDITIONAL_DATA>>> {data} <<<END_ADDITIONAL_DATA>>>
Create a detailed structured data (JSON-LD) script for SEO purposes.
The structured data should help search engines understand the content and features of the webpage, enhancing its visibility and potential for rich snippets in search results.
Detailed Steps:
Parse the HTML content to extract relevant information like the title, main heading, and body content.
Use the contentType to determine the structured data type (e.g., Article, Product, Recipe).
Integrate the additional relevant data (e.g., author, datePublished, keywords) into the structured data.
Ensure all URLs, images, and other attributes are correctly formatted and included.
Validate the generated JSON-LD to ensure it meets schema.org standards and is free of errors.
Expected Output:
Generate a JSON-LD structured data snippet based on the provided inputs."""
try:
response = llm_text_gen(prompt)
return response
except Exception as err:
st.error(f"Exit: Failed to get response from LLM: {err}")
def ai_structured_data():
st.title("📝 Generate Structured Data for SEO 🚀")
st.markdown("**Make your content more discoverable with rich snippets.**")
content_type = st.selectbox("**Select Content Type**", list(schema_types.keys()))
details = {}
schema_fields = schema_types[content_type]["fields"]
num_fields = len(schema_fields)
url = st.text_input("**URL :**", placeholder="Enter the URL of your webpage")
for i in range(0, num_fields, 2):
cols = st.columns(2)
for j in range(2):
if i + j < num_fields:
field = schema_fields[i + j]
if "Date" in field:
details[field] = cols[j].date_input(field)
else:
details[field] = cols[j].text_input(field, placeholder=f"Enter {field.lower()}")
if st.button("Generate Structured Data"):
structured_data = generate_json_data(content_type, details, url)
if structured_data:
st.subheader("Generated Structured Data (JSON-LD):")
st.markdown(structured_data)
st.download_button(
label="Download JSON-LD",
data=structured_data,
file_name=f"{content_type}_structured_data.json",
mime="application/json",
)

14
lib/utils/alwrity_utils.py
View File

@@ -36,6 +36,7 @@ from lib.ai_writers.youtube_ai_writer import write_yt_title, write_yt_descriptio
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.ai_seo_tools.seo_structured_data import ai_structured_data
from lib.gpt_providers.text_to_image_generation.main_generate_image_from_prompt import generate_image
from lib.content_planning_calender.content_planning_agents_alwrity_crew import ai_agents_planner
@@ -110,6 +111,19 @@ def process_input(input_text, uploaded_file):
return None
def ai_seo_tools():
""" Collection SEO tools for content creators. """
options = [
"Generate Structured Data - Rich Snippet",
"AI SEO Audit",
"Quit"
]
choice = st.selectbox("**👇Select AI SEO Tool:**", options, index=0, format_func=lambda x: f"📝 {x}")
if choice == "Generate Structured Data - Rich Snippet":
ai_structured_data()
def blog_from_keyword():
""" Input blog keywords, research and write a factual blog."""
st.title("Blog Content Writer")

2
lib/workspace/alwrity_config/main_config.json
View File

@@ -2,7 +2,7 @@
"Blog Content Characteristics": {
"Blog Length": "2000",
"Blog Tone": "Casual",
"Blog Demographic": "Digital Marketing",
"Blog Demographic": "Professional",
"Blog Type": "Informational",
"Blog Language": "English",
"Blog Output Format": "markdown"