summedia 24.2.3

SumMedia library is a powerful Python package used for extracting and parsing newspaper articles. It simplifies the process of web scraping, article downloading and working with openai API. The plugin enables various functionalities related to news content personalization and categorization.

---

SumMedia in a News Web package

The SumMedia is a powerful Python tool used for extracting and parsing newspaper articles. It simplifies the process of web scraping, article downloading and working with openai API. The plugin enables various functionalities related to news content personalization and categorization. Here's an overview of its key features and functionalities:

Table of Contents

1. Article Extraction
2. SumMedia for Filtering and Categorizing Articles
3. SumMedia as a Personal Assistant for Reading Articles
4. SumMedia for Generating Post for Social Media
5. Multi-language Support

---

Article Extraction

Download articles from a given URL and extract useful information like the text, authors, publish date, images, videos, and more.

from summedia.fetching_data import (
    get_text,
    get_time_read,
    get_images,
    get_publishing_date,
    get_authors,
    get_title,
    get_movies,
    get_meta_description,
    get_meta_keywords
)

URL = "www.example.url"

text_article = get_text(URL)
time_read = get_time_read(URL, words_per_minute=238)
img_urls = get_images(URL)
publish_date = get_publishing_date(URL)
authors = get_authors(URL)
title = get_title(URL)
movies = get_movies(URL)
meta_description = get_meta_description(URL)
meta_keywords = get_meta_keywords(URL)

---

Filtering and Categorizing Articles

The work can involve using ChatGPT to analyze and filter news or inappropriate content. You can also develop an algorithm for categorizing articles based on topic, location, date, and other factors.

import os
from summedia.text import Text

txt = Text(api_key=os.environ.get("OPENAI_API_KEY"))
tag_and_categorize_text = txt.tag_and_categorize_text("your text here", model_type="gpt-3.5-turbo-1106")

---

Personal Assistant for Reading Articles

Browse various news websites, fetch article headlines and brief summaries, and then deliver them in a user-friendly manner.

import os
from summedia.text import Text
from summedia.level import SimplificationLevel

text = Text(api_key=os.environ.get("OPENAI_API_KEY"))
summary_article = text.summarize_text("www.example.url", max_number_words=150, model_type="gpt-3.5-turbo-1106")
analyze_sentiment = text.analyze_sentiment("www.example.url", max_number_words=150, model_type="gpt-3.5-turbo-1106")
to_bullet_list = text.to_bullet_list("www.example.url", model_type="gpt-3.5-turbo-1106")
adjust_text_complexity = text.adjust_text_complexity("www.example.url", level = SimplificationLevel.STUDENT, model_type="gpt-3.5-turbo-1106")

---

Generating Post for Social Media

Automate posts to Twitter/X and facebook by just specifying the url for article.

import os
from summedia.social_media import SocialMedia

social_media = SocialMedia(api_key=os.environ.get("OPENAI_API_KEY"))

post_to_facebook = social_media.post_to_facebook(
    "your text here", model_type="gpt-3.5-turbo-1106"
)

condense_text_to_tweet = social_media.condense_text_to_tweet(
    "your text here", model_type="gpt-3.5-turbo-1106"
)

---

Multi-language Support

Handling articles in different languages, making it versatile for international applications. You can also use it as an article translator.

import os
from summedia.text import Text

txt = Text(api_key=os.environ.get("OPENAI_API_KEY"))
translate_text = txt.translate_text("your text here", model_type="gpt-3.5-turbo-1106", language_to_translate="en")

---

Create your own prompt

Create a prompt tailored to your needs.

import os
from summedia.elastic import ElasticAPIRequester

your_prompt = ElasticAPIRequester(api_key=os.environ.get("OPENAI_API_KEY"))
content_system_prompt = "YOUR SYSTEM PROMPT HERE"
content_user_prompt = "YOUR USER PROMPT HERE"
elastic_prompt_result = your_prompt.elastic_prompt(content_system_prompt, content_user_prompt,  model_type="gpt-3.5-turbo-1106")

---

Requirements & Costs

You'll need a paid OpenAI account and an API key.

Check out more here: https://openai.com/pricing

Installation

pip install summedia

How to run tests

pytest --cov=summedia