utils / base_utils.py
not-lain's picture
add background removal & restructure code into multiple files
c577758
from pptx import Presentation
from pdf2image import convert_from_path
import pdfplumber
from docx import Document
import subprocess
import os
from typing import Optional, List
import string
import random
import re
import requests
from bs4 import BeautifulSoup
import logging
import time
from urllib.parse import urlparse
class URLTextExtractor:
"""
A comprehensive utility for extracting text content from web pages with advanced features.
Features:
- Rotating User-Agents to mimic different browsers
- Robust error handling and retry mechanism
- Section preservation for maintaining document structure
- Configurable extraction options
- Logging support
Attributes:
USER_AGENTS (list): A comprehensive list of user agent strings to rotate through.
logger (logging.Logger): Logger for tracking extraction attempts and errors.
Example:
>>> extractor = URLTextExtractor()
>>> text = extractor.extract_text_from_url('https://example.com')
>>> print(text)
"""
# Expanded list of user agents including mobile and less common browsers
USER_AGENTS = [
# Desktop Browsers
"Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36",
"Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/15.1 Safari/605.1.15",
"Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:95.0) Gecko/20100101 Firefox/95.0",
# Mobile Browsers
"Mozilla/5.0 (iPhone; CPU iPhone OS 14_6 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/14.1.1 Mobile/15E148 Safari/604.1",
"Mozilla/5.0 (Linux; Android 10; SM-G970F) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.101 Mobile Safari/537.36",
]
def __init__(self, logger=None):
"""
Initialize the URLTextExtractor.
Args:
logger (logging.Logger, optional): Custom logger.
If not provided, creates a default logger.
"""
self.logger = logger or self._create_default_logger()
def _create_default_logger(self):
"""
Create a default logger for tracking extraction process.
Returns:
logging.Logger: Configured logger instance
"""
logger = logging.getLogger(__name__)
logger.setLevel(logging.INFO)
handler = logging.StreamHandler()
formatter = logging.Formatter("%(asctime)s - %(levelname)s - %(message)s")
handler.setFormatter(formatter)
logger.addHandler(handler)
return logger
def _process_element_text(self, element):
"""
Process text within an element, handling anchor tags specially.
Args:
element (bs4.element.Tag): BeautifulSoup element to process
Returns:
str: Processed text with proper spacing
"""
# Replace anchor tags with spaced text
for a_tag in element.find_all("a"):
# Add spaces around the anchor text
a_tag.replace_with(f" {a_tag.get_text(strip=True)} ")
# Get text with separator
return element.get_text(separator=" ", strip=True)
def extract_text_from_url(
self,
url,
max_retries=3,
preserve_sections=True,
min_section_length=30,
allowed_tags=None,
):
"""
Extract text content from a given URL with advanced configuration.
Args:
url (str): The URL of the webpage to extract text from.
max_retries (int, optional): Maximum number of retry attempts. Defaults to 3.
preserve_sections (bool, optional): Whether to preserve section separations. Defaults to True.
min_section_length (int, optional): Minimum length of text sections to include. Defaults to 30.
allowed_tags (list, optional): Specific HTML tags to extract text from.
If None, uses a default set of content-rich tags.
Returns:
str: Extracted text content from the webpage
Raises:
ValueError: If URL cannot be fetched after maximum retries
requests.RequestException: For network-related errors
Examples:
>>> extractor = URLTextExtractor()
>>> text = extractor.extract_text_from_url('https://example.com')
>>> text = extractor.extract_text_from_url('https://example.com', preserve_sections=False)
"""
# Default allowed tags if not specified
if allowed_tags is None:
allowed_tags = [
"p",
"div",
"article",
"section",
"main",
"h1",
"h2",
"h3",
"h4",
"h5",
"h6",
]
# Validate URL
try:
parsed_url = urlparse(url)
if not all([parsed_url.scheme, parsed_url.netloc]):
# raise ValueError("Invalid URL format")
return None
except Exception as e:
self.logger.error(f"URL parsing error: {e}")
raise
for attempt in range(max_retries):
try:
# Randomly select a user agent
headers = {
"User-Agent": random.choice(self.USER_AGENTS),
"Accept-Language": "en-US,en;q=0.9",
"Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,*/*;q=0.8",
}
# Send a GET request to the URL
response = requests.get(
url, headers=headers, timeout=10, allow_redirects=True
)
# Raise an exception for bad status codes
response.raise_for_status()
# Log successful fetch
self.logger.info(f"Successfully fetched URL: {url}")
# Parse the HTML content
soup = BeautifulSoup(response.text, "html.parser")
# Remove unwanted elements
for script in soup(
["script", "style", "head", "header", "footer", "nav"]
):
script.decompose()
# Extract text with section preservation
if preserve_sections:
# Extract text from specified tags
sections = []
for tag in allowed_tags:
for element in soup.find_all(tag):
# Process element text, handling anchor tags
section_text = self._process_element_text(element)
# Only add sections meeting minimum length
if len(section_text) >= min_section_length:
sections.append(section_text)
# Join sections with newline
text = "\n".join(sections)
else:
# If not preserving sections, use modified text extraction
text = " ".join(
self._process_element_text(element)
for tag in allowed_tags
for element in soup.find_all(tag)
)
# Remove excessive whitespace and empty lines
text = "\n".join(
line.strip() for line in text.split("\n") if line.strip()
)
return text
except (requests.RequestException, ValueError) as e:
# Log error details
self.logger.warning(f"Attempt {attempt + 1} failed: {e}")
# If it's the last retry, raise the error
if attempt == max_retries - 1:
self.logger.error(
f"Failed to fetch URL after {max_retries} attempts"
)
raise ValueError(
f"Error fetching URL after {max_retries} attempts: {e}"
)
# Exponential backoff
wait_time = 2**attempt
self.logger.info(f"Waiting {wait_time} seconds before retry")
time.sleep(wait_time)
# Fallback (though this should never be reached due to the raise in the loop)
return None
def extract_text_from_pptx(file_path):
prs = Presentation(file_path)
text_content = []
for slide in prs.slides:
slide_text = []
for shape in slide.shapes:
if hasattr(shape, "text"):
slide_text.append(shape.text)
text_content.append("\n".join(slide_text))
return "\n\n".join(text_content)
def extract_text_from_ppt(file_path):
try:
print("file_path = ", file_path)
# Convert PPT to PPTX using unoconv
pptx_file_path = os.path.splitext(file_path)[0] + ".pptx"
subprocess.run(["unoconv", "-f", "pptx", file_path], check=True)
# Extract text from PPTX
presentation = Presentation(pptx_file_path)
text_content = []
for slide in presentation.slides:
slide_text = []
for shape in slide.shapes:
if hasattr(shape, "text"):
slide_text.append(shape.text)
text_content.append("\n".join(slide_text))
# Remove the converted PPTX file
os.remove(pptx_file_path)
out = "\n\n".join(text_content)
return out
except Exception as e:
print(f"Error extracting text from PPT file: {e}")
return "Error extracting text from PPT file"
# def extract_text_from_ppt_or_pptx(file_path):
# if file_path.endswith(".pptx"):
# return extract_text_from_pptx(file_path)
# elif file_path.endswith(".ppt"):
# return extract_text_from_ppt(file_path)
# else:
# return "Unsupported file type. Please provide a .ppt or .pptx file."
def convert_pdf_to_image(file):
images = convert_from_path(file)
return images
def extract_text_from_pdf(file):
text = ""
with pdfplumber.open(file) as pdf:
for page in pdf.pages:
text += page.extract_text() + "\n"
return text
def extract_text_from_docx(file_path):
text = ""
doc = Document(file_path.name)
for paragraph in doc.paragraphs:
text += paragraph.text + "\n"
return text
def convert_doc_to_text(file_path):
try:
subprocess.run(
["unoconv", "--format", "txt", file_path],
capture_output=True,
text=True,
check=True,
)
txt_file_path = file_path.replace(".doc", ".txt")
with open(txt_file_path, "r") as f:
text = f.read()
text = text.lstrip("\ufeff")
os.remove(txt_file_path)
return text
except subprocess.CalledProcessError as e:
print(f"Error converting {file_path} to text: {e}")
return ""
# function that generates a random string
def generate_random_string(length=23):
characters = string.ascii_letters + string.digits # Includes letters and digits
random_string = "".join(random.choice(characters) for _ in range(length))
return random_string
# function that adds the necessary json fields
def handle_json_output(json_list: list):
n = len(json_list)
for i in range(n):
# not last element
random_string1 = generate_random_string()
random_string2 = generate_random_string()
element = json_list[i]
front = element["frontText"]
back = element["backText"]
element["frontHTML"] = (
f'<div id="element-richtextarea-{random_string1}" style="position:absolute;left:100px;top:50px;width:800px;height:300px;text-align:center;display:flex;align-items:center;font-size:40px;">'
f"<p>{front}</p></div>"
)
element["backHTML"] = (
f'<div id="element-richtextarea-{random_string2}" style="position:absolute;left:100px;top:50px;width:800px;height:300px;text-align:center;display:flex;align-items:center;font-size:40px;">'
f"<p>{back}</p></div>"
)
element["termType"] = "basic"
cloze_matches = re.findall(r"_{2,}", front)
# match only the first one, if there is multiple don't do anything
if (cloze_matches != []) & (len(cloze_matches) <= 2):
# It's a cloze type card
element["termType"] = "cloze"
# inject the back in a span format into the front
def replace_cloze(match):
return f'</p><p><span class="closure">{back}</span></p><p>'
front_html = re.sub(r"_{2,}", replace_cloze, front)
element["frontHTML"] = (
f'<div id="element-richtextarea-{random_string1}" style="position:absolute;left:100px;top:50px;width:800px;height:300px;text-align:center;display:flex;align-items:center;font-size:40px;">'
f"<p>{front_html}</p></div>"
)
def replace_underscores(match):
return f" {back} "
element["frontText"] = re.sub(r"_{2,}", replace_underscores, front)
element["backText"] = ""
element["backHTML"] = (
f'<div id="element-richtextarea-{random_string2}" style="position:absolute;left:100px;top:50px;width:800px;height:300px;text-align:center;display:flex;align-items:center;font-size:40px;">'
f"<p><br></p></div>"
)
return json_list
def sanitize_list_of_lists(text: str) -> Optional[List[List]]:
left = text.find("[")
right = text.rfind("]")
text = text[left : right + 1]
try:
# Safely evaluate the string to a Python object
list_of_lists = eval(text)
if isinstance(list_of_lists, list): # Ensure it's a list
out = []
try:
# parse list of lists
for front, back in list_of_lists:
out.append({"frontText": front, "backText": back})
return handle_json_output(out)
# errors
except Exception as e:
print(e)
# return anything that was already parsed
if out != []:
return handle_json_output(out)
# original schedma is not respected
else:
return None
else:
print("The evaluated object is not a list.")
return None
except Exception as e:
print(f"Error parsing the list of lists: {e}")
return None
extractor = URLTextExtractor()
def parse_url(url):
return extractor.extract_text_from_url(url)