MBilal-72's picture
Upload Utils and its files
c76bc58 verified
raw
history blame
20.7 kB
"""
Content Parsing Module
Handles extraction of content from PDFs, text, and webpages
"""
import requests
from bs4 import BeautifulSoup
from urllib.parse import urljoin, urlparse
from typing import List, Dict, Any
import time
from langchain_community.document_loaders import PyPDFLoader
from langchain.schema import Document
class BaseParser:
"""Base class for all content parsers"""
def __init__(self):
self.supported_formats = []
def parse(self, source: str) -> List[Document]:
"""Parse content from source and return LangChain Documents"""
raise NotImplementedError("Subclasses must implement parse method")
def validate_source(self, source: str) -> bool:
"""Validate if the source can be processed"""
return True
class PDFParser(BaseParser):
"""Parser for PDF documents"""
def __init__(self):
super().__init__()
self.supported_formats = ['.pdf']
def parse(self, pdf_path: str) -> List[Document]:
"""
Parse PDF file and return list of Document objects
Args:
pdf_path (str): Path to the PDF file
Returns:
List[Document]: List of parsed documents with metadata
"""
try:
loader = PyPDFLoader(pdf_path)
documents = loader.load_and_split()
# Add additional metadata
for i, doc in enumerate(documents):
doc.metadata.update({
'source_type': 'pdf',
'page_number': i + 1,
'total_pages': len(documents),
'parser': 'PDFParser'
})
return documents
except Exception as e:
raise Exception(f"Error parsing PDF: {str(e)}")
def get_pdf_metadata(self, pdf_path: str) -> Dict[str, Any]:
"""Extract metadata from PDF file"""
try:
loader = PyPDFLoader(pdf_path)
documents = loader.load()
total_pages = len(documents)
total_words = sum(len(doc.page_content.split()) for doc in documents)
return {
'total_pages': total_pages,
'total_words': total_words,
'average_words_per_page': total_words / total_pages if total_pages > 0 else 0,
'file_type': 'PDF',
'parser_used': 'PyPDFLoader'
}
except Exception as e:
return {'error': f"Could not extract metadata: {str(e)}"}
class TextParser(BaseParser):
"""Parser for plain text content"""
def __init__(self):
super().__init__()
self.supported_formats = ['.txt', 'plain_text']
self.chunk_size = 1000 # Default chunk size for long texts
def parse(self, text_content: str, chunk_size: int = None) -> List[Document]:
"""
Parse text content and return list of Document objects
Args:
text_content (str): Raw text content
chunk_size (int): Optional chunk size for splitting long texts
Returns:
List[Document]: List of documents, potentially chunked
"""
try:
if not text_content.strip():
raise ValueError("Empty text content provided")
chunk_size = chunk_size or self.chunk_size
# If text is short, return as single document
if len(text_content) <= chunk_size:
doc = Document(
page_content=text_content,
metadata={
'source_type': 'text',
'word_count': len(text_content.split()),
'char_count': len(text_content),
'chunk_index': 0,
'total_chunks': 1,
'parser': 'TextParser'
}
)
return [doc]
# Split long text into chunks
chunks = self._split_text_into_chunks(text_content, chunk_size)
documents = []
for i, chunk in enumerate(chunks):
doc = Document(
page_content=chunk,
metadata={
'source_type': 'text',
'word_count': len(chunk.split()),
'char_count': len(chunk),
'chunk_index': i,
'total_chunks': len(chunks),
'parser': 'TextParser'
}
)
documents.append(doc)
return documents
except Exception as e:
raise Exception(f"Error parsing text: {str(e)}")
def _split_text_into_chunks(self, text: str, chunk_size: int) -> List[str]:
"""Split text into chunks while preserving sentence boundaries"""
sentences = text.split('. ')
chunks = []
current_chunk = ""
for sentence in sentences:
# Add sentence to current chunk if it fits
test_chunk = current_chunk + sentence + ". "
if len(test_chunk) <= chunk_size:
current_chunk = test_chunk
else:
# Start new chunk if current chunk has content
if current_chunk.strip():
chunks.append(current_chunk.strip())
current_chunk = sentence + ". "
# Add final chunk if it has content
if current_chunk.strip():
chunks.append(current_chunk.strip())
return chunks
def analyze_text_structure(self, text_content: str) -> Dict[str, Any]:
"""Analyze the structure and characteristics of text content"""
try:
lines = text_content.split('\n')
words = text_content.split()
sentences = text_content.split('.')
# Count different elements
paragraphs = [p.strip() for p in text_content.split('\n\n') if p.strip()]
return {
'total_words': len(words),
'total_sentences': len([s for s in sentences if s.strip()]),
'total_lines': len(lines),
'total_paragraphs': len(paragraphs),
'average_words_per_sentence': len(words) / len(sentences) if sentences else 0,
'average_sentences_per_paragraph': len(sentences) / len(paragraphs) if paragraphs else 0,
'character_count': len(text_content),
'reading_time_minutes': len(words) / 200, # Assuming 200 words per minute
'complexity_score': self._calculate_text_complexity(text_content)
}
except Exception as e:
return {'error': f"Could not analyze text structure: {str(e)}"}
def _calculate_text_complexity(self, text: str) -> float:
"""Calculate a simple text complexity score"""
words = text.split()
sentences = [s for s in text.split('.') if s.strip()]
if not sentences:
return 0.0
# Average words per sentence (higher = more complex)
avg_words_per_sentence = len(words) / len(sentences)
# Average characters per word (higher = more complex)
avg_chars_per_word = sum(len(word) for word in words) / len(words) if words else 0
# Simple complexity score (normalized to 1-10 scale)
complexity = (avg_words_per_sentence * 0.1) + (avg_chars_per_word * 0.5)
return min(complexity, 10.0)
class WebpageParser(BaseParser):
"""Parser for web content"""
def __init__(self):
super().__init__()
self.supported_formats = ['http', 'https']
self.headers = {
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36'
}
self.timeout = 10
self.max_retries = 3
def parse_website(self, url: str, max_pages: int = 1, include_subpages: bool = False) -> List[Dict[str, Any]]:
"""
Parse website content and return structured data
Args:
url (str): Website URL to parse
max_pages (int): Maximum number of pages to parse
include_subpages (bool): Whether to include subpages
Returns:
List[Dict]: List of page data with content and metadata
"""
try:
pages_data = []
urls_to_process = [url]
processed_urls = set()
# If including subpages, find additional URLs
if include_subpages and max_pages > 1:
subpage_urls = self._find_subpages(url, max_pages - 1)
urls_to_process.extend(subpage_urls)
# Process each URL
for current_url in urls_to_process[:max_pages]:
if current_url in processed_urls:
continue
page_data = self._parse_single_page(current_url)
if page_data:
pages_data.append(page_data)
processed_urls.add(current_url)
# Add small delay to be respectful
time.sleep(1)
return pages_data
except Exception as e:
raise Exception(f"Error parsing website: {str(e)}")
def _parse_single_page(self, url: str) -> Dict[str, Any]:
"""Parse a single webpage and extract content"""
try:
# Make request with retries
response = None
for attempt in range(self.max_retries):
try:
response = requests.get(url, headers=self.headers, timeout=self.timeout)
response.raise_for_status()
break
except requests.RequestException as e:
if attempt == self.max_retries - 1:
raise e
time.sleep(2 ** attempt) # Exponential backoff
if not response:
return None
# Parse HTML content
soup = BeautifulSoup(response.content, 'html.parser')
# Remove unwanted elements
for element in soup(['script', 'style', 'nav', 'footer', 'header', 'aside']):
element.decompose()
# Extract main content
main_content = self._extract_main_content(soup)
# Extract metadata
title = self._extract_title(soup)
description = self._extract_description(soup)
headings = self._extract_headings(soup)
links = self._extract_links(soup, url)
# Clean and process text
cleaned_text = self._clean_text_content(main_content)
return {
'url': url,
'title': title,
'description': description,
'content': cleaned_text,
'headings': headings,
'internal_links': links['internal'],
'external_links': links['external'],
'word_count': len(cleaned_text.split()),
'char_count': len(cleaned_text),
'meta_keywords': self._extract_meta_keywords(soup),
'images': self._extract_images(soup, url),
'parser': 'WebpageParser',
'parsed_at': time.strftime('%Y-%m-%d %H:%M:%S')
}
except Exception as e:
return {'url': url, 'error': f"Failed to parse page: {str(e)}"}
def _extract_main_content(self, soup: BeautifulSoup) -> str:
"""Extract the main content from the page"""
# Try to find main content in order of preference
content_selectors = [
'main',
'article',
'[role="main"]',
'.content',
'.main-content',
'#content',
'#main',
'.post-content',
'.entry-content'
]
for selector in content_selectors:
element = soup.select_one(selector)
if element:
return element.get_text(separator=' ', strip=True)
# Fallback to body content
body = soup.find('body')
if body:
return body.get_text(separator=' ', strip=True)
return soup.get_text(separator=' ', strip=True)
def _extract_title(self, soup: BeautifulSoup) -> str:
"""Extract page title"""
title_tag = soup.find('title')
if title_tag:
return title_tag.get_text().strip()
# Fallback to h1
h1 = soup.find('h1')
if h1:
return h1.get_text().strip()
return "No Title Found"
def _extract_description(self, soup: BeautifulSoup) -> str:
"""Extract meta description"""
meta_desc = soup.find('meta', attrs={'name': 'description'})
if meta_desc and meta_desc.get('content'):
return meta_desc['content'].strip()
# Fallback to Open Graph description
og_desc = soup.find('meta', attrs={'property': 'og:description'})
if og_desc and og_desc.get('content'):
return og_desc['content'].strip()
return "No Description Found"
def _extract_headings(self, soup: BeautifulSoup) -> List[Dict[str, Any]]:
"""Extract all headings with their hierarchy"""
headings = []
for i in range(1, 7): # h1 to h6
for heading in soup.find_all(f'h{i}'):
text = heading.get_text(strip=True)
if text:
headings.append({
'level': i,
'text': text,
'id': heading.get('id', ''),
'class': heading.get('class', [])
})
return headings
def _extract_links(self, soup: BeautifulSoup, base_url: str) -> Dict[str, List[str]]:
"""Extract internal and external links"""
internal_links = []
external_links = []
base_domain = urlparse(base_url).netloc
for link in soup.find_all('a', href=True):
href = link['href']
full_url = urljoin(base_url, href)
parsed_url = urlparse(full_url)
if parsed_url.netloc == base_domain:
internal_links.append(full_url)
elif parsed_url.netloc: # External link with domain
external_links.append(full_url)
return {
'internal': list(set(internal_links)),
'external': list(set(external_links))
}
def _extract_meta_keywords(self, soup: BeautifulSoup) -> List[str]:
"""Extract meta keywords if available"""
meta_keywords = soup.find('meta', attrs={'name': 'keywords'})
if meta_keywords and meta_keywords.get('content'):
keywords = meta_keywords['content'].split(',')
return [kw.strip() for kw in keywords if kw.strip()]
return []
def _extract_images(self, soup: BeautifulSoup, base_url: str) -> List[Dict[str, str]]:
"""Extract image information"""
images = []
for img in soup.find_all('img'):
src = img.get('src')
if src:
full_url = urljoin(base_url, src)
images.append({
'src': full_url,
'alt': img.get('alt', ''),
'title': img.get('title', '')
})
return images
def _clean_text_content(self, text: str) -> str:
"""Clean and normalize text content"""
if not text:
return ""
# Split into lines and clean each line
lines = text.split('\n')
cleaned_lines = []
for line in lines:
line = line.strip()
if line and len(line) > 1: # Skip empty lines and single characters
cleaned_lines.append(line)
# Join lines with single spaces
cleaned_text = ' '.join(cleaned_lines)
# Remove multiple spaces
while ' ' in cleaned_text:
cleaned_text = cleaned_text.replace(' ', ' ')
return cleaned_text
def _find_subpages(self, url: str, max_subpages: int) -> List[str]:
"""Find subpages from the main page"""
try:
response = requests.get(url, headers=self.headers, timeout=self.timeout)
response.raise_for_status()
soup = BeautifulSoup(response.content, 'html.parser')
base_domain = urlparse(url).netloc
subpages = set()
# Find internal links
for link in soup.find_all('a', href=True):
href = link['href']
full_url = urljoin(url, href)
parsed_url = urlparse(full_url)
# Only include internal links from same domain
if (parsed_url.netloc == base_domain and
full_url != url and
not any(ext in full_url.lower() for ext in ['.pdf', '.jpg', '.png', '.gif', '.zip'])):
subpages.add(full_url)
if len(subpages) >= max_subpages:
break
return list(subpages)[:max_subpages]
except Exception:
return []
def validate_url(self, url: str) -> bool:
"""Validate if URL is accessible"""
try:
response = requests.head(url, headers=self.headers, timeout=5)
return response.status_code == 200
except:
return False
def get_website_info(self, url: str) -> Dict[str, Any]:
"""Get basic information about a website"""
try:
response = requests.get(url, headers=self.headers, timeout=self.timeout)
response.raise_for_status()
soup = BeautifulSoup(response.content, 'html.parser')
return {
'url': url,
'title': self._extract_title(soup),
'description': self._extract_description(soup),
'meta_keywords': self._extract_meta_keywords(soup),
'has_robots_meta': bool(soup.find('meta', attrs={'name': 'robots'})),
'has_viewport_meta': bool(soup.find('meta', attrs={'name': 'viewport'})),
'language': soup.get('lang', 'unknown'),
'status_code': response.status_code,
'content_type': response.headers.get('content-type', 'unknown'),
'server': response.headers.get('server', 'unknown')
}
except Exception as e:
return {'url': url, 'error': f"Could not get website info: {str(e)}"}
class ParserFactory:
"""Factory class to create appropriate parsers"""
@staticmethod
def get_parser(source_type: str):
"""Get the appropriate parser for the source type"""
parsers = {
'pdf': PDFParser(),
'text': TextParser(),
'webpage': WebpageParser(),
'url': WebpageParser()
}
return parsers.get(source_type.lower())
@staticmethod
def detect_source_type(source: str) -> str:
"""Detect the type of content source"""
if source.startswith(('http://', 'https://')):
return 'webpage'
elif source.endswith('.pdf'):
return 'pdf'
else:
return 'text'