Spaces:
Sleeping
Sleeping
File size: 13,966 Bytes
519c06d |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 |
"""
ArXiv Fetcher Component
Fetches and processes research papers from ArXiv
"""
import re
import time
import requests
from typing import List, Dict, Optional, Any
from datetime import datetime, timedelta
import arxiv
class ArxivFetcher:
"""
Fetches research papers from ArXiv
Provides search, download, and metadata extraction capabilities
"""
def __init__(self, config = None):
# Import Config only when needed to avoid dependency issues
if config is None:
try:
from .config import Config
self.config = Config()
except ImportError:
# Fallback to None if Config cannot be imported
self.config = None
else:
self.config = config
self.client = arxiv.Client()
def search_papers(self,
query: str,
max_results: int = 10,
sort_by: str = "relevance",
category: str = None,
date_range: int = None) -> List[Dict[str, Any]]:
"""
Search for papers on ArXiv
Args:
query: Search query
max_results: Maximum number of results
sort_by: Sort criteria ('relevance', 'lastUpdatedDate', 'submittedDate')
category: ArXiv category filter (e.g., 'cs.AI', 'cs.LG')
date_range: Days back to search (e.g., 7, 30, 365)
Returns:
List of paper dictionaries
"""
try:
print(f"Searching ArXiv for: '{query}'")
# Build search query
search_query = query
if category:
search_query = f"cat:{category} AND {query}"
# Set sort criteria
sort_criteria = {
"relevance": arxiv.SortCriterion.Relevance,
"lastUpdatedDate": arxiv.SortCriterion.LastUpdatedDate,
"submittedDate": arxiv.SortCriterion.SubmittedDate
}.get(sort_by, arxiv.SortCriterion.Relevance)
# Create search
search = arxiv.Search(
query=search_query,
max_results=max_results,
sort_by=sort_criteria,
sort_order=arxiv.SortOrder.Descending
)
papers = []
for result in self.client.results(search):
# Date filtering
if date_range:
cutoff_date = datetime.now() - timedelta(days=date_range)
if result.published.replace(tzinfo=None) < cutoff_date:
continue
# Extract paper information
paper = self._extract_paper_info(result)
papers.append(paper)
print(f"Found {len(papers)} papers")
return papers
except Exception as e:
print(f"Error searching ArXiv: {e}")
return []
def get_paper_by_id(self, arxiv_id: str) -> Optional[Dict[str, Any]]:
"""
Get a specific paper by ArXiv ID
Args:
arxiv_id: ArXiv paper ID (e.g., '2301.12345')
Returns:
Paper dictionary or None
"""
try:
print(f"Fetching paper: {arxiv_id}")
search = arxiv.Search(id_list=[arxiv_id])
results = list(self.client.results(search))
if results:
paper = self._extract_paper_info(results[0])
print(f"Retrieved paper: {paper['title']}")
return paper
else:
print(f"Paper not found: {arxiv_id}")
return None
except Exception as e:
print(f"Error fetching paper {arxiv_id}: {e}")
return None
def search_by_author(self, author: str, max_results: int = 20) -> List[Dict[str, Any]]:
"""
Search for papers by author
Args:
author: Author name
max_results: Maximum number of results
Returns:
List of paper dictionaries
"""
query = f"au:{author}"
return self.search_papers(query, max_results=max_results, sort_by="lastUpdatedDate")
def search_by_category(self, category: str, max_results: int = 20) -> List[Dict[str, Any]]:
"""
Search for papers by category
Args:
category: ArXiv category (e.g., 'cs.AI', 'cs.LG', 'stat.ML')
max_results: Maximum number of results
Returns:
List of paper dictionaries
"""
query = f"cat:{category}"
return self.search_papers(query, max_results=max_results, sort_by="lastUpdatedDate")
def get_trending_papers(self, category: str = "cs.AI", days: int = 7, max_results: int = 10) -> List[Dict[str, Any]]:
"""
Get trending papers in a category
Args:
category: ArXiv category
days: Days back to look for papers
max_results: Maximum number of results
Returns:
List of paper dictionaries
"""
return self.search_by_category(category, max_results=max_results)
def _extract_paper_info(self, result) -> Dict[str, Any]:
"""
Extract paper information from ArXiv result
Args:
result: ArXiv search result
Returns:
Paper dictionary
"""
try:
# Extract ArXiv ID
arxiv_id = result.entry_id.split('/')[-1]
# Clean and format data
paper = {
'arxiv_id': arxiv_id,
'title': result.title.strip(),
'authors': [author.name for author in result.authors],
'summary': result.summary.strip(),
'published': result.published.isoformat(),
'updated': result.updated.isoformat(),
'categories': result.categories,
'primary_category': result.primary_category,
'pdf_url': result.pdf_url,
'entry_id': result.entry_id,
'journal_ref': result.journal_ref,
'doi': result.doi,
'comment': result.comment,
'links': [{'title': link.title, 'href': link.href} for link in result.links],
'fetched_at': datetime.now().isoformat()
}
# Add formatted metadata
paper['authors_str'] = ', '.join(paper['authors'][:3]) + ('...' if len(paper['authors']) > 3 else '')
paper['categories_str'] = ', '.join(paper['categories'][:3]) + ('...' if len(paper['categories']) > 3 else '')
paper['year'] = result.published.year
paper['month'] = result.published.month
return paper
except Exception as e:
print(f"Error extracting paper info: {e}")
return {
'arxiv_id': 'unknown',
'title': 'Error extracting title',
'authors': [],
'summary': 'Error extracting summary',
'error': str(e)
}
def download_pdf(self, paper: Dict[str, Any], download_dir: str = "downloads") -> Optional[str]:
"""
Download PDF for a paper
Args:
paper: Paper dictionary
download_dir: Directory to save PDF
Returns:
Path to downloaded PDF or None
"""
try:
import os
os.makedirs(download_dir, exist_ok=True)
pdf_url = paper.get('pdf_url')
if not pdf_url:
print(f"No PDF URL for paper: {paper.get('title', 'Unknown')}")
return None
arxiv_id = paper.get('arxiv_id', 'unknown')
filename = f"{arxiv_id}.pdf"
filepath = os.path.join(download_dir, filename)
if os.path.exists(filepath):
print(f"PDF already exists: {filepath}")
return filepath
print(f"Downloading PDF: {paper.get('title', 'Unknown')}")
response = requests.get(pdf_url, timeout=30)
response.raise_for_status()
with open(filepath, 'wb') as f:
f.write(response.content)
print(f"PDF downloaded: {filepath}")
return filepath
except Exception as e:
print(f"Error downloading PDF: {e}")
return None
def get_paper_recommendations(self, paper_id: str, max_results: int = 5) -> List[Dict[str, Any]]:
"""
Get paper recommendations based on a paper's content
Args:
paper_id: ArXiv ID of the base paper
max_results: Number of recommendations
Returns:
List of recommended papers
"""
try:
# Get the base paper
base_paper = self.get_paper_by_id(paper_id)
if not base_paper:
return []
# Extract key terms from title and summary
title = base_paper.get('title', '')
summary = base_paper.get('summary', '')
categories = base_paper.get('categories', [])
# Simple keyword extraction (can be improved with NLP)
keywords = self._extract_keywords(title + ' ' + summary)
# Search for related papers
query = ' '.join(keywords[:5]) # Use top 5 keywords
related_papers = self.search_papers(
query=query,
max_results=max_results + 5, # Get more to filter out the original
sort_by="relevance"
)
# Filter out the original paper
recommendations = [p for p in related_papers if p.get('arxiv_id') != paper_id]
return recommendations[:max_results]
except Exception as e:
print(f"Error getting recommendations: {e}")
return []
def _extract_keywords(self, text: str) -> List[str]:
"""
Simple keyword extraction from text
Args:
text: Input text
Returns:
List of keywords
"""
# Simple implementation - can be improved with NLP libraries
import re
from collections import Counter
# Remove common stop words
stop_words = {'the', 'and', 'or', 'but', 'in', 'on', 'at', 'to', 'for', 'of', 'with', 'by', 'a', 'an', 'as', 'is', 'was', 'are', 'were', 'be', 'been', 'have', 'has', 'had', 'do', 'does', 'did', 'will', 'would', 'could', 'should', 'may', 'might', 'must', 'can', 'this', 'that', 'these', 'those', 'we', 'us', 'our', 'you', 'your', 'he', 'him', 'his', 'she', 'her', 'it', 'its', 'they', 'them', 'their'}
# Extract words
words = re.findall(r'\b[a-zA-Z]{3,}\b', text.lower())
# Filter and count
filtered_words = [word for word in words if word not in stop_words]
word_counts = Counter(filtered_words)
# Return most common words
return [word for word, count in word_counts.most_common(20)]
def get_categories(self) -> Dict[str, str]:
"""
Get available ArXiv categories
Returns:
Dictionary of category codes and descriptions
"""
return {
'cs.AI': 'Artificial Intelligence',
'cs.LG': 'Machine Learning',
'cs.CV': 'Computer Vision',
'cs.CL': 'Computation and Language',
'cs.NE': 'Neural and Evolutionary Computing',
'cs.RO': 'Robotics',
'cs.CR': 'Cryptography and Security',
'cs.DC': 'Distributed, Parallel, and Cluster Computing',
'cs.DB': 'Databases',
'cs.DS': 'Data Structures and Algorithms',
'cs.HC': 'Human-Computer Interaction',
'cs.IR': 'Information Retrieval',
'cs.IT': 'Information Theory',
'cs.MM': 'Multimedia',
'cs.NI': 'Networking and Internet Architecture',
'cs.OS': 'Operating Systems',
'cs.PL': 'Programming Languages',
'cs.SE': 'Software Engineering',
'cs.SY': 'Systems and Control',
'stat.ML': 'Machine Learning (Statistics)',
'stat.AP': 'Applications (Statistics)',
'stat.CO': 'Computation (Statistics)',
'stat.ME': 'Methodology (Statistics)',
'stat.TH': 'Statistics Theory',
'math.ST': 'Statistics Theory (Mathematics)',
'math.PR': 'Probability (Mathematics)',
'math.OC': 'Optimization and Control',
'math.NA': 'Numerical Analysis',
'eess.AS': 'Audio and Speech Processing',
'eess.IV': 'Image and Video Processing',
'eess.SP': 'Signal Processing',
'eess.SY': 'Systems and Control',
'q-bio.QM': 'Quantitative Methods',
'q-bio.NC': 'Neurons and Cognition',
'physics.data-an': 'Data Analysis, Statistics and Probability'
}
|