#!/usr/bin/env python3 """Generate ~10M+ seed URLs from multiple sources for the web crawler.""" import random, os, sys, hashlib OUTPUT = "/workspace/mega_seeds_v2.txt" urls = set() def log(msg): print(f"[seed-gen] {msg}", flush=True) # ========== 1. Top domains with deep path generation ========== TOP_DOMAINS = [ # News & Media (60+) "bbc.com", "reuters.com", "theguardian.com", "nytimes.com", "washingtonpost.com", "cnn.com", "aljazeera.com", "apnews.com", "npr.org", "pbs.org", "theatlantic.com", "newyorker.com", "economist.com", "ft.com", "bloomberg.com", "wired.com", "arstechnica.com", "theverge.com", "techcrunch.com", "engadget.com", "vice.com", "slate.com", "salon.com", "vox.com", "politico.com", "bbc.co.uk", "independent.co.uk", "telegraph.co.uk", "sky.com", "abc.net.au", "smh.com.au", "stuff.co.nz", "cbc.ca", "globeandmail.com", "spiegel.de", "zeit.de", "lemonde.fr", "elpais.com", "corriere.it", "straitstimes.com", "scmp.com", "japantimes.co.jp", "hindustantimes.com", "timesofindia.indiatimes.com", "ndtv.com", "thehindu.com", "dawn.com", "bangkokpost.com", "koreaherald.com", "dw.com", "france24.com", "euronews.com", "rfi.fr", "channelnewsasia.com", "rt.com", "tass.com", "latimes.com", "chicagotribune.com", "bostonglobe.com", "sfgate.com", "seattletimes.com", "denverpost.com", "dallasnews.com", "miamiherald.com", "usatoday.com", "foxnews.com", "nbcnews.com", "cbsnews.com", "abcnews.go.com", "newsweek.com", "time.com", "usnews.com", "thehill.com", "axios.com", "propublica.org", "theintercept.com", "motherjones.com", "thedailybeast.com", "rawstory.com", "huffpost.com", "buzzfeednews.com", # Science & Education (40+) "nature.com", "sciencedirect.com", "plos.org", "arxiv.org", "ncbi.nlm.nih.gov", "pubmed.ncbi.nlm.nih.gov", "nih.gov", "nasa.gov", "esa.int", "cern.ch", "britannica.com", "newscientist.com", "scientificamerican.com", "smithsonianmag.com", "nationalgeographic.com", "khanacademy.org", "coursera.org", "edx.org", "sciencemag.org", "cell.com", "thelancet.com", "bmj.com", "physicstoday.org", "aps.org", "acs.org", "rsc.org", "ieee.org", "acm.org", "springer.com", "wiley.com", "elsevier.com", "jstor.org", "researchgate.net", "academia.edu", "livescience.com", "sciencenews.org", "phys.org", "sciencedaily.com", "popularmechanics.com", "popularsciencemag.com", "discovermagazine.com", # Tech & Dev (40+) "github.com", "stackoverflow.com", "dev.to", "medium.com", "hackernoon.com", "lobste.rs", "slashdot.org", "docs.python.org", "developer.mozilla.org", "w3schools.com", "geeksforgeeks.org", "tutorialspoint.com", "realpython.com", "freecodecamp.org", "digitalocean.com", "linux.die.net", "man7.org", "kernel.org", "infoq.com", "dzone.com", "baeldung.com", "vogella.com", "tldp.org", "linuxjournal.com", "lwn.net", "css-tricks.com", "smashingmagazine.com", "alistapart.com", "martinfowler.com", "joelonsoftware.com", "paulgraham.com", "blog.codinghorror.com", "scottberkun.com", "towardsdatascience.com", "kdnuggets.com", "analyticsvidhya.com", "machinelearningmastery.com", "distill.pub", "openai.com", "huggingface.co", "paperswithcode.com", # Reference & Knowledge (20+) "en.wikipedia.org", "en.wiktionary.org", "en.wikisource.org", "en.wikiquote.org", "en.wikibooks.org", "commons.wikimedia.org", "simple.wikipedia.org", "archive.org", "gutenberg.org", "openlibrary.org", "plato.stanford.edu", "iep.utm.edu", "snopes.com", "factcheck.org", "politifact.com", "dictionary.com", "merriam-webster.com", "etymonline.com", "howstuffworks.com", "explainthatstuff.com", # Government (40+) "cdc.gov", "nih.gov", "fda.gov", "epa.gov", "nasa.gov", "noaa.gov", "usgs.gov", "energy.gov", "education.gov", "treasury.gov", "state.gov", "nist.gov", "nsf.gov", "doi.gov", "usda.gov", "loc.gov", "archives.gov", "bls.gov", "census.gov", "weather.gov", "nps.gov", "fcc.gov", "ftc.gov", "sec.gov", "gov.uk", "nhs.uk", "parliament.uk", "ons.gov.uk", "canada.ca", "abs.gov.au", "stats.govt.nz", "europa.eu", "who.int", "un.org", "worldbank.org", "imf.org", "oecd.org", "wto.org", "iaea.org", "unesco.org", # Culture & Arts (20+) "imdb.com", "rottentomatoes.com", "metacritic.com", "allmusic.com", "discogs.com", "genius.com", "goodreads.com", "librarything.com", "moma.org", "metmuseum.org", "britishmuseum.org", "artnet.com", "artsy.net", "christies.com", "sothebys.com", # Health (15+) "webmd.com", "mayoclinic.org", "healthline.com", "medlineplus.gov", "drugs.com", "medscape.com", "clevelandclinic.org", "hopkinsmedicine.org", "mountsinai.org", "patient.info", "nhs.uk", # Business & Finance (15+) "investopedia.com", "fool.com", "seekingalpha.com", "marketwatch.com", "cnbc.com", "wsj.com", "forbes.com", "businessinsider.com", "inc.com", "fastcompany.com", "hbr.org", "mckinsey.com", "bcg.com", # Food & Lifestyle (10+) "allrecipes.com", "foodnetwork.com", "epicurious.com", "seriouseats.com", "bonappetit.com", "cookinglight.com", "food52.com", "tasteatlas.com", # Travel & Geography (10+) "lonelyplanet.com", "tripadvisor.com", "atlasobscura.com", "worldatlas.com", "roughguides.com", ] # 200 topic words for path generation WORDS = [ "climate-change", "artificial-intelligence", "quantum-computing", "renewable-energy", "machine-learning", "biodiversity", "cybersecurity", "blockchain", "nanotechnology", "gene-therapy", "space-exploration", "ocean-conservation", "nuclear-fusion", "autonomous-vehicles", "brain-computer-interface", "crispr", "dark-matter", "electric-vehicles", "food-security", "global-health", "hydrogen-energy", "internet-of-things", "lithium-batteries", "mars-colonization", "neural-networks", "organic-chemistry", "particle-physics", "quantum-entanglement", "robotics", "semiconductor", "vaccine-development", "water-purification", "ancient-civilizations", "medieval-history", "industrial-revolution", "world-war-two", "cold-war", "roman-empire", "byzantine-empire", "ottoman-empire", "mongol-empire", "silk-road", "renaissance", "enlightenment", "french-revolution", "american-revolution", "civil-rights-movement", "decolonization", "globalization", "democracy", "economics", "philosophy", "psychology", "sociology", "anthropology", "archaeology", "linguistics", "mathematics", "statistics", "calculus", "algebra", "geometry", "topology", "biology", "chemistry", "physics", "astronomy", "geology", "ecology", "genetics", "evolution", "immunology", "virology", "neuroscience", "pharmacology", "epidemiology", "nutrition", "architecture", "engineering", "materials-science", "aeronautics", "computer-science", "data-science", "cryptography", "algorithms", "operating-systems", "databases", "networking", "compilers", "distributed-systems", "cloud-computing", "microservices", "python-programming", "javascript", "rust-language", "golang", "music-theory", "film-history", "literature", "poetry", "painting", "sculpture", "photography", "theater", "dance", "cooking-techniques", "fermentation", "agriculture", "forestry", "marine-biology", "volcanology", "seismology", "meteorology", "climate-science", "paleontology", "entomology", "botany", "zoology", "mycology", "microbiology", "biochemistry", "organic-synthesis", "polymer-science", "thermodynamics", "fluid-dynamics", "optics", "acoustics", "electromagnetism", "superconductivity", "plasma-physics", "astrophysics", "cosmology", "general-relativity", "string-theory", "cognitive-science", "behavioral-economics", "game-theory", "information-theory", "chaos-theory", "complexity-theory", "number-theory", "graph-theory", "combinatorics", "differential-equations", "linear-algebra", "probability", "machine-vision", "natural-language-processing", "reinforcement-learning", "transformer-models", "generative-ai", "computer-graphics", "virtual-reality", "augmented-reality", "human-computer-interaction", "software-engineering", "devops", "agile-methodology", "design-patterns", "functional-programming", "deep-learning", "attention-mechanisms", "diffusion-models", "genetic-algorithms", "cellular-automata", "swarm-intelligence", "sustainability", "urban-planning", "public-health", "mental-health", "education-reform", "immigration", "trade-policy", "fiscal-policy", "monetary-policy", "supply-chain", "logistics", "manufacturing", "automation", "3d-printing", "biotechnology", "synthetic-biology", "bioinformatics", "proteomics", "genomics", "transcriptomics", "metabolomics", "spectroscopy", "chromatography", "crystallography", "microscopy", "signal-processing", "control-theory", "optimization", "stochastic-processes", "bayesian-inference", "regression-analysis", "time-series", "dimensionality-reduction", "clustering", "anomaly-detection", "recommender-systems", "knowledge-graphs", "federated-learning", "transfer-learning", "few-shot-learning", "self-supervised-learning", "contrastive-learning", "meta-learning", "multi-agent-systems", "evolutionary-computation", "fuzzy-logic", "formal-verification", "type-theory", "category-theory", "abstract-algebra", "real-analysis", "complex-analysis", "measure-theory", "functional-analysis", "harmonic-analysis", ] PATH_PATTERNS = [ "/wiki/{word}", "/article/{word}", "/news/{word}", "/topic/{word}", "/category/{word}", "/tag/{word}", "/blog/{word}", "/posts/{word}", "/science/{word}", "/technology/{word}", "/health/{word}", "/politics/{word}", "/business/{word}", "/culture/{word}", "/world/{word}", "/opinion/{word}", "/analysis/{word}", "/features/{word}", "/reviews/{word}", "/guides/{word}", "/how-to/{word}", "/tutorial/{word}", "/learn/{word}", "/research/{word}", "/papers/{word}", "/docs/{word}", "/history/{word}", "/environment/{word}", "/education/{word}", "/en/{word}", "/en/article/{word}", "/en/news/{word}", "/topics/{word}", "/sections/{word}", "/archives/{word}", "/explore/{word}", "/discover/{word}", "/search?q={word}", "/{word}", "/stories/{word}", "/insights/{word}", ] log("Phase 1: Domain + path URLs...") for domain in TOP_DOMAINS: urls.add(f"https://{domain}/") urls.add(f"https://www.{domain}/") for pattern in PATH_PATTERNS: for word in WORDS: url = f"https://{domain}{pattern}".replace("{word}", word) urls.add(url) # Also with www url2 = f"https://www.{domain}{pattern}".replace("{word}", word) urls.add(url2) # Date-based archives for year in range(2020, 2027): for month in range(1, 13): urls.add(f"https://{domain}/{year}/{month:02d}/") urls.add(f"https://www.{domain}/{year}/{month:02d}/") for word in random.sample(WORDS, 5): urls.add(f"https://{domain}/{year}/{month:02d}/{word}") log(f" Domain+path: {len(urls):,} URLs") # ========== 2. Wikipedia — massive generation ========== log("Phase 2: Wikipedia URLs...") # Prefix + subject combinations WIKI_PREFIXES = [ "History_of_", "List_of_", "Geography_of_", "Economy_of_", "Culture_of_", "Demographics_of_", "Politics_of_", "Education_in_", "Religion_in_", "Music_of_", "Sport_in_", "Climate_of_", "Cuisine_of_", "Transport_in_", "Health_in_", "Science_and_technology_in_", "Tourism_in_", "Military_of_", "Architecture_of_", "Literature_of_", "Timeline_of_", "Outline_of_", "Index_of_", "Flag_of_", "Coat_of_arms_of_", "Capital_of_", "Languages_of_", "Ethnic_groups_in_", "Biodiversity_of_", "National_symbols_of_", "Human_rights_in_", "Taxation_in_", "Energy_policy_of_", "Foreign_relations_of_", "Law_of_", "Media_of_", "Cinema_of_", "Theatre_in_", "Dance_in_", ] # Countries, cities, regions PLACES = [ "the_United_States", "the_United_Kingdom", "France", "Germany", "Japan", "China", "India", "Brazil", "Russia", "Canada", "Australia", "Mexico", "Italy", "Spain", "South_Korea", "Netherlands", "Sweden", "Norway", "Switzerland", "Poland", "Turkey", "Egypt", "South_Africa", "Nigeria", "Kenya", "Argentina", "Chile", "Colombia", "Peru", "Indonesia", "Thailand", "Vietnam", "Malaysia", "Philippines", "Pakistan", "Bangladesh", "Iran", "Iraq", "Saudi_Arabia", "Israel", "New_Zealand", "Ireland", "Scotland", "Wales", "Portugal", "Greece", "Czech_Republic", "Hungary", "Romania", "Bulgaria", "Croatia", "Serbia", "Ukraine", "Belarus", "Lithuania", "Latvia", "Estonia", "Finland", "Denmark", "Iceland", "Luxembourg", "Belgium", "Austria", "Slovakia", "Slovenia", "North_Macedonia", "Albania", "Montenegro", "Bosnia_and_Herzegovina", "Moldova", "Georgia_(country)", "Armenia", "Azerbaijan", "Kazakhstan", "Uzbekistan", "Turkmenistan", "Kyrgyzstan", "Tajikistan", "Mongolia", "Nepal", "Sri_Lanka", "Myanmar", "Cambodia", "Laos", "Brunei", "East_Timor", "Papua_New_Guinea", "Fiji", "Samoa", "Tonga", "Morocco", "Algeria", "Tunisia", "Libya", "Sudan", "Ethiopia", "Somalia", "Tanzania", "Uganda", "Rwanda", "Burundi", "Democratic_Republic_of_the_Congo", "Republic_of_the_Congo", "Cameroon", "Ghana", "Senegal", "Mali", "Niger", "Burkina_Faso", "Ivory_Coast", "Guinea", "Sierra_Leone", "Liberia", "Togo", "Benin", "Mauritania", "Madagascar", "Mozambique", "Zimbabwe", "Zambia", "Malawi", "Botswana", "Namibia", "Angola", "Gabon", "Equatorial_Guinea", "Cuba", "Haiti", "Dominican_Republic", "Jamaica", "Trinidad_and_Tobago", "Panama", "Costa_Rica", "Honduras", "Guatemala", "El_Salvador", "Nicaragua", "Belize", "Paraguay", "Uruguay", "Ecuador", "Bolivia", "Venezuela", "Guyana", "Suriname", # Cities "New_York_City", "London", "Paris", "Tokyo", "Berlin", "Rome", "Moscow", "Beijing", "Mumbai", "Sydney", "Toronto", "Dubai", "Singapore", "Hong_Kong", "Seoul", "Istanbul", "Cairo", "Los_Angeles", "Chicago", "Houston", "Phoenix", "Philadelphia", "San_Antonio", "San_Diego", "Dallas", "San_Jose", "Austin", "San_Francisco", "Seattle", "Denver", "Boston", "Nashville", "Portland", "Las_Vegas", "Miami", "Atlanta", "Minneapolis", "Detroit", "Pittsburgh", "Cleveland", "Cincinnati", "Milwaukee", "Bangkok", "Jakarta", "Manila", "Ho_Chi_Minh_City", "Kuala_Lumpur", "Shanghai", "Shenzhen", "Guangzhou", "Chengdu", "Wuhan", "Osaka", "Kyoto", "Nagoya", "Yokohama", "Sapporo", "Delhi", "Kolkata", "Chennai", "Bangalore", "Hyderabad", "Karachi", "Lahore", "Dhaka", "Colombo", "Kathmandu", "Lagos", "Nairobi", "Addis_Ababa", "Dar_es_Salaam", "Johannesburg", "Cape_Town", "Casablanca", "Algiers", "Tunis", "Accra", "Buenos_Aires", "Sao_Paulo", "Rio_de_Janeiro", "Lima", "Bogota", "Santiago", "Quito", "Caracas", "Montevideo", "La_Paz", "Mexico_City", "Havana", "Kingston", "Panama_City", "San_Juan", "Dublin", "Edinburgh", "Manchester", "Birmingham", "Glasgow", "Amsterdam", "Rotterdam", "Brussels", "Antwerp", "Zurich", "Geneva", "Vienna", "Prague", "Budapest", "Warsaw", "Krakow", "Bucharest", "Sofia", "Athens", "Lisbon", "Porto", "Barcelona", "Madrid", "Valencia", "Seville", "Milan", "Naples", "Florence", "Venice", "Turin", "Munich", "Hamburg", "Frankfurt", "Cologne", "Stuttgart", "Copenhagen", "Stockholm", "Oslo", "Helsinki", "Reykjavik", "St._Petersburg", "Kyiv", "Minsk", "Tbilisi", "Baku", ] for prefix in WIKI_PREFIXES: for place in PLACES: urls.add(f"https://en.wikipedia.org/wiki/{prefix}{place}") # Standalone articles — big list WIKI_ARTICLES = [ "Photosynthesis", "DNA", "RNA", "Protein", "Enzyme", "Cell_(biology)", "Mitochondrion", "Chloroplast", "Ribosome", "Chromosome", "Gene", "Evolution", "Natural_selection", "Speciation", "Taxonomy_(biology)", "Ecosystem", "Biome", "Food_web", "Biodiversity", "Conservation_biology", "Plate_tectonics", "Earthquake", "Volcano", "Tsunami", "Continental_drift", "Electron", "Proton", "Neutron", "Quark", "Photon", "Neutrino", "Higgs_boson", "Standard_Model", "Quantum_mechanics", "General_relativity", "Electromagnetism", "Thermodynamics", "Entropy", "Wave", "Frequency", "Periodic_table", "Chemical_element", "Chemical_compound", "Chemical_reaction", "Algorithm", "Data_structure", "Turing_machine", "P_versus_NP_problem", "Artificial_intelligence", "Machine_learning", "Neural_network", "Operating_system", "Computer_network", "Database", "Compiler", "Internet", "World_Wide_Web", "HTTP", "Encryption", "Blockchain", "Solar_System", "Sun", "Mercury_(planet)", "Venus", "Earth", "Mars", "Jupiter", "Saturn", "Uranus", "Neptune", "Moon", "Galaxy", "Milky_Way", "Black_hole", "Neutron_star", "Supernova", "Dark_matter", "Dark_energy", "Big_Bang", "Cosmic_microwave_background", "Calculus", "Linear_algebra", "Number_theory", "Set_theory", "Group_theory", "Topology", "Pi", "Prime_number", "Fibonacci_number", "Python_(programming_language)", "JavaScript", "Java_(programming_language)", "C_(programming_language)", "Rust_(programming_language)", "Go_(programming_language)", "Linux", "Unix", "Automobile", "Airplane", "Steam_engine", "Electric_motor", "Battery_(electricity)", "Solar_cell", "Wind_turbine", "Nuclear_power", "Semiconductor", "Transistor", "Integrated_circuit", "Printing_press", "Telescope", "Microscope", "Compass", "Clock", "Alexander_the_Great", "Julius_Caesar", "Genghis_Khan", "Napoleon", "Leonardo_da_Vinci", "Isaac_Newton", "Albert_Einstein", "Charles_Darwin", "Marie_Curie", "Nikola_Tesla", "Alan_Turing", "William_Shakespeare", "Democracy", "Capitalism", "Socialism", "United_Nations", "Olympic_Games", "FIFA_World_Cup", "Association_football", "Human_brain", "Heart", "Immune_system", "Vaccine", "Antibiotic", "Cancer", "Diabetes", "Malaria", "HIV/AIDS", "Agriculture", "Rice", "Wheat", "Coffee", "Tea", "Steel", "Iron", "Gold", "Concrete", "Plastic", "Bridge", "Dam", "Pyramid", "Great_Wall_of_China", "Oxygen", "Hydrogen", "Carbon", "Nitrogen", "Helium", "Lithium", "Sodium", "Potassium", "Calcium", "Magnesium", "Phosphorus", "Sulfur", "Chlorine", "Argon", "Silicon", "Aluminum", "Copper", "Zinc", "Tin", "Lead", "Mercury_(element)", "Uranium", "Plutonium", "Water", "Ammonia", "Methane", "Ethanol", "Glucose", "Cellulose", "Starch", "Lignin", "Keratin", "Collagen", "Hemoglobin", "Insulin", "Dopamine", "Serotonin", "Adrenaline", "Cortisol", "Testosterone", "Estrogen", "Melatonin", "Oxytocin", "Endorphin", "Bacteria", "Virus", "Fungus", "Archaea", "Protozoa", "Antibiotic_resistance", "Prion", "Bacteriophage", "Plasmid", "CRISPR", "Polymerase_chain_reaction", "Gel_electrophoresis", "Mass_spectrometry", "X-ray_crystallography", "Nuclear_magnetic_resonance", "Electron_microscope", "Scanning_tunneling_microscope", "Hubble_Space_Telescope", "James_Webb_Space_Telescope", "Large_Hadron_Collider", "ITER", "International_Space_Station", "Apollo_program", "Space_Shuttle", "SpaceX", "Mars_rover", "Voyager_program", "Cassini-Huygens", "New_Horizons", "Global_Positioning_System", "Satellite", "Fiber_optics", "5G", "Wi-Fi", "Bluetooth", "RFID", "Barcode", "Laser", "LED", "OLED", "Liquid_crystal_display", "Cathode-ray_tube", "Vacuum_tube", "Diode", "Capacitor", "Resistor", "Inductor", "Transformer", "Relay", "Electric_generator", "Electric_power_transmission", "Alternating_current", "Direct_current", "Superconductor", "Magnet", "Electromagnet", "Magnetic_resonance_imaging", "Computed_tomography", "Ultrasound", "Positron_emission_tomography", "Radioactivity", "Nuclear_fission", "Nuclear_fusion", "Isotope", "Carbon_dating", "Geiger_counter", "Dosimeter", "Renewable_energy", "Fossil_fuel", "Coal", "Petroleum", "Natural_gas", "Geothermal_energy", "Tidal_power", "Wave_power", "Biomass", "Photovoltaics", "Concentrated_solar_power", "Wind_farm", "Hydroelectricity", "Pumped-storage_hydroelectricity", "Climate_change", "Greenhouse_gas", "Carbon_dioxide", "Methane", "Ozone_layer", "Acid_rain", "Air_pollution", "Water_pollution", "Deforestation", "Desertification", "Coral_bleaching", "Mass_extinction", "Endangered_species", "Conservation", "National_park", "World_Heritage_Site", "Biosphere_reserve", "Amazon_rainforest", "Great_Barrier_Reef", "Sahara", "Arctic", "Antarctic", "Mariana_Trench", "Mount_Everest", "Grand_Canyon", "Niagara_Falls", "Victoria_Falls", "Nile", "Amazon_River", "Mississippi_River", "Yangtze", "Ganges", "Danube", "Rhine", "Thames", "Seine", "Pacific_Ocean", "Atlantic_Ocean", "Indian_Ocean", "Arctic_Ocean", "Mediterranean_Sea", "Caribbean_Sea", "South_China_Sea", "Pangaea", "Gondwana", "Laurasia", "Rodinia", "Cambrian_explosion", "Permian-Triassic_extinction_event", "Cretaceous-Paleogene_extinction_event", "Dinosaur", "Tyrannosaurus", "Triceratops", "Velociraptor", "Brachiosaurus", "Mammoth", "Saber-toothed_cat", "Megalodon", "Homo_sapiens", "Homo_erectus", "Neanderthal", "Australopithecus", "Stone_Age", "Bronze_Age", "Iron_Age", "Mesopotamia", "Ancient_Egypt", "Indus_Valley_Civilisation", "Ancient_Greece", "Ancient_Rome", "Han_dynasty", "Tang_dynasty", "Ming_dynasty", "Qing_dynasty", "Mughal_Empire", "Gupta_Empire", "Maurya_Empire", "Achaemenid_Empire", "Sassanid_Empire", "Maya_civilization", "Aztec_Empire", "Inca_Empire", "Viking_Age", "Crusades", "Black_Death", "Hundred_Years_War", "Age_of_Discovery", "Columbian_exchange", "Atlantic_slave_trade", "Scientific_Revolution", "Protestant_Reformation", "Thirty_Years_War", "Seven_Years_War", "Napoleonic_Wars", "Congress_of_Vienna", "Scramble_for_Africa", "Meiji_Restoration", "World_War_I", "Russian_Revolution", "Great_Depression", "World_War_II", "Holocaust", "Cold_War", "Korean_War", "Vietnam_War", "Space_Race", "Cuban_Missile_Crisis", "Berlin_Wall", "Dissolution_of_the_Soviet_Union", "September_11_attacks", "War_on_terror", "Iraq_War", "Arab_Spring", "COVID-19_pandemic", "Plato", "Aristotle", "Socrates", "Confucius", "Laozi", "Buddha", "Muhammad", "Jesus", "Moses", "Abraham", "Immanuel_Kant", "Friedrich_Nietzsche", "Karl_Marx", "John_Locke", "Thomas_Hobbes", "Jean-Jacques_Rousseau", "Voltaire", "David_Hume", "Rene_Descartes", "Baruch_Spinoza", "Georg_Wilhelm_Friedrich_Hegel", "Arthur_Schopenhauer", "Soren_Kierkegaard", "Ludwig_Wittgenstein", "Bertrand_Russell", "Noam_Chomsky", "Michel_Foucault", "Jacques_Derrida", "Simone_de_Beauvoir", "Hannah_Arendt", "John_Rawls", "Adam_Smith", "John_Maynard_Keynes", "Milton_Friedman", "Friedrich_Hayek", "Joseph_Schumpeter", "Amartya_Sen", "Galileo_Galilei", "Johannes_Kepler", "Nicolaus_Copernicus", "Tycho_Brahe", "Robert_Hooke", "Gottfried_Wilhelm_Leibniz", "Leonhard_Euler", "Carl_Friedrich_Gauss", "Bernhard_Riemann", "Henri_Poincare", "David_Hilbert", "Emmy_Noether", "Srinivasa_Ramanujan", "Kurt_Godel", "John_von_Neumann", "Claude_Shannon", "Norbert_Wiener", "Richard_Feynman", "Niels_Bohr", "Werner_Heisenberg", "Erwin_Schrodinger", "Paul_Dirac", "Max_Planck", "James_Clerk_Maxwell", "Michael_Faraday", "Andre-Marie_Ampere", "Georg_Ohm", "Heinrich_Hertz", "Guglielmo_Marconi", "Alexander_Graham_Bell", "Thomas_Edison", "Nikolai_Lobachevsky", "George_Boole", "Charles_Babbage", "Ada_Lovelace", "Grace_Hopper", "Tim_Berners-Lee", "Vint_Cerf", "Dennis_Ritchie", "Linus_Torvalds", "Steve_Jobs", "Bill_Gates", "Louis_Pasteur", "Robert_Koch", "Alexander_Fleming", "Jonas_Salk", "Francis_Crick", "James_Watson", "Rosalind_Franklin", "Barbara_McClintock", "Rachel_Carson", "Jane_Goodall", "David_Attenborough", ] for article in WIKI_ARTICLES: urls.add(f"https://en.wikipedia.org/wiki/{article}") urls.add(f"https://en.wikipedia.org/wiki/Talk:{article}") # Years for year in range(1, 2027): urls.add(f"https://en.wikipedia.org/wiki/{year}") for year in range(1800, 2027): for suffix in ["_in_science", "_in_literature", "_in_music", "_in_art", "_in_film", "_in_television", "_in_sports", "_in_aviation", "_in_spaceflight", "_in_archaeology"]: urls.add(f"https://en.wikipedia.org/wiki/{year}{suffix}") # curid-based random articles — 500K of them for _ in range(500000): curid = random.randint(1, 75000000) urls.add(f"https://en.wikipedia.org/w/index.php?curid={curid}") # Special:Random with cache busters for i in range(100000): urls.add(f"https://en.wikipedia.org/wiki/Special:Random?x={i}") # Other language wikis for lang in ["de", "fr", "es", "it", "pt", "ja", "zh", "ru", "ko", "ar", "nl", "sv", "pl", "vi", "id", "uk", "cs", "fi", "hu", "ro", "da", "no", "he", "th", "el", "bg", "hr", "sr", "sk", "sl", "et", "lt", "lv", "ca", "eu", "gl", "ms", "tl", "hi", "bn", "ta", "te", "mr", "ur", "fa", "tr"]: for _ in range(5000): curid = random.randint(1, 5000000) urls.add(f"https://{lang}.wikipedia.org/w/index.php?curid={curid}") log(f" Wikipedia: {len(urls):,} URLs") # ========== 3. StackOverflow deep links ========== log("Phase 3: StackOverflow...") SO_TAGS = [ "python", "javascript", "java", "c%23", "php", "android", "html", "css", "node.js", "sql", "mysql", "r", "reactjs", "c%2b%2b", "angular", "typescript", "linux", "git", "docker", "kubernetes", "rust", "go", "swift", "kotlin", "machine-learning", "deep-learning", "tensorflow", "pytorch", "numpy", "pandas", "django", "flask", "spring", "express", "vue.js", "svelte", "next.js", "aws", "azure", "google-cloud", "terraform", "ansible", "postgresql", "mongodb", "redis", "elasticsearch", "apache-kafka", "algorithms", "data-structures", "regex", "bash", "powershell", "networking", "security", "cryptography", "api", "rest", "graphql", "testing", "unit-testing", "selenium", "ci-cd", "jenkins", ] for tag in SO_TAGS: for page in range(1, 201): urls.add(f"https://stackoverflow.com/questions/tagged/{tag}?tab=votes&page={page}") urls.add(f"https://stackoverflow.com/questions/tagged/{tag}?tab=newest&page={page}") # Random SO question IDs (questions go up to ~80M) for _ in range(200000): qid = random.randint(1, 80000000) urls.add(f"https://stackoverflow.com/questions/{qid}") log(f" StackOverflow: {len(urls):,} URLs") # ========== 4. GitHub trending & topics ========== log("Phase 4: GitHub...") GH_TOPICS = [ "python", "javascript", "typescript", "rust", "go", "java", "cpp", "machine-learning", "deep-learning", "artificial-intelligence", "web", "api", "cli", "database", "security", "devops", "linux", "react", "vue", "angular", "svelte", "nextjs", "django", "flask", "docker", "kubernetes", "terraform", "ansible", "algorithms", "data-structures", "compilers", "operating-systems", ] for topic in GH_TOPICS: urls.add(f"https://github.com/topics/{topic}") for page in range(1, 21): urls.add(f"https://github.com/topics/{topic}?page={page}") # Popular GitHub repos README pages GH_REPOS = [ "torvalds/linux", "python/cpython", "rust-lang/rust", "golang/go", "tensorflow/tensorflow", "pytorch/pytorch", "microsoft/vscode", "facebook/react", "vuejs/vue", "angular/angular", "nodejs/node", "django/django", "pallets/flask", "kubernetes/kubernetes", "docker/compose", "apache/spark", "elastic/elasticsearch", "redis/redis", "postgres/postgres", "openai/gpt-2", "huggingface/transformers", "scikit-learn/scikit-learn", "pandas-dev/pandas", "numpy/numpy", ] for repo in GH_REPOS: urls.add(f"https://github.com/{repo}") urls.add(f"https://github.com/{repo}/wiki") urls.add(f"https://github.com/{repo}/issues") urls.add(f"https://github.com/{repo}/pulls") log(f" GitHub: {len(urls):,} URLs") # ========== 5. Reddit deep links ========== log("Phase 5: Reddit...") SUBREDDITS = [ "science", "technology", "worldnews", "askscience", "explainlikeimfive", "todayilearned", "history", "space", "physics", "chemistry", "biology", "math", "programming", "machinelearning", "datascience", "philosophy", "books", "movies", "music", "food", "travel", "fitness", "medicine", "engineering", "economics", "psychology", "linguistics", "anthropology", "archaeology", "geology", "astronomy", "environment", "energy", "climate", "oceanography", "neuroscience", "genetics", "ecology", "evolution", "botanicalporn", "mycology", "entomology", "askhistorians", "askphilosophy", "askengineers", "askeconomics", "learnprogramming", "compsci", "netsec", "reverseengineering", "artificial", "deeplearning", "statistics", "dataisbeautiful", "mapporn", "historyporn", "earthporn", "spaceporn", "cityporn", "architectureporn", "educationalgifs", "interestingasfuck", "damnthatsinteresting", "coolguides", "futurology", "collapse", "geopolitics", "neutralpolitics", "truereddit", "depthub", "bestof", "changemyview", "unpopularopinion", "documentaries", "lectures", "mealtimevideos", "python", "javascript", "rust", "golang", "java", "cpp", "linux", "sysadmin", "devops", "homelab", "selfhosted", "mechanicalkeyboards", "buildapc", "hardware", "overclocking", "cooking", "recipes", "seriouseats", "breadit", "fermentation", "gardening", "permaculture", "composting", "photography", "videography", "filmmakers", "writing", "screenwriting", "worldbuilding", "chess", "boardgames", "rpg", "dnd", "running", "cycling", "swimming", "climbing", "hiking", "camping", "backpacking", "ultralight", ] for sub in SUBREDDITS: urls.add(f"https://www.reddit.com/r/{sub}/") urls.add(f"https://old.reddit.com/r/{sub}/") for sort in ["top", "hot", "new", "controversial"]: urls.add(f"https://www.reddit.com/r/{sub}/{sort}/") urls.add(f"https://old.reddit.com/r/{sub}/{sort}/") for time in ["all", "year", "month", "week"]: urls.add(f"https://www.reddit.com/r/{sub}/top/?t={time}") log(f" Reddit: {len(urls):,} URLs") # ========== 6. News archive deep date links ========== log("Phase 6: News archives...") NEWS_DOMAINS = [ "bbc.com/news", "reuters.com", "apnews.com", "theguardian.com", "nytimes.com", "washingtonpost.com", "cnn.com", "aljazeera.com", "npr.org", "theatlantic.com", "wired.com", "arstechnica.com", "latimes.com", "chicagotribune.com", "usatoday.com", ] NEWS_SECTIONS = [ "world", "us", "uk", "business", "technology", "science", "health", "entertainment", "sports", "politics", "environment", "education", "opinion", "analysis", "features", "investigations", "asia", "europe", "africa", "americas", "middle-east", ] for site in NEWS_DOMAINS: for section in NEWS_SECTIONS: for year in range(2015, 2027): for month in range(1, 13): urls.add(f"https://www.{site}/{section}/{year}/{month:02d}") for day in [1, 5, 10, 15, 20, 25]: urls.add(f"https://www.{site}/{section}/{year}/{month:02d}/{day:02d}") log(f" News: {len(urls):,} URLs") # ========== 7. Academic & research sites ========== log("Phase 7: Academic URLs...") UNIVERSITIES = [ "mit.edu", "stanford.edu", "harvard.edu", "berkeley.edu", "caltech.edu", "princeton.edu", "yale.edu", "columbia.edu", "cornell.edu", "upenn.edu", "uchicago.edu", "duke.edu", "northwestern.edu", "jhu.edu", "cmu.edu", "gatech.edu", "umich.edu", "uw.edu", "ucla.edu", "ucsd.edu", "uiuc.edu", "purdue.edu", "wisc.edu", "utexas.edu", "psu.edu", "cam.ac.uk", "ox.ac.uk", "imperial.ac.uk", "ucl.ac.uk", "ed.ac.uk", "ethz.ch", "epfl.ch", "tum.de", "lmu.de", "anu.edu.au", "unimelb.edu.au", "utoronto.ca", "ubc.ca", "mcgill.ca", ] UNI_PATHS = [ "research", "news", "about", "academics", "departments", "faculty", "library", "publications", "programs", "courses", "events", "computer-science", "physics", "mathematics", "chemistry", "biology", "engineering", "medicine", "law", "business", "economics", "history", "philosophy", "psychology", "linguistics", "sociology", "electrical-engineering", "mechanical-engineering", "civil-engineering", "environmental-science", "materials-science", "biomedical-engineering", "neuroscience", "cognitive-science", "political-science", "public-health", "epidemiology", "statistics", ] for uni in UNIVERSITIES: for path in UNI_PATHS: urls.add(f"https://www.{uni}/{path}/") urls.add(f"https://{uni}/{path}/") # arXiv papers (IDs go up to ~2503.xxxxx) for year in range(14, 26): for month in range(1, 13): for paper in range(1, 501): urls.add(f"https://arxiv.org/abs/{year:02d}{month:02d}.{paper:05d}") # PubMed articles for _ in range(200000): pmid = random.randint(1, 40000000) urls.add(f"https://pubmed.ncbi.nlm.nih.gov/{pmid}/") log(f" Academic: {len(urls):,} URLs") # ========== 8. Misc high-quality sites ========== log("Phase 8: Misc sites...") # Medium articles by topic MEDIUM_TAGS = [ "artificial-intelligence", "machine-learning", "data-science", "programming", "python", "javascript", "technology", "science", "startup", "design", "productivity", "self-improvement", "writing", "marketing", "business", "blockchain", "cryptocurrency", "cybersecurity", "ux-design", "web-development", "deep-learning", "nlp", "computer-vision", "robotics", "iot", "cloud-computing", "devops", "microservices", "api", "databases", "algorithms", "software-engineering", "system-design", "architecture", "leadership", "management", "culture", "education", "health", "mental-health", "neuroscience", "psychology", "philosophy", "history", "economics", "politics", "climate-change", "sustainability", "energy", ] for tag in MEDIUM_TAGS: urls.add(f"https://medium.com/tag/{tag}") urls.add(f"https://medium.com/tag/{tag}/latest") urls.add(f"https://medium.com/tag/{tag}/top/all-time") for year in range(2020, 2027): for month in range(1, 13): urls.add(f"https://medium.com/tag/{tag}/archive/{year}/{month:02d}") # HackerNews items (IDs go up to ~40M+) for _ in range(200000): item_id = random.randint(1, 42000000) urls.add(f"https://news.ycombinator.com/item?id={item_id}") # IMDb titles for _ in range(100000): tid = random.randint(1, 30000000) urls.add(f"https://www.imdb.com/title/tt{tid:07d}/") # Goodreads books for _ in range(50000): bid = random.randint(1, 70000000) urls.add(f"https://www.goodreads.com/book/show/{bid}") log(f" Misc: {len(urls):,} URLs") # ========== 9. Documentation sites ========== log("Phase 9: Documentation sites...") PYTHON_MODULES = [ "os", "sys", "json", "re", "math", "datetime", "collections", "itertools", "functools", "pathlib", "asyncio", "typing", "logging", "unittest", "argparse", "subprocess", "threading", "multiprocessing", "socket", "http", "urllib", "xml", "csv", "sqlite3", "hashlib", "random", "statistics", "decimal", "array", "queue", "heapq", "bisect", "struct", "io", "time", "string", "textwrap", "difflib", "enum", "dataclasses", "abc", "contextlib", "warnings", "copy", "pprint", "reprlib", "operator", "inspect", "importlib", "pkgutil", "zipimport", "compileall", "dis", "ast", "symtable", "token", "tokenize", "pickle", "shelve", "marshal", "dbm", "gzip", "bz2", "lzma", "zipfile", "tarfile", "tempfile", "glob", "fnmatch", "shutil", "filecmp", "stat", "fileinput", "signal", "mmap", "ctypes", "select", "ssl", "ftplib", "smtplib", "imaplib", "cgi", "html", "webbrowser", ] for mod in PYTHON_MODULES: urls.add(f"https://docs.python.org/3/library/{mod}.html") MDN_PATHS = [ "HTML", "CSS", "JavaScript", "API", "HTTP", "SVG", "MathML", "Web/HTML/Element", "Web/CSS/Reference", "Web/JavaScript/Reference", "Web/API/Document", "Web/API/Window", "Web/API/Element", "Web/API/Fetch_API", "Web/API/Canvas_API", "Web/API/WebSocket", "Web/API/Web_Workers_API", "Web/API/Service_Worker_API", "Web/API/IndexedDB_API", "Web/API/Geolocation_API", "Web/API/Web_Audio_API", "Web/API/WebGL_API", "Web/API/WebRTC_API", "Web/HTTP/Headers", "Web/HTTP/Methods", "Web/HTTP/Status", ] for path in MDN_PATHS: urls.add(f"https://developer.mozilla.org/en-US/docs/{path}") log(f" Docs: {len(urls):,} URLs") # ========== 10. Generate numeric URL patterns for high-throughput sites ========== log("Phase 10: Numeric pattern generation for remaining URLs...") # Britannica articles for _ in range(100000): word = random.choice(WORDS + [a.replace("_", "-").lower() for a in WIKI_ARTICLES[:200]]) urls.add(f"https://www.britannica.com/topic/{word}") urls.add(f"https://www.britannica.com/science/{word}") urls.add(f"https://www.britannica.com/technology/{word}") urls.add(f"https://www.britannica.com/place/{word}") urls.add(f"https://www.britannica.com/biography/{word}") urls.add(f"https://www.britannica.com/event/{word}") # Additional random Wikipedia curids to hit 10M remaining = 10000000 - len(urls) if remaining > 0: log(f" Generating {remaining:,} more Wikipedia curid URLs to reach 10M...") batch = set() while len(batch) < remaining: curid = random.randint(1, 75000000) batch.add(f"https://en.wikipedia.org/w/index.php?curid={curid}") if len(batch) % 500000 == 0: log(f" {len(batch):,} / {remaining:,}") urls.update(batch) log(f" Final total: {len(urls):,} URLs") # ========== WRITE OUTPUT ========== url_list = list(urls) random.shuffle(url_list) log(f"Writing {len(url_list):,} URLs to {OUTPUT}...") with open(OUTPUT, "w") as f: for url in url_list: f.write(url + "\n") sz = os.path.getsize(OUTPUT) log(f"Done! {len(url_list):,} unique URLs written ({sz/(1024*1024):.1f}MB)")