ferret 0.9.6 → 0.10.0

data/test/unit/index/tc_index_writer.rb

@@ -7,6 +7,8 @@ class IndexWriterTest < Test::Unit::TestCase
 
   def setup()
     @dir = Ferret::Store::RAMDirectory.new
+    fis = FieldInfos.new()
+    fis.create_index(@dir)
   end
 
   def tear_down()
@@ -18,7 +20,7 @@ class IndexWriterTest < Test::Unit::TestCase
     clock = @dir.make_lock(IndexWriter::COMMIT_LOCK_NAME)
     assert(! wlock.locked?)
     assert(! clock.locked?)
-    iw = IndexWriter.new(@dir, :create => true)
+    iw = IndexWriter.new(:dir => @dir, :create => true)
     assert(@dir.exists?("segments"))
     assert(wlock.locked?)
     iw.close()
@@ -28,26 +30,408 @@ class IndexWriterTest < Test::Unit::TestCase
   end
 
   def test_add_document
-    iw = IndexWriter.new(@dir, :analyzer => StandardAnalyzer.new(), :create => true)
-    doc = IndexTestHelper.prepare_document()
-    iw.add_document(doc)
+    iw = IndexWriter.new(:dir => @dir,
+                         :analyzer => StandardAnalyzer.new(),
+                         :create => true)
+    iw << {:title => "first doc", :content => ["contents of", "first doc"]}
     assert_equal(1, iw.doc_count)
+    iw << ["contents of", "second doc"]
+    assert_equal(2, iw.doc_count)
+    iw << "contents of third doc"
+    assert_equal(3, iw.doc_count)
     iw.close()
   end
 
-  def test_add_documents
-    iw = IndexWriter.new(@dir, :analyzer => StandardAnalyzer.new(), :create => true)
-    # uncomment the following line to see logging
-    #iw.info_stream = $stdout
+  def test_add_documents_fuzzy
+    iw = IndexWriter.new(:dir => @dir,
+                         :analyzer => StandardAnalyzer.new())
     iw.merge_factor = 3
-    iw.min_merge_docs = 3
-    docs = IndexTestHelper.prepare_book_list()
-    docs.each_with_index do |doc, i|
-      #puts "Index doc " + i.to_s
+    iw.max_buffered_docs = 3
+
+    # add 100 documents
+    100.times do
+      doc = random_doc()
       iw.add_document(doc)
     end
-    assert_equal(37, iw.doc_count)
+    assert_equal(100, iw.doc_count)
     iw.close()
   end
 
+  private
+
+  WORDS = [
+    "desirous", "hollowness's", "camp's", "Senegal", "broadcaster's",
+    "pecking", "Provence", "paternalism", "premonition", "Dumbo's",
+    "Darlene's", "Elbert's", "substrate", "Camille", "Menkalinan", "Cooper",
+    "decamps", "abatement's", "bindings", "scrubby", "subset", "ancestor's",
+    "pelagic", "abscissa", "loofah's", "gleans", "boudoir", "disappointingly",
+    "guardianship's", "settlers", "Mylar", "timetable's", "parabolic",
+    "madams", "bootlegger's", "monotonically", "gage", "Karyn's", "deposed",
+    "boozy", "swordfish's", "Chevron", "Victrola", "Tameka", "impels",
+    "carrels", "salami's", "celibate", "resistance's", "duration",
+    "abscissae", "Kilroy's", "corrosive", "flight's", "flapper", "scare",
+    "peppiest", "Pygmies", "Menzies", "wrist's", "enumerable", "housecoats",
+    "Khwarizmi's", "stampeding", "hungering", "steeping", "Yemenis",
+    "entangles", "solver", "mishapping", "Rand's", "ninety", "Boris",
+    "impedimenta", "predators", "ridge", "wretchedness's", "crapping", "Head",
+    "Edwards", "Claude's", "geodesics", "verities", "botch", "Short's",
+    "vellum's", "coruscates", "hydrogenates", "Haas's", "deceitfulness",
+    "cohort's", "Cepheus", "totes", "Cortez's", "napalm", "fruitcake",
+    "coordinated", "Coulomb", "desperation", "behoves", "contractor's",
+    "vacationed", "Wanamaker's", "leotard", "filtrated", "cringes", "Lugosi",
+    "sheath's", "orb", "jawed", "Isidro", "geophysics", "persons", "Asians",
+    "booze's", "eight's", "backslappers", "hankered", "dos", "helpings",
+    "tough", "interlarding", "gouger", "inflect", "Juneau's", "hay's",
+    "sardining", "spays", "Brandi", "depressant", "space", "assess",
+    "reappearance's", "Eli's", "Cote", "Enoch", "chants", "ruffianing",
+    "moralised", "unsuccessfully", "or", "Maryland's", "mildest", "unsafer",
+    "dutiful", "Pribilof", "teas", "vagued", "microbiologists", "hedgerow",
+    "speller's", "conservators", "catharsis", "drawbacks", "whooshed",
+    "unlawful", "revolve", "craftsmanship", "destabilise", "Margarito",
+    "Asgard's", "spawn's", "Annabel's", "canonicals", "buttermilk",
+    "exaltation's", "pothole", "reprints", "approximately", "homage",
+    "Wassermann's", "Atlantic's", "exacerbated", "Huerta", "keypunching",
+    "engagements", "dilate", "ponchos", "Helvetius", "Krakatoa", "basket's",
+    "stepmother", "schlock's", "drippings", "cardiology's", "northwesterly",
+    "cruddier", "poesies", "rustproof", "climb", "miscalled", "Belgians",
+    "Iago", "brownout", "nurseries", "hooliganism's", "concourse's",
+    "advocate", "sunrise's", "hyper", "octopus's", "erecting",
+    "counterattacking", "redesign", "studies", "nitrating", "milestone",
+    "bawls", "Nereid", "inferring", "Ontario's", "annexed", "treasury",
+    "cosmogony's", "scandalised", "shindig's", "detention's",
+    "Lollobrigida's", "eradicating", "magpie", "supertankers", "Adventist's",
+    "dozes", "Artaxerxes", "accumulate", "dankest", "telephony", "flows",
+    "Srivijaya's", "fourteen's", "antonym", "rancid", "briefing's",
+    "theologian", "Jacuzzi", "gracing", "chameleon's", "Brittney's",
+    "Pullmans", "Robitussin's", "jitterier", "mayonnaise's", "fort",
+    "closeouts", "amatory", "Drew's", "cockfight", "pyre", "Laura's",
+    "Bradley's", "obstructionists", "interventions", "tenderness's",
+    "loadstones", "castigation's", "undercut", "volubly", "meditated",
+    "Ypsilanti", "Jannie's", "tams", "drummer's", "inaugurations", "mawing",
+    "Anglophile", "Sherpa", "footholds", "Gonzalo", "removers",
+    "customisation", "procurement's", "allured", "grimaced", "captaining",
+    "liberates", "grandeur's", "Windsor", "screwdrivers", "Flynn's",
+    "extortionists", "carnivorous", "thinned", "panhandlers", "trust's",
+    "bemoaned", "untwisted", "cantors", "rectifies", "speculation",
+    "niacin's", "soppy", "condom", "halberd", "Leadbelly", "vocation's",
+    "tanners", "chanticleer", "secretariats", "Ecuador's", "suppurated",
+    "users", "slag's", "atrocity's", "pillar", "sleeveless", "bulldozers",
+    "turners", "hemline", "astounded", "rosaries", "Mallarmé", "crucifies",
+    "Maidenform", "contribution", "evolve", "chemicals", "uteri",
+    "expostulation", "roamers", "daiquiris", "arraignment", "ribs", "King's",
+    "Persepolis", "arsenic's", "blindfolds", "bloodsucker's", "restocks",
+    "falconry", "Olympia's", "Colosseum's", "vigils", "Louie's",
+    "unwillingly", "sealed", "potatoes", "Argentine", "audit's", "outworn",
+    "boggles", "likely", "alleging", "Tinkerbell", "redistribution's",
+    "Normandy", "Cortes", "porter's", "buntings", "cornucopias", "rosewoods",
+    "shelf's", "airdrops", "summits", "Rosalyn", "redecorating", "twirlers",
+    "monsters", "directed", "semiautomatics", "Foch", "Hobart", "mutilates",
+    "Wilma's", "ornamenting", "Clifford's", "pyromania", "Strasbourg",
+    "bleeders", "additions", "super", "effortlessly", "piecing", "vacations",
+    "gybes", "warranted", "Ting", "her", "histrionic", "marshaled", "spore's",
+    "villainy's", "brat", "confusion", "amphitheatre's", "adjourns",
+    "guzzled", "Visayans", "rogue's", "morsels", "candlestick", "flaks",
+    "Waterbury", "pulp's", "endorser's", "postdoc", "coffining", "swallowing",
+    "Wrangell", "Marcie's", "Marley", "untapped", "fear's", "Kant",
+    "pursuit's", "normally", "jackals", "orals", "Paramaribo's", "Marilyn's",
+    "Diem's", "narrower", "medicinally", "chickweed's", "pretentiousness",
+    "Lardner", "baritone's", "purrs", "Pam's", "pestles", "Philip's",
+    "Titania", "eccentrics", "Albion's", "greed's", "raggediest",
+    "importations", "Truman", "incentives", "typified", "incurred",
+    "bandstands", "Minnie's", "pleasant", "Sandy's", "perplexities",
+    "crease's", "obliques", "backstop", "Nair's", "perusing", "Quixote's",
+    "sicknesses", "vapour's", "butte", "lariats", "disfavours", "McGuffey",
+    "paediatric", "filtered", "whiff's", "gunboats", "devolved",
+    "extravaganza's", "organism", "giggling", "citadel's", "counterbalances",
+    "executrixes", "Cathay", "marshmallow's", "iniquitous", "Katmai", "Siva",
+    "welled", "impertinence's", "plunger", "rice", "forgers", "Larousse",
+    "pollution's", "medium", "residue's", "rumbas", "Odis", "arrogant",
+    "Jasper's", "panged", "doubted", "vistaing", "decibel's", "modulus's",
+    "chickpea's", "mugger's", "potentates", "sequesters", "academy's",
+    "Turk's", "pharmacology's", "defogger", "clomp", "soulless", "elastic",
+    "la's", "shards", "unfortunate", "counterclaim's", "objections", "towel",
+    "converged", "z", "ionisation", "stirrups", "antiquarians", "constructor",
+    "virtuosity's", "Göteborg", "centigramme's", "translators", "dalliance's",
+    "us", "bullfight", "drawer's", "nonconformist", "handcrafts", "Magritte",
+    "tulle", "plant's", "routine", "colour's", "latency's", "repertoire's",
+    "photocopies", "catalyse", "ashrams", "lagging", "flapjack's",
+    "ayatollahs", "decentest", "pitted", "conformity", "jack", "batsman",
+    "electrifies", "Unitarians", "obtain", "medicates", "tumour's",
+    "nutritionally", "haystack", "bustles", "slut", "satirising", "birettas",
+    "starring", "Kubrick's", "flogs", "chequering", "Menkalinan's",
+    "Barbados's", "Bioko", "swinish", "hades", "perjured", "timing's",
+    "cocaine", "ejecting", "rationalises", "dilettante's", "umping",
+    "capsized", "frogmen", "matt", "prostituting", "bola's", "devolution's",
+    "poxing", "Maritza's", "snob's", "scoped", "Costco", "feral", "sirocco",
+    "rebating", "truculence", "junkier", "nabs", "elicit", "allegiance",
+    "care", "arteriosclerosis's", "nonproliferation's", "doxologies",
+    "disconsolate", "bodega", "designers", "Rembrandt", "apostasies",
+    "garrulousness", "Hertzsprung's", "hayseeds", "noncooperation's",
+    "resentment", "cuticles", "sandboxes", "gimmicks", "magnolia",
+    "invalidity's", "pulverised", "Tinkerbell's", "hypoglycemics",
+    "gunboat's", "workbench's", "fleetingly's", "sportsman's", "trots",
+    "decomposes", "discrepancies", "owls", "obscener", "organic", "stoutness",
+    "councillor's", "Philippine's", "Aline", "coarsening", "suffocated",
+    "infighting's", "peculiarity", "roof's", "premier", "sucked", "churl",
+    "remounts", "intends", "wiles", "unfold", "unperturbed", "wainscotings",
+    "restfuller", "ashtray's", "wader's", "decanters", "gild", "tandems",
+    "spooked", "galling", "annuity's", "opacity", "clamour's", "flaccid",
+    "caroming", "savvying", "mammalian's", "toadstool's", "doohickey", "jibs",
+    "conquests", "dishes", "effusively", "distinctions", "curly", "Peckinpah",
+    "whining", "quasar", "sponge", "infrequent", "Novembers", "cowling",
+    "poem's", "muzzles", "Sufi", "authoritarians", "prompts", "Gavin's",
+    "morphology's", "shenanigan", "narrated", "rapprochement", "Heine",
+    "propane's", "addition", "prefect's", "pining", "dwindles",
+    "compulsiveness's", "objectors", "trudging", "segregates", "language",
+    "enthralled", "explosiveness", "toeing", "drainers", "Merrimack's",
+    "smarten", "bigwig's", "embroiders", "Medicaids", "grammar's", "behest's",
+    "chiseled", "equalled", "factual", "Casablanca's", "dams",
+    "disillusioned", "turtleneck", "Baden", "provinces", "bushwhacked", "fey",
+    "Yangtze", "loan's", "decent", "strobe", "challenger's", "hometown",
+    "Neal", "Ernestine's", "magnetises", "minute", "patrol", "Starbucks",
+    "Bernstein", "signal", "interplanetary", "tweak", "archdeacon",
+    "untoward", "transducer", "azaleas", "levied", "worlds", "talks",
+    "Tancred", "hairsplitting's", "edibility's", "confab", "rosetted",
+    "Spanish", "Americanisation", "Charley", "realm's", "incongruities",
+    "chinstraps", "dollhouses", "binocular", "popgun", "physiotherapy's",
+    "knave's", "angelically", "heartbreaking", "clarions", "bespeaks",
+    "pivotal", "Zosma", "ungrammatical", "dilution", "tidily", "Dejesus's",
+    "taller", "pennyweight's", "freshman", "Jamestown", "chiefer", "amen",
+    "attiring", "appurtenance's", "opiates", "mottoes", "towellings", "ashen",
+    "font's", "spoors", "pupil", "groom's", "skimpy", "achieves",
+    "intolerance's", "ardour's", "exorcist", "bottoming", "snag's",
+    "Frenches", "hysteric's", "ladyfinger's", "differences", "seed",
+    "clubfoot's", "glades", "Elton's", "jargon", "Waldo", "grinning",
+    "coherence's", "winos", "turnround", "appended", "Ethelred's", "delete",
+    "steadfastness's", "miss", "thermoplastic", "depraves", "unctuous",
+    "reanimates", "transfusing", "protects", "Babbage's", "foists", "inn",
+    "etched", "sanctimoniously", "idling", "timepiece", "holistic",
+    "waterside", "ulna's", "swindled", "employables", "zebra", "nieces",
+    "pertained", "usages", "vamp's", "Larry's", "cooler's", "holographs",
+    "clewing", "stubborning", "peaked", "underfeeds", "marshmallows",
+    "agreeable", "beards", "Slovenia's", "nitroglycerin", "palls", "impurer",
+    "armours", "stomachaches", "notification's", "Dixieland's", "crozier's",
+    "neurotic", "kudos", "Tania's", "M", "soundtrack's", "territory's",
+    "sped", "house's", "divisibility", "ingress's", "pummelled", "Isabel",
+    "Dewitt", "seemly", "hutched", "calliope", "lengthwise", "flubs",
+    "Moldavia's", "Mercia", "McBride's", "Lenten", "pulverise", "football",
+    "oligarchy", "Max", "scribbler", "acclimatize", "brainwashes",
+    "apprenticed", "benevolences", "two", "Wodehouse", "crew's", "massacre",
+    "proportionals", "Jewishness's", "instep's", "emissary", "folder",
+    "nonentity's", "convinced", "caption", "kangarooed", "dogie",
+    "vagabonding", "auction's", "appraising", "antimony", "part's",
+    "longitude's", "inconsiderateness's", "pawning", "serer", "solos",
+    "histories", "mushy", "parturition", "munched", "oregano", "inanest",
+    "dryness", "kitchenware", "unexpected", "covens", "cheesecakes",
+    "stakeout's", "Pulaski's", "Yoknapatawpha's", "pinhead", "drifted",
+    "guzzler's", "funking", "sou'wester", "oesophagus's", "highbrow",
+    "contralto", "meningitis", "Mazzini", "raggedest", "vaginas", "misfiring",
+    "margaritas", "wedder", "pointed", "slicked", "garlanded", "comeuppances",
+    "vassals", "Sui", "Concord", "bozos", "Garry's", "Maribel's", "epileptic",
+    "Jehoshaphat's", "revolutionary's", "kneecaps", "songbird", "actively",
+    "Meredith", "toddler", "distrusting", "fuchsias", "perusal", "instills",
+    "deathbed", "sunspot's", "spatula's", "Muscovy", "humaniser", "Keats",
+    "regrets", "deflect", "theories", "nonpluses", "populating", "leniency's",
+    "penicillin's", "gaol's", "borough", "moose's", "dogmata",
+    "transcendentally", "supposition's", "nursed", "Gagarin's", "honest",
+    "Chandrasekhar's", "mudslinger's", "parable", "bonged", "Wyeth's",
+    "Ochoa's", "Grenoble", "steamy", "halter's", "rotisserie's", "pagoda's",
+    "wallaby's", "Yank", "pretzel", "rapist's", "estrange", "hectored",
+    "Puebla's", "conniver", "creditor's", "dole's", "Fotomat", "patents",
+    "heckling", "thickener", "etches", "yogi", "hemstitched", "obverses",
+    "Lipizzaner", "divert", "Strong's", "sagest", "Alabama", "He", "Carrie's",
+    "obligation's", "verity's", "outed", "Rhee", "bluffed", "codas",
+    "crèche's", "unpalatable", "dilettanti", "vestment", "purse's",
+    "inflammation's", "bookmarked", "doing's", "whinnying", "impersonators",
+    "Theiler", "scurried", "resistor", "southerners", "Anacreon",
+    "reconstruction's", "footage", "trespassing", "Kafka", "bottling",
+    "stays", "Gretzky", "overburdening", "princesses", "weathercock's",
+    "atolls", "cheerier", "packet", "surrenders", "teacup", "Sabik's",
+    "undecidable", "lollygagged", "pawl's", "anaesthesiology", "sublimely",
+    "contortionists", "motorcades", "Maureen", "lamasery", "yourselves",
+    "Creighton", "poliomyelitis's", "civil", "outmanoeuvre", "lauded",
+    "closeness", "Humboldt's", "pretzels", "ungrudging", "blackguard's",
+    "sickles", "typo", "narcotics", "linesman", "psychotics", "pictured",
+    "deviltry", "Yahtzee", "Lovelace's", "cerebra", "airiness's", "bewitch",
+    "how", "motherland's", "crate's", "Keenan's", "turnstile's",
+    "pedometer's", "carted", "slipping", "fallow", "Canadian", "ladybird's",
+    "thump", "shopper's", "enters", "scowls", "nematode", "focused",
+    "Riley's", "grainiest", "novas", "snuffled", "leftovers", "deify",
+    "Samoan", "pruning", "contenting", "Khachaturian's", "triads",
+    "genealogies", "psalmist", "shaming", "appropriated", "ignominies",
+    "Beadle's", "MHz", "peerages", "facile", "Seoul", "Janna's", "jig's",
+    "mousiness's", "funnier", "delimiter", "watermark", "sheik's", "Reasoner",
+    "ipecac's", "curdles", "wronged", "Segovia's", "solders", "Dunne's",
+    "contractor", "awards", "hostels", "pinkie's", "Herzl", "misplace",
+    "shuttle", "innovative", "vestries", "cosmoses", "trikes", "Casandra's",
+    "hokier", "carouser's", "summerhouses", "renascence", "decomposed",
+    "Balzac's", "outlast", "shod", "squalling", "smugging", "weighing",
+    "omega's", "selects", "fleetingly", "Finland", "petted", "disrespects",
+    "fetter", "confound", "brads", "Bosnia's", "preposition's", "guy's",
+    "different", "tracts", "paediatrics's", "polygon", "eyetooth's", "Aesop",
+    "pentagons", "professions", "homeowner", "looter's", "intimidated",
+    "lustre's", "loneliness", "catnapped", "counties", "pailful",
+    "Christendom's", "Barents", "penis", "Mumford's", "Nigel", "éclairs",
+    "splats", "diabolical", "popularly", "quart", "abjected", "Rasalgethi",
+    "camel's", "inimical", "overweening", "distention's", "Advil", "casement",
+    "seamier", "avaricious", "sierra's", "caparison's", "moldered", "Cortez",
+    "handmaid's", "disappointment", "billowed", "overpopulated", "outsets",
+    "ray", "smoother", "overkill", "somber", "tiller's", "zigzag", "adviser",
+    "absorption's", "sturdily", "hairy", "bloodmobile", "investiture's",
+    "creature", "ripeness's", "Jonathon", "arborvitae's", "skulduggery",
+    "bog", "skeleton's", "Kit's", "Panamas", "Ashlee's", "jazzy", "snit",
+    "divisive", "caribous", "permuting", "frankest", "annotated", "oak's",
+    "meg's", "Gill", "burrito", "dormancy's", "offings", "Nike",
+    "outnumbered", "skater's", "Portugal", "deficit", "Cannon's", "pockmark",
+    "sediment's", "mailbox", "innuendoed", "retire", "wolfhound's",
+    "nicotine's", "brigade's", "mettle's", "softhearted", "hooey's",
+    "abdication", "Orval", "Jaime", "ship", "hyphenations", "sectarians",
+    "Alabaman", "tagging", "ultras", "schizoids", "medicines", "undersized",
+    "Gray", "maternity's", "bandaging", "scooping", "coercion's", "serapes",
+    "celebrate", "Listerine's", "throve", "crypt's", "nearsighted",
+    "metallurgists", "Delicious", "cotton's", "yoked", "cogitates",
+    "underage", "cigarette's", "hallways", "Cointreau", "ma'am", "spacing's",
+    "foresight", "parkway's", "Edwardian", "mediator", "Turner", "Derrida's",
+    "motorist's", "hobo", "equivalences", "sophism", "peeping", "telescoped",
+    "overproduce", "ductility", "Leblanc", "refractory", "passé", "decodes",
+    "womanising", "flax's", "pond's", "infrequency", "talkativeness's",
+    "settlement's", "Prince", "bating", "multimillionaire", "Schultz",
+    "premiss", "quackery", "bathhouse", "Leno's", "Monday's", "Hung's",
+    "undaunted", "bewaring", "tension's", "Chile's", "Rostand's", "platoons",
+    "rodeo's", "Dionne", "Dyson's", "gingivitis's", "fewer",
+    "electromagnetism's", "scrubbier", "ensconced", "wretcheder", "mica's",
+    "expectorant", "snapper's", "chastised", "habitation", "spry", "bathing",
+    "stealth's", "champagnes", "baleful", "fencing's", "threaded", "codicils",
+    "disgraced", "redcaps", "addends", "Olivier", "clasped", "Gwendolyn",
+    "foment", "angularity's", "strenuously", "gorilla", "misbehaved",
+    "surplus's", "newsier", "positioned", "bloodmobiles", "circumstantials",
+    "person's", "varicose", "Calliope", "plethora", "Olmsted",
+    "reconciliation", "Brendan's", "beset", "totters", "sailors",
+    "parliamentarians", "Whitaker", "hilts", "pummelling", "academician's",
+    "ruse", "discreeter", "appetisingly", "perfections", "anus", "overrode",
+    "pedantry's", "possessed", "germs", "unscrews", "expired",
+    "semitrailer's", "Cupid's", "nonsmoker", "Marathon", "secs", "Hopkins",
+    "freeing", "libelled", "furious", "staccatos", "electroencephalogram's",
+    "malingerer's", "impulses", "briars", "Tran", "hilltops", "sulks",
+    "quailed", "fads", "retrenches", "spouted", "outtake", "puncture's",
+    "rats", "kibitzed", "berets", "omnivorous", "flange", "Mons", "glints",
+    "mansards", "thou", "cuing", "suspected", "Kaiser's", "savvier", "skits",
+    "interdict's", "Booker", "Rubinstein", "Tm's", "crossing's", "dewlap",
+    "guarantor's", "edification's", "joyfullest", "crossed", "chowdering",
+    "sillier", "reloading", "commodity's", "bodkins", "conduced", "coughs",
+    "nucleus's", "sixtieth", "proverbially", "comprehensive", "ineluctably",
+    "patrolmen", "resuscitating", "carpetbag's", "Darrin's", "Yeager",
+    "Bataan's", "spoonsful", "proceeds", "wrongdoer", "Karroo", "heart",
+    "poison", "typifying", "endowment's", "aquanauts", "deaconesses",
+    "homosexuality", "Maxine", "haunching", "centred", "Peking's",
+    "toothiest", "growers", "firebombs", "throbs", "Downy", "contribution's",
+    "sago's", "Cole", "Knoxville", "leftmost", "Nell's", "Baffin", "barrings",
+    "contagions", "disencumbers", "countdown", "quintuple", "perihelion",
+    "creationism's", "actioning", "admiralty", "Mt's", "durability's",
+    "sewer's", "replicas", "oxide", "ripened", "Pisces's", "Cinerama's",
+    "catheters", "oppressive", "roosting", "foggiest", "properly", "Kareem",
+    "Ollie", "minuted", "vehicles", "eel", "remunerates", "swashbuckler's",
+    "remunerative", "sanguining", "Belem's", "forlornly", "rudders",
+    "officialdom", "countertenors", "Upton", "whoop", "animations", "arouses",
+    "millionths", "videocassette", "fledgling", "shake", "exterminated",
+    "Cain's", "trendiest", "wariest", "torpedoes", "airmails", "Cameron's",
+    "discord's", "spitefulness's", "thudded", "menaced", "takeovers",
+    "solicited", "wallpapers", "economic", "cache", "rechargeable", "gongs",
+    "droning", "exemption", "Alaskans", "toothed", "snifter", "Stephens",
+    "prejudge", "doctor's", "bobolinks", "rotates", "valuation's", "narrator",
+    "weaning", "uncle", "shelter", "destitution's", "Edgardo's", "gauge",
+    "Nice", "Adolf's", "rheumatics", "inheritances", "undesirables",
+    "Eileen's", "flyweight's", "scope", "possessiveness", "tipsily",
+    "effulgence", "rematch", "Baltic", "unsteadiest", "rodeos", "gloaming's",
+    "ringers", "randomised", "commissars", "destroyer's", "router",
+    "disengaging", "it's", "Albert", "rampantly", "varmint", "Adkins",
+    "chevron", "insomniac", "bobsledded", "masochist's", "chronometers",
+    "compaction", "Mauro", "sidled", "Highlander's", "snail's", "syllabifies",
+    "application's", "symmetrical", "blacking", "accent's", "sentimentalists",
+    "sonatas", "profanities", "sloping", "Araby", "percolate", "repeated",
+    "youthfulness's", "Loyola", "deliriously", "matriarch's", "tailors",
+    "rerouting", "hairpin", "dispersal", "endowment", "disquieting", "swat",
+    "neckerchieves", "wrinkles", "amoebas", "Darcy", "orthodontics's",
+    "milder", "sneezing", "prescience's", "pads", "wrought", "perspicuity's",
+    "materialist", "pull", "laundryman's", "lazily", "protractor's", "Vic",
+    "photocopier", "guardrooms", "cablecasting", "confirms", "excretions",
+    "combatant", "counterfeiters", "periwig", "genteelest", "router's",
+    "springy", "procreated", "syphon", "parent's", "bigwigs", "rebelled",
+    "milkmaids", "McGee's", "seaworthier", "Bellatrix's", "tenement",
+    "embryologists", "Vaselining", "burrow's", "tonnage's", "Petty's",
+    "chancels", "scouring", "mouser", "recompensed", "guarding", "editor",
+    "raster", "bourgeoisie's", "interpolating", "skinflint's", "transport",
+    "bullfinch", "needlessly", "withholds", "counterclockwise", "panicking",
+    "Ahriman", "flambeing", "contrary", "heartstrings", "whittled", "crib's",
+    "highlighter", "extroverted", "Martinique's", "racquets", "Maldivian",
+    "physiognomy", "Hammarskjold", "massage", "shingling", "neighbourhood",
+    "boobed", "vulture", "intercontinental", "cobblers", "peddlers",
+    "forthrightly", "germicide", "raindrop's", "fir's", "decaffeinates",
+    "wobblier", "abnegated", "cruiser's", "satiety", "trilled", "impending",
+    "gulf", "mountebank", "beltway", "reappointment", "cinematographer",
+    "pylon", "penthouses", "morally", "installs", "Walsh's", "drawstring",
+    "circus's", "Khayyam's", "Myrtle's", "ventrals", "category's",
+    "opportunistic", "grovelling", "warier", "upchuck", "hairdresser's",
+    "Montanans", "jobber", "dazzle", "encirclement's", "muffin's", "coronets",
+    "focus's", "footfall's", "subjunctives", "late", "pedagogued",
+    "dignitaries", "content", "blockbusters", "reminiscent", "mayor",
+    "specifier", "extinction", "nutshell's", "catbird's", "bundle",
+    "gracefulness", "exceed", "estranges", "chancy", "bankrupted", "Avery",
+    "Barnett", "succulence", "stacking", "ensnare", "truck", "embargo",
+    "persecutes", "translation's", "muskrat's", "illumines", "undercoat's",
+    "fleecier", "brick", "qualities", "imprecision", "reprisals", "discounts",
+    "harmonics", "Mann's", "terrorism", "interminable", "Santiago's",
+    "deepness", "tramples", "golder", "voyeurism's", "tent", "particle's",
+    "minuend", "waxwings", "knobby", "trustee", "funnily", "hotheadedness's",
+    "Kristin", "what", "bite", "murmur's", "pustule's", "weeknights",
+    "rocked", "athlete", "ventilates", "impresses", "daguerreotyping",
+    "Gross", "gambols", "villa", "maraud", "disapproval", "apostrophe's",
+    "sheaf", "noisemaker's", "autonomy's", "massing", "daemon's", "Thackeray",
+    "fermenting", "whammy", "philosophise", "empathy", "calamities",
+    "sunbathe", "Qom", "yahoo's", "coxcomb's", "move", "school's",
+    "rainmakers", "shipwreck", "potbelly's", "courageously", "current",
+    "Aleut", "treaties", "U", "always", "Bosch", "impregnating", "bud's",
+    "carat", "centrists", "acquaintance's", "convoy's", "chichis",
+    "restraint's", "Cosby", "factotums", "handshaking", "paragon's",
+    "mileages", "Tammie", "cartoonists", "lemmas", "lowliness's", "onion's",
+    "E's", "Bible", "Cranmer", "fob's", "minks", "overstocking", "Willamette",
+    "needle's", "scuppers", "Carborundum", "upwardly", "tallies", "aptitude",
+    "synod", "nasturtium's", "Pensacola", "snappish", "merino", "sups",
+    "fingerboard's", "prodigy's", "narcissism's", "substantial", "lug",
+    "establishing", "Vergil's", "patrimonies", "shorted", "forestation",
+    "undeniable", "Katmandu", "lamination", "trollop's", "odd", "stanza",
+    "paraplegic", "melanin", "Rico", "foreman", "stereotypes", "affinity's",
+    "cleansing", "sautéing", "epochs", "crooners", "manicured", "undisclosed",
+    "propel", "usage", "Alioth's", "Aurelia's", "peruse", "Vassar's",
+    "Demosthenes's", "Brazos", "supermarket", "scribbles", "Jekyll's",
+    "discomfort's", "mastiffs", "ballasting", "Figueroa", "turnstiles",
+    "convince", "Shelton's", "Gustavo", "shunting", "Fujitsu's", "fining's",
+    "hippos", "dam's", "expressionists", "peewee", "troop's"
+  ]
+  WORDS_SIZE = WORDS.size
+
+  def random_word
+    return WORDS[rand(WORDS_SIZE)]
+  end
+
+  def random_sentence(max_len)
+    sentence = ""
+    (1 + rand(max_len)).times { sentence << " " << random_word }
+    return sentence
+  end
+
+  def random_doc(max_fields = 10, max_elements = 10, max_len = 100)
+    doc = {}
+    (1 + rand(max_fields)).times do
+      field = random_word.intern
+      elem_count = rand(max_elements) + 1
+      if (elem_count == 1)
+        doc[field] = random_sentence(max_len)
+      else
+        doc[field] = []
+        elem_count.times { doc[field] << random_sentence(max_len)}
+      end
+    end
+    return doc
+  end
 end

data/test/unit/index/th_doc.rb

@@ -1,8 +1,4 @@
-require File.dirname(__FILE__) + "/../../test_helper"
-
-
 module IndexTestHelper
-  include Ferret::Document
   include Ferret::Index
   include Ferret::Analysis
   include Ferret::Search
@@ -16,23 +12,34 @@ module IndexTestHelper
   BINARY_DATA = IndexTestHelper.make_binary(256)
   COMPRESSED_BINARY_DATA = IndexTestHelper.make_binary(56)
 
-  def IndexTestHelper.prepare_document
-    doc = Document.new()
-
-    doc << Field.new("text_field1", "field one text", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::NO)
-    doc << Field.new("text_field2", "field field field two text", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    doc << Field.new("key_field", "keyword", Field::Store::YES, Field::Index::UNTOKENIZED)
-    doc << Field.new("unindexed_field", "unindexed field text", Field::Store::YES, Field::Index::NO)
-    doc << Field.new("unstored_field1", "unstored field text one", Field::Store::NO, Field::Index::TOKENIZED, Field::TermVector::NO)
-    doc << Field.new("unstored_field2", "unstored field text two", Field::Store::NO, Field::Index::TOKENIZED, Field::TermVector::YES)
-    doc << Field.new("compressed_field", "compressed text", Field::Store::COMPRESS, Field::Index::TOKENIZED, Field::TermVector::YES)
-    doc << Field.new_binary_field("binary_field", BINARY_DATA, Field::Store::YES)
-    doc << Field.new_binary_field("compressed_binary_field", COMPRESSED_BINARY_DATA, Field::Store::COMPRESS)
-    return doc
+  def IndexTestHelper.prepare_document(dir)
+    fis = FieldInfos.new
+    fis.add_field(:text_field1, :term_vector => :no)
+    fis.add_field(:text_field2)
+    fis.add_field(:key_field, :index => :untokenized)
+    fis.add_field(:unindexed_field, :index => :no)
+    fis.add_field(:unstored_field1, :store => :no, :term_vector => :no)
+    fis.add_field(:unstored_field2, :store => :no, :term_vector => :yes)
+    fis.add_field(:compressed_field, :store => :compressed, :term_vector => :yes)
+    fis.add_field(:binary_field, :index => :no, :term_vector => :no)
+    fis.add_field(:compressed_binary_field, :store => :compressed,
+                  :index => :no, :term_vector => :no)
+    doc = {
+      :text_field1             => "field one text",
+      :text_field2             => "field field field two text",
+      :key_field               => "keyword",
+      :unindexed_field         => "unindexed field text",
+      :unstored_field1         => "unstored field text one",
+      :unstored_field2         => "unstored field text two",
+      :compressed_field        => "compressed text",
+      :binary_field            => BINARY_DATA,
+      :compressed_binary_field => COMPRESSED_BINARY_DATA
+    }
+    return doc, fis
   end
 
   def IndexTestHelper.prepare_documents
-    data = [
+    [
       ["apple", "green"],
       ["apple", "red"],
       ["orange", "orange"],
@@ -41,212 +48,268 @@ module IndexTestHelper
       ["mandarin", "orange"],
       ["peach", "orange"],
       ["apricot", "orange"]
-    ]
-
-    docs = []
-
-    data.each do |food|
-      doc = Document.new()
-      doc << Field.new("name", food[0], Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-      doc << Field.new("colour", food[1], Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-      docs << doc
-    end
-    return docs
-  end
-
-  def IndexTestHelper.write_document(dir, doc, segment="test", analyzer = WhiteSpaceAnalyzer.new(), similarity = Similarity.default())
-    writer = DocumentWriter.new(dir, analyzer, similarity, 50)
-    writer.add_document(segment, doc)
+    ].map { |food| {"name" => food[0], "colour" => food[1]} }
   end
 
   def IndexTestHelper.prepare_book_list
     books = [
-      {"author" => "P.H. Newby", "title" => "Something To Answer For", "year" => "1969"},
-      {"author" => "Bernice Rubens", "title" => "The Elected Member", "year" => "1970"},
-      {"author" => "V. S. Naipaul", "title" => "In a Free State", "year" => "1971"},
-      {"author" => "John Berger", "title" => "G", "year" => "1972"},
-      {"author" => "J. G. Farrell", "title" => "The Siege of Krishnapur", "year" => "1973"},
-      {"author" => "Stanley Middleton", "title" => "Holiday", "year" => "1974"},
-      {"author" => "Nadine Gordimer", "title" => "The Conservationist", "year" => "1974"},
-      {"author" => "Ruth Prawer Jhabvala", "title" => "Heat and Dust", "year" => "1975"},
-      {"author" => "David Storey", "title" => "Saville", "year" => "1976"},
-      {"author" => "Paul Scott", "title" => "Staying On", "year" => "1977"},
-      {"author" => "Iris Murdoch", "title" => "The Sea", "year" => "1978"},
-      {"author" => "Penelope Fitzgerald", "title" => "Offshore", "year" => "1979"},
-      {"author" => "William Golding", "title" => "Rites of Passage", "year" => "1980"},
-      {"author" => "Salman Rushdie", "title" => "Midnight's Children", "year" => "1981"},
-      {"author" => "Thomas Keneally", "title" => "Schindler's Ark", "year" => "1982"},
-      {"author" => "J. M. Coetzee", "title" => "Life and Times of Michael K", "year" => "1983"},
-      {"author" => "Anita Brookner", "title" => "Hotel du Lac", "year" => "1984"},
-      {"author" => "Keri Hulme", "title" => "The Bone People", "year" => "1985"},
-      {"author" => "Kingsley Amis", "title" => "The Old Devils", "year" => "1986"},
-      {"author" => "Penelope Lively", "title" => "Moon Tiger", "year" => "1987"},
-      {"author" => "Peter Carey", "title" => "Oscar and Lucinda", "year" => "1988"},
-      {"author" => "Kazuo Ishiguro", "title" => "The Remains of the Day", "year" => "1989"},
-      {"author" => "A. S. Byatt", "title" => "Possession", "year" => "1990"},
-      {"author" => "Ben Okri", "title" => "The Famished Road", "year" => "1991"},
-      {"author" => "Michael Ondaatje", "title" => "The English Patient", "year" => "1992"},
-      {"author" => "Barry Unsworth", "title" => "Sacred Hunger", "year" => "1992"},
-      {"author" => "Roddy Doyle", "title" => "Paddy Clarke Ha Ha Ha", "year" => "1993"},
-      {"author" => "James Kelman", "title" => "How Late It Was, How Late", "year" => "1994"},
-      {"author" => "Pat Barker", "title" => "The Ghost Road", "year" => "1995"},
-      {"author" => "Graham Swift", "title" => "Last Orders", "year" => "1996"},
-      {"author" => "Arundati Roy", "title" => "The God of Small Things", "year" => "1997"},
-      {"author" => "Ian McEwan", "title" => "Amsterdam", "year" => "1998"},
-      {"author" => "J. M. Coetzee", "title" => "Disgrace", "year" => "1999"},
-      {"author" => "Margaret Atwood", "title" => "The Blind Assassin", "year" => "2000"},
-      {"author" => "Peter Carey", "title" => "True History of the Kelly Gang", "year" => "2001"},
-      {"author" => "Yann Martel", "title" => "The Life of Pi", "year" => "2002"},
-      {"author" => "DBC Pierre", "title" => "Vernon God Little", "year" => "2003"}
+      {"author" => "P.H. Newby",
+        "title" => "Something To Answer For",
+         "year" => "1969"},
+      {"author" => "Bernice Rubens",
+        "title" => "The Elected Member",
+         "year" => "1970"},
+      {"author" => "V. S. Naipaul",
+        "title" => "In a Free State",
+         "year" => "1971"},
+      {"author" => "John Berger",
+        "title" => "G",
+         "year" => "1972"},
+      {"author" => "J. G. Farrell",
+        "title" => "The Siege of Krishnapur",
+         "year" => "1973"},
+      {"author" => "Stanley Middleton",
+        "title" => "Holiday",
+         "year" => "1974"},
+      {"author" => "Nadine Gordimer",
+        "title" => "The Conservationist",
+         "year" => "1974"},
+      {"author" => "Ruth Prawer Jhabvala",
+        "title" => "Heat and Dust",
+         "year" => "1975"},
+      {"author" => "David Storey",
+        "title" => "Saville",
+         "year" => "1976"},
+      {"author" => "Paul Scott",
+        "title" => "Staying On",
+         "year" => "1977"},
+      {"author" => "Iris Murdoch",
+        "title" => "The Sea",
+         "year" => "1978"},
+      {"author" => "Penelope Fitzgerald",
+        "title" => "Offshore",
+         "year" => "1979"},
+      {"author" => "William Golding",
+        "title" => "Rites of Passage",
+         "year" => "1980"},
+      {"author" => "Salman Rushdie",
+        "title" => "Midnight's Children",
+         "year" => "1981"},
+      {"author" => "Thomas Keneally",
+        "title" => "Schindler's Ark",
+         "year" => "1982"},
+      {"author" => "J. M. Coetzee",
+        "title" => "Life and Times of Michael K",
+         "year" => "1983"},
+      {"author" => "Anita Brookner",
+        "title" => "Hotel du Lac",
+         "year" => "1984"},
+      {"author" => "Keri Hulme",
+        "title" => "The Bone People",
+         "year" => "1985"},
+      {"author" => "Kingsley Amis",
+        "title" => "The Old Devils",
+         "year" => "1986"},
+      {"author" => "Penelope Lively",
+        "title" => "Moon Tiger",
+         "year" => "1987"},
+      {"author" => "Peter Carey",
+        "title" => "Oscar and Lucinda",
+         "year" => "1988"},
+      {"author" => "Kazuo Ishiguro",
+        "title" => "The Remains of the Day",
+         "year" => "1989"},
+      {"author" => "A. S. Byatt",
+        "title" => "Possession",
+         "year" => "1990"},
+      {"author" => "Ben Okri",
+        "title" => "The Famished Road",
+         "year" => "1991"},
+      {"author" => "Michael Ondaatje",
+        "title" => "The English Patient",
+         "year" => "1992"},
+      {"author" => "Barry Unsworth",
+        "title" => "Sacred Hunger",
+         "year" => "1992"},
+      {"author" => "Roddy Doyle",
+        "title" => "Paddy Clarke Ha Ha Ha",
+         "year" => "1993"},
+      {"author" => "James Kelman",
+        "title" => "How Late It Was, How Late",
+         "year" => "1994"},
+      {"author" => "Pat Barker",
+        "title" => "The Ghost Road",
+         "year" => "1995"},
+      {"author" => "Graham Swift",
+        "title" => "Last Orders",
+         "year" => "1996"},
+      {"author" => "Arundati Roy",
+        "title" => "The God of Small Things",
+         "year" => "1997"},
+      {"author" => "Ian McEwan",
+        "title" => "Amsterdam",
+         "year" => "1998"},
+      {"author" => "J. M. Coetzee",
+        "title" => "Disgrace",
+         "year" => "1999"},
+      {"author" => "Margaret Atwood",
+        "title" => "The Blind Assassin",
+         "year" => "2000"},
+      {"author" => "Peter Carey",
+        "title" => "True History of the Kelly Gang",
+         "year" => "2001"},
+      {"author" => "Yann Martel",
+        "title" => "The Life of Pi",
+         "year" => "2002"},
+      {"author" => "DBC Pierre",
+        "title" => "Vernon God Little",
+         "year" => "2003"}
     ]
-    docs = []
-
-    books.each do |book|
-      doc = Document.new()
-      doc << Field.new("author", book["author"], Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-      doc << Field.new("title", book["title"], Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-      doc << Field.new("year", book["year"], Field::Store::YES, Field::Index::NO, Field::TermVector::NO)
-      docs << doc
-    end
-    return docs
   end
 
-  IR_TEST_DOC_CNT = 64
-
-  def IndexTestHelper.prepare_ir_test_docs()
-    body = "body"
-    title = "title"
-    author = "author"
-    text = "text"
-    year = "year"
-    changing_field = "changing_field"
+  def self.prepare_ir_test_fis
+    fis = FieldInfos.new
+    fis.add_field(:body)
+    fis.add_field(:changing_field, :term_vector => :no)
+    fis.add_field(:title, :index => :untokenized, :term_vector => :with_offsets)
+    fis.add_field(:author, :term_vector => :with_positions)
+    fis.add_field(:year, :index => :no, :term_vector => :no)
+    fis.add_field(:text, :store => :no, :term_vector => :no)
+  end
 
-    docs = Array.new(IR_TEST_DOC_CNT)
-    docs[0] = Document.new()
-    docs[0] << Field.new(body, "Where is Wally", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    docs[0] << Field.new(changing_field, "word3 word4 word1 word2 word1 word3 word4 word1 word3 word3", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::NO)
-    docs[1] = Document.new()
-    docs[1] << Field.new(body, "Some Random Sentence read", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    docs[2] = Document.new()
-    docs[2] << Field.new(body, "Some read Random Sentence read", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    docs[3] = Document.new()
-    docs[3] << Field.new(title, "War And Peace", Field::Store::YES, Field::Index::UNTOKENIZED, Field::TermVector::WITH_OFFSETS)
-    docs[3] << Field.new(body, "word3 word4 word1 word2 word1 word3 word4 word1 word3 word3", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    docs[3] << Field.new(author, "Leo Tolstoy", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS)
-    docs[3] << Field.new(year, "1865", Field::Store::YES, Field::Index::NO, Field::TermVector::NO)
-    docs[3] << Field.new(text, "more text which is not stored", Field::Store::NO, Field::Index::TOKENIZED, Field::TermVector::NO)
-    docs[4] = Document.new()
-    docs[4] << Field.new(body, "Some Random Sentence", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    docs[5] = Document.new()
-    docs[5] << Field.new(body, "Here's Wally", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    docs[6] = Document.new()
-    docs[6] << Field.new(body, "Some Random Sentence read read read read", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    docs[7] = Document.new()
-    docs[7] << Field.new(body, "Some Random Sentence", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    docs[8] = Document.new()
-    docs[8] << Field.new(body, "Some Random Sentence", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    docs[9] = Document.new()
-    docs[9] << Field.new(body, "read Some Random Sentence read this will be used after unfinished next position read", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    docs[10] = Document.new()
-    docs[10] << Field.new(body, "Some read Random Sentence", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    docs[10] << Field.new(changing_field, "word3 word4 word1 word2 word1 word3 word4 word1 word3 word3", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::YES)
-    docs[11] = Document.new()
-    docs[11] << Field.new(body, "And here too. Well, maybe Not", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    docs[12] = Document.new()
-    docs[12] << Field.new(body, "Some Random Sentence", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    docs[13] = Document.new()
-    docs[13] << Field.new(body, "Some Random Sentence", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    docs[14] = Document.new()
-    docs[14] << Field.new(body, "Some Random Sentence", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    docs[15] = Document.new()
-    docs[15] << Field.new(body, "Some read Random Sentence", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    docs[16] = Document.new()
-    docs[16] << Field.new(body, "Some Random read read Sentence", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    docs[17] = Document.new()
-    docs[17] << Field.new(body, "Some Random read Sentence", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    docs[17] << Field.new(changing_field, "word3 word4 word1 word2 word1 word3 word4 word1 word3 word3", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS)
-    docs[18] = Document.new()
-    docs[18] << Field.new(body, "Wally Wally Wally", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    docs[19] = Document.new()
-    docs[19] << Field.new(body, "Some Random Sentence", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    docs[19] << Field.new(changing_field, "word3 word4 word1 word2 word1 word3 word4 word1 word3 word3", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_OFFSETS)
-    docs[20] = Document.new()
-    docs[20] << Field.new(body, "Wally is where Wally usually likes to go. Wally Mart! Wally likes shopping there for Where's Wally books. Wally likes to read", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    docs[20] << Field.new(changing_field, "word3 word4 word1 word2 word1 word3 word4 word1 word3 word3", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    docs[21] = Document.new()
-    docs[21] << Field.new(body, "Some Random Sentence read read read and more read read read", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
-    docs[21] << Field.new(changing_field, "word3 word4 word1 word2 word1 word3 word4 word1 word3 word3", Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::NO)
+  INDEX_TEST_DOC_COUNT = 64
+  def self.prepare_ir_test_docs
+    docs = []
+    docs[0] = {
+      :body => "Where is Wally",
+      :changing_field => "word3 word4 word1 word2 word1 word3 word4 word1 " +
+                         "word3 word3",
+    }
+    docs[1] = {
+      :body => "Some Random Sentence read"
+    }
+    docs[2] = {
+      :body => "Some read Random Sentence read"
+    }
+    docs[3] = {
+      :title  => "War And Peace",
+      :body   => "word3 word4 word1 word2 word1 word3 word4 word1 word3 word3",
+      :author => "Leo Tolstoy",
+      :year   => "1865",
+      :text   => "more text which is not stored"
+    }
+    docs[4] = {
+      :body => "Some Random Sentence"
+    }
+    docs[5] = {
+      :body => "Here's Wally"
+    }
+    docs[6] = {
+      :body => "Some Random Sentence read read read read"
+    }
+    docs[7] = {
+      :body => "Some Random Sentence"
+    }
+    docs[8] = {
+      :body => "Some Random Sentence"
+    }
+    docs[9] = {
+      :body => "read Some Random Sentence read this will be used after " +
+               "unfinished next position read"
+    }
+    docs[10] = {
+      :body => "Some read Random Sentence",
+      :changing_field => "word3 word4 word1 word2 word1 word3 word4 word1 " +
+                         "word3 word3"
+    }
+    docs[11] = {
+      :body => "And here too. Well, maybe Not"
+    }
+    docs[12] = {
+      :body => "Some Random Sentence"
+    }
+    docs[13] = {
+      :body => "Some Random Sentence"
+    }
+    docs[14] = {
+      :body => "Some Random Sentence"
+    }
+    docs[15] = {
+      :body => "Some Random Sentence"
+    }
+    docs[16] = {
+      :body => "Some Random read read Sentence"
+    }
+    docs[17] = {
+      :body => "Some Random read Sentence",
+      :changing_field => "word3 word4 word1 word2 word1 word3 word4 word1 " +
+                         "word3 word3"
+    }
+    docs[18] = {
+      :body => "Wally Wally Wally"
+    }
+    docs[19] = {
+      :body => "Some Random Sentence",
+      :changing_field => "word3 word4 word1 word2 word1 word3 word4 word1 " +
+                         "word3 word3"
+    }
+    docs[20] = {
+      :body => "Wally is where Wally usually likes to go. Wally Mart! Wally " +
+               "likes shopping there for Where's Wally books. Wally likes " +
+               "to read",
+      :changing_field => "word3 word4 word1 word2 word1 word3 word4 word1 " +
+                         "word3 word3"
+    }
+    docs[21] = {
+      :body => "Some Random Sentence read read read and more read read read",
+      :changing_field => "word3 word4 word1 word2 word1 word3 word4 word1 " +
+                         "word3 word3"
+    }
 
     buf = ""
     21.times { buf << "skip " }
-    22.upto(IR_TEST_DOC_CNT) do |i|
+    22.upto(INDEX_TEST_DOC_COUNT-1) do |i|
       buf << "skip "
-      docs[i] = Document.new()
-      docs[i] << Field.new(text, buf.clone, Field::Store::NO, Field::Index::TOKENIZED, Field::TermVector::WITH_POSITIONS_OFFSETS)
+      docs[i] = {:text => buf.clone}
     end
     return docs
   end
 
-  def IndexTestHelper.prepare_search_docs
-    data = [
-      {"date" => "20050930", "field" => "word1",
-        "cat" => "cat1/"},
-      {"date" => "20051001", "field" => "word1 word2 the quick brown fox",
-        "cat" => "cat1/sub1"},
-      {"date" => "20051002", "field" => "word1 word3",
-        "cat" => "cat1/sub1/subsub1"},
-      {"date" => "20051003", "field" => "word1 word3",
-        "cat" => "cat1/sub2"},
-      {"date" => "20051004", "field" => "word1 word2",
-        "cat" => "cat1/sub2/subsub2"},
-      {"date" => "20051005", "field" => "word1",
-        "cat" => "cat2/sub1"},
-      {"date" => "20051006", "field" => "word1 word3",
-        "cat" => "cat2/sub1"},
-      {"date" => "20051007", "field" => "word1",
-        "cat" => "cat2/sub1"},
-      {"date" => "20051008", "field" => "word1 word2 word3 the fast brown fox",
-        "cat" => "cat2/sub1"},
-      {"date" => "20051009", "field" => "word1",
-        "cat" => "cat3/sub1"},
-      {"date" => "20051010", "field" => "word1",
-        "cat" => "cat3/sub1"},
-      {"date" => "20051011", "field" => "word1 word3 the quick red fox",
-        "cat" => "cat3/sub1"},
-      {"date" => "20051012", "field" => "word1",
-        "cat" => "cat3/sub1"},
-      {"date" => "20051013", "field" => "word1",
-        "cat" => "cat1/sub2"},
-      {"date" => "20051014", "field" => "word1 word3 the quick hairy fox",
-        "cat" => "cat1/sub1"},
-      {"date" => "20051015", "field" => "word1",
-        "cat" => "cat1/sub2/subsub1"},
-      {"date" => "20051016",
-        "field" => "word1 the quick fox is brown and hairy and a little red",
-        "cat" => "cat1/sub1/subsub2"},
-      {"date" => "20051017", "field" => "word1 the brown fox is quick and red",
-        "cat" => "cat1/"}
-    ]
-
-    docs = []
-    data.each_with_index do |fields, i|
-      doc = Document.new()
-      doc.boost = i+1
+  INDEX_TEST_DOCS = self.prepare_ir_test_docs()
+  INDEX_TEST_FIS = self.prepare_ir_test_fis()
 
-      fields.each_pair do |field, text|
-        doc << Field.new(field, text, Field::Store::YES, Field::Index::TOKENIZED, Field::TermVector::NO, false)
-      end
-      docs << doc
+  def self.prepare_search_docs
+    i = 1
+    [
+      ["20050930", "cat1/",             "word1"                               ],
+      ["20051001", "cat1/sub1",         "word1 word2 the quick brown fox"     ],
+      ["20051002", "cat1/sub1/subsub1", "word1 word3"                         ],
+      ["20051003", "cat1/sub2",         "word1 word3"                         ],
+      ["20051004", "cat1/sub2/subsub2", "word1 word2"                         ],
+      ["20051005", "cat2/sub1",         "word1"                               ],
+      ["20051006", "cat2/sub1",         "word1 word3"                         ],
+      ["20051007", "cat2/sub1",         "word1"                               ],
+      ["20051008", "cat2/sub1",         "word1 word2 word3 the fast brown fox"],
+      ["20051009", "cat3/sub1",         "word1"                               ],
+      ["20051010", "cat3/sub1",         "word1"                               ],
+      ["20051011", "cat3/sub1",         "word1 word3 the quick red fox"       ],
+      ["20051012", "cat3/sub1",         "word1"                               ],
+      ["20051013", "cat1/sub2",         "word1"                               ],
+      ["20051014", "cat1/sub1",         "word1 word3 the quick hairy fox"     ],
+      ["20051015", "cat1/sub2/subsub1", "word1"                               ],
+      ["20051016", "cat1/sub1/subsub2",
+        "word1 the quick fox is brown and hairy and a little red"             ],
+      ["20051017", "cat1/",
+        "word1 the brown fox is quick and red"                                ]
+    ].map do |date, category, field| 
+      doc = Ferret::Document.new(i)
+      i += 1
+      doc[:date] = date
+      doc[:category] = category
+      doc[:field] = field
+      doc
     end
-    return docs
-  end
-
-  def IndexTestHelper.explain (query, searcher, field)
-    top_docs = searcher.search(query)
-    top_docs.score_docs.each { |sd|
-      puts "\nDoc #{sd.doc}: #{searcher.doc(sd.doc)[field]}\n#{searcher.explain(query, sd.doc).to_s}\n"
-    }
   end
 
+  SEARCH_TEST_DOCS = self.prepare_search_docs()
 end
-