1# Copyright (c) 2012-2015 International Business Machines 2# Corporation and others. All Rights Reserved. 3# 4# This file should be in UTF-8 with a signature byte sequence ("BOM"). 5# 6# collationtest.txt: Collation test data. 7# 8# created on: 2012apr13 9# created by: Markus W. Scherer 10 11# A line with "** test: description" is used for verbose and error output. 12 13# A collator can be set with "@ root" or "@ locale language-tag", 14# for example "@ locale de-u-co-phonebk". 15# An old-style locale ID can also be used, for example "@ locale de@collation=phonebook". 16 17# A collator can be built with "@ rules". 18# An "@ rules" line is followed by one or more lines with the tailoring rules. 19 20# A collator can be modified with "% attribute=value". 21 22# "* compare" tests the order (= or <) of the following strings. 23# The relation can be "=" or "<" (the level of the difference is not specified) 24# or "<1", "<2", "<c", "<3", "<4" (indicating the level of the difference). 25 26# Test sections ("* compare") are terminated by 27# definitions of new collators, changing attributes, or new test sections. 28 29** test: simple CEs & expansions 30# Many types of mappings are tested elsewhere, including via the UCA conformance tests. 31# Here we mostly cover a few unusual mappings. 32@ rules 33&\x01 # most control codes are ignorable 34<<<\u0300 # tertiary CE 35&9<\x00 # NUL not ignorable 36&\uA00A\uA00B=\uA002 # two long-primary CEs 37&\uA00A\uA00B\u00050005=\uA003 # three CEs, require 64 bits 38 39* compare 40= \x01 41= \x02 42<3 \u0300 43<1 9 44<1 \x00 45= \x01\x00\x02 46<1 a 47<3 a\u0300 48<2 a\u0308 49= ä 50<1 b 51<1 か # Hiragana Ka (U+304B) 52<2 か\u3099 # plus voiced sound mark 53= が # Hiragana Ga (U+304C) 54<1 \uA00A\uA00B 55= \uA002 56<1 \uA00A\uA00B\u00050004 57<1 \uA00A\uA00B\u00050005 58= \uA003 59<1 \uA00A\uA00B\u00050006 60 61** test: contractions 62# Create some interesting mappings, and map some normalization-inert characters 63# (which are not subject to canonical reordering) 64# to some of the same CEs to check the sequence of CEs. 65@ rules 66 67# Contractions starting with 'a' should not continue with any character < U+0300 68# so that we can test a shortcut for that. 69&a=ⓐ 70&b<bz=ⓑ 71&d<dz\u0301=ⓓ # d+z+acute 72&z 73<a\u0301=Ⓐ # a+acute sorts after z 74<a\u0301\u0301=Ⓑ # a+acute+acute 75<a\u0301\u0301\u0358=Ⓒ # a+acute+acute+dot above right 76<a\u030a=Ⓓ # a+ring 77<a\u0323=Ⓔ # a+dot below 78<a\u0323\u0358=Ⓕ # a+dot below+dot above right 79<a\u0327\u0323\u030a=Ⓖ # a+cedilla+dot below+ring 80<a\u0327\u0323bz=Ⓗ # a+cedilla+dot below+b+z 81 82&\U0001D158=⁰ # musical notehead black (has a symbol primary) 83<\U0001D158\U0001D165=¼ # musical quarter note 84 85# deliberately missing prefix contractions: 86# dz 87# a\u0327 88# a\u0327\u0323 89# a\u0327\u0323b 90 91&\x01 92<<<\U0001D165=¹ # musical stem (ccc=216) 93<<<\U0001D16D=² # musical augmentation dot (ccc=226) 94<<<\U0001D165\U0001D16D=³ # stem+dot (ccc=216 226) 95&\u0301=❶ # acute (ccc=230) 96&\u030a=❷ # ring (ccc=230) 97&\u0308=❸ # diaeresis (ccc=230) 98<<\u0308\u0301=❹ # diaeresis+acute (=dialytika tonos) (ccc=230 230) 99&\u0327=❺ # cedilla (ccc=202) 100&\u0323=❻ # dot below (ccc=220) 101&\u0331=❼ # macron below (ccc=220) 102<<\u0331\u0358=❽ # macron below+dot above right (ccc=220 232) 103&\u0334=❾ # tilde overlay (ccc=1) 104&\u0358=❿ # dot above right (ccc=232) 105 106&\u0f71=① # tibetan vowel sign aa 107&\u0f72=② # tibetan vowel sign i 108# \u0f71\u0f72 # tibetan vowel sign aa + i = ii = U+0F73 109&\u0f73=③ # tibetan vowel sign ii (ccc=0 but lccc=129) 110 111** test: simple contractions 112 113# Some strings are chosen to cause incremental contiguous contraction matching to 114# go into partial matches for prefixes of contractions 115# (where the prefixes are deliberately not also contractions). 116# When there is no complete match, then the matching code must back out of those 117# so that discontiguous contractions work as specified. 118 119* compare 120# contraction starter with no following text, or mismatch, or blocked 121<1 a 122= ⓐ 123<1 aa 124= ⓐⓐ 125<1 ab 126= ⓐb 127<1 az 128= ⓐz 129 130* compare 131<1 a 132<2 a\u0308\u030a # ring blocked by diaeresis 133= ⓐ❸❷ 134<2 a\u0327 135= ⓐ❺ 136 137* compare 138<2 \u0308 139= ❸ 140<2 \u0308\u030a\u0301 # acute blocked by ring 141= ❸❷❶ 142 143* compare 144<1 \U0001D158 145= ⁰ 146<1 \U0001D158\U0001D165 147= ¼ 148 149# no discontiguous contraction because of missing prefix contraction d+z, 150# and a starter ('z') after the 'd' 151* compare 152<1 dz\u0323\u0301 153= dz❻❶ 154 155# contiguous contractions 156* compare 157<1 abz 158= ⓐⓑ 159<1 abzz 160= ⓐⓑz 161 162* compare 163<1 a 164<1 z 165<1 a\u0301 166= Ⓐ 167<1 a\u0301\u0301 168= Ⓑ 169<1 a\u0301\u0301\u0358 170= Ⓒ 171<1 a\u030a 172= Ⓓ 173<1 a\u0323\u0358 174= Ⓕ 175<1 a\u0327\u0323\u030a # match despite missing prefix 176= Ⓖ 177<1 a\u0327\u0323bz 178= Ⓗ 179 180* compare 181<2 \u0308\u0308\u0301 # acute blocked from first diaeresis, contracts with second 182= ❸❹ 183 184* compare 185<1 \U0001D158\U0001D165 186= ¼ 187 188* compare 189<3 \U0001D165\U0001D16D 190= ³ 191 192** test: discontiguous contractions 193* compare 194<1 a\u0327\u030a # a+ring skips cedilla 195= Ⓓ❺ 196<2 a\u0327\u0327\u030a # a+ring skips 2 cedillas 197= Ⓓ❺❺ 198<2 a\u0327\u0327\u0327\u030a # a+ring skips 3 cedillas 199= Ⓓ❺❺❺ 200<2 a\u0334\u0327\u0327\u030a # a+ring skips tilde overlay & 2 cedillas 201= Ⓓ❾❺❺ 202<1 a\u0327\u0323 # a+dot below skips cedilla 203= Ⓔ❺ 204<1 a\u0323\u0301\u0358 # a+dot below+dot ab.r.: 2-char match, then skips acute 205= Ⓕ❶ 206<2 a\u0334\u0323\u0358 # a+dot below skips tilde overlay 207= Ⓕ❾ 208 209* compare 210<2 \u0331\u0331\u0358 # macron below+dot ab.r. skips the second macron below 211= ❽❼ 212 213* compare 214<1 a\u0327\u0331\u0323\u030a # a+ring skips cedilla, macron below, dot below (dot blocked by macron) 215= Ⓓ❺❼❻ 216<1 a\u0327\u0323\U0001D16D\u030a # a+dot below skips cedilla 217= Ⓔ❺²❷ 218<2 a\u0327\u0327\u0323\u030a # a+dot below skips 2 cedillas 219= Ⓔ❺❺❷ 220<2 a\u0327\u0323\u0323\u030a # a+dot below skips cedilla 221= Ⓔ❺❻❷ 222<2 a\u0334\u0327\u0323\u030a # a+dot below skips tilde overlay & cedilla 223= Ⓔ❾❺❷ 224 225* compare 226<1 \U0001D158\u0327\U0001D165 # quarter note skips cedilla 227= ¼❺ 228<1 a\U0001D165\u0323 # a+dot below skips stem 229= Ⓔ¹ 230 231# partial contiguous match, backs up, matches discontiguous contraction 232<1 a\u0327\u0323b 233= Ⓔ❺b 234<1 a\u0327\u0323ba 235= Ⓔ❺bⓐ 236 237# a+acute+acute+dot above right skips cedilla, continues matching 2 same-ccc combining marks 238* compare 239<1 a\u0327\u0301\u0301\u0358 240= Ⓒ❺ 241 242# FCD but not NFD 243* compare 244<1 a\u0f73\u0301 # a+acute skips tibetan ii 245= Ⓐ③ 246 247# FCD but the 0f71 inside the 0f73 must be skipped 248# to match the discontiguous contraction of the first 0f71 with the trailing 0f72 inside the 0f73 249* compare 250<1 \u0f71\u0f73 # == \u0f73\u0f71 == \u0f71\u0f71\u0f72 251= ③① 252 253** test: discontiguous contractions with nested contractions 254* compare 255<1 a\u0323\u0308\u0301\u0358 256= Ⓕ❹ 257<2 a\u0323\u0308\u0301\u0308\u0301\u0358 258= Ⓕ❹❹ 259 260** test: discontiguous contractions with interleaved contractions 261* compare 262# a+ring & cedilla & macron below+dot above right 263<1 a\u0327\u0331\u030a\u0358 264= Ⓓ❺❽ 265 266# a+ring & 1x..3x macron below+dot above right 267<2 a\u0331\u030a\u0358 268= Ⓓ❽ 269<2 a\u0331\u0331\u030a\u0358\u0358 270= Ⓓ❽❽ 271# also skips acute 272<2 a\u0331\u0331\u0331\u030a\u0301\u0358\u0358\u0358 273= Ⓓ❽❽❽❶ 274 275# a+dot below & stem+augmentation dot, followed by contiguous d+z+acute 276<1 a\U0001D165\u0323\U0001D16Ddz\u0301 277= Ⓔ³ⓓ 278 279** test: some simple string comparisons 280@ root 281* compare 282# first string compares against "" 283= \u0000 284< a 285<1 b 286<3 B 287= \u0000B\u0000 288 289** test: compare with strength=primary 290% strength=primary 291* compare 292<1 a 293<1 b 294= B 295 296** test: compare with strength=secondary 297% strength=secondary 298* compare 299<1 a 300<1 b 301= B 302 303** test: compare with strength=tertiary 304% strength=tertiary 305* compare 306<1 a 307<1 b 308<3 B 309 310** test: compare with strength=quaternary 311% strength=quaternary 312* compare 313<1 a 314<1 b 315<3 B 316 317** test: compare with strength=identical 318% strength=identical 319* compare 320<1 a 321<1 b 322<3 B 323 324** test: côté with forwards secondary 325@ root 326* compare 327<1 cote 328<2 coté 329<2 côte 330<2 côté 331 332** test: côté with forwards secondary vs. U+FFFE merge separator 333# Merged sort keys: On each level, any difference in the first segment 334# must trump any further difference. 335* compare 336<1 cote\uFFFEcôté 337<2 coté\uFFFEcôte 338<2 côte\uFFFEcoté 339<2 côté\uFFFEcote 340 341** test: côté with backwards secondary 342% backwards=on 343* compare 344<1 cote 345<2 côte 346<2 coté 347<2 côté 348 349** test: côté with backwards secondary vs. U+FFFE merge separator 350# Merged sort keys: On each level, any difference in the first segment 351# must trump any further difference. 352* compare 353<1 cote\uFFFEcôté 354<2 côte\uFFFEcoté 355<2 coté\uFFFEcôte 356<2 côté\uFFFEcote 357 358** test: U+FFFE on identical level 359@ root 360% strength=identical 361* compare 362# All of these control codes are completely-ignorable, so that 363# their low code points are compared with the merge separator. 364# The merge separator must compare less than any other character. 365<1 \uFFFE\u0001\u0002\u0003 366<i \u0001\uFFFE\u0002\u0003 367<i \u0001\u0002\uFFFE\u0003 368<i \u0001\u0002\u0003\uFFFE 369 370* compare 371# The merge separator must even compare less than U+0000. 372<1 \uFFFE\u0000\u0000 373<i \u0000\uFFFE\u0000 374<i \u0000\u0000\uFFFE 375 376** test: Hani < surrogates < U+FFFD 377# Note: compareUTF8() treats unpaired surrogates like U+FFFD, 378# so with that the strings with surrogates will compare equal to each other 379# and equal to the string with U+FFFD. 380@ root 381% strength=identical 382* compare 383<1 abz 384<1 a\u4e00z 385<1 a\U00020000z 386<1 a\ud800z 387<1 a\udbffz 388<1 a\udc00z 389<1 a\udfffz 390<1 a\ufffdz 391 392** test: script reordering 393@ root 394% reorder Hani Zzzz digit 395* compare 396<1 ? 397<1 + 398<1 丂 399<1 a 400<1 α 401<1 5 402 403% reorder default 404* compare 405<1 ? 406<1 + 407<1 5 408<1 a 409<1 α 410<1 丂 411 412** test: empty rules 413@ rules 414* compare 415<1 a 416<2 ä 417<3 Ä 418<1 b 419 420** test: very simple rules 421@ rules 422&a=e<<<<q<<<<r<x<<<X<<y<<<Y;z,Z 423% strength=quaternary 424* compare 425<1 a 426= e 427<4 q 428<4 r 429<1 x 430<3 X 431<2 y 432<3 Y 433<2 z 434<3 Z 435 436** test: tailoring twice before a root position: primary 437@ rules 438&[before 1]b<p 439&[before 1]b<q 440* compare 441<1 a 442<1 p 443<1 q 444<1 b 445 446** test: tailoring twice before a root position: secondary 447@ rules 448&[before 2]ſ<<p 449&[before 2]ſ<<q 450* compare 451<1 s 452<2 p 453<2 q 454<2 ſ 455 456# secondary-before common weight 457@ rules 458&[before 2]b<<p 459&[before 2]b<<q 460* compare 461<1 a 462<1 p 463<2 q 464<2 b 465 466** test: tailoring twice before a root position: tertiary 467@ rules 468&[before 3]B<<<p 469&[before 3]B<<<q 470* compare 471<1 b 472<3 p 473<3 q 474<3 B 475 476# tertiary-before common weight 477@ rules 478&[before 3]b<<<p 479&[before 3]b<<<q 480* compare 481<1 a 482<1 p 483<3 q 484<3 b 485 486@ rules 487&[before 2]b<<s 488&[before 3]s<<<p 489&[before 3]s<<<q 490* compare 491<1 a 492<1 p 493<3 q 494<3 s 495<2 b 496 497** test: tailor after completely ignorable 498@ rules 499&\x00<<<x<<y 500* compare 501= \x00 502= \x1F 503<3 x 504<2 y 505 506** test: secondary tailoring gaps, ICU ticket 9362 507@ rules 508&[before 2]s<<'_' 509&s<<r # secondary between s and ſ (long s) 510&ſ<<*a-q # more than 15 between ſ and secondary CE boundary 511&[before 2][first primary ignorable]<<u<<v # between secondary CE boundary & lowest secondary CE 512&[last primary ignorable]<<y<<z 513 514* compare 515<2 u 516<2 v 517<2 \u0332 # lowest secondary CE 518<2 \u0308 519<2 y 520<2 z 521<1 s_ 522<2 ss 523<2 sr 524<2 sſ 525<2 sa 526<2 sb 527<2 sp 528<2 sq 529<2 sus 530<2 svs 531<2 rs 532 533** test: tertiary tailoring gaps, ICU ticket 9362 534@ rules 535&[before 3]t<<<'_' 536&t<<<r # tertiary between t and fullwidth t 537&ᵀ<<<*a-q # more than 15 between ᵀ (modifier letter T) and tertiary CE boundary 538&[before 3][first secondary ignorable]<<<u<<<v # between tertiary CE boundary & lowest tertiary CE 539&[last secondary ignorable]<<<y<<<z 540 541* compare 542<3 u 543<3 v 544# Note: The root collator currently does not map any characters to tertiary CEs. 545<3 y 546<3 z 547<1 t_ 548<3 tt 549<3 tr 550<3 tｔ 551<3 tᵀ 552<3 ta 553<3 tb 554<3 tp 555<3 tq 556<3 tut 557<3 tvt 558<3 rt 559 560** test: secondary & tertiary around root character 561@ rules 562&[before 2]m<<r 563&m<<s 564&[before 3]m<<<u 565&m<<<v 566* compare 567<1 l 568<1 r 569<2 u 570<3 m 571<3 v 572<2 s 573<1 n 574 575** test: secondary & tertiary around tailored item 576@ rules 577&m<x 578&[before 2]x<<r 579&x<<s 580&[before 3]x<<<u 581&x<<<v 582* compare 583<1 m 584<1 r 585<2 u 586<3 x 587<3 v 588<2 s 589<1 n 590 591** test: more nesting of secondary & tertiary before 592@ rules 593&[before 3]m<<<u 594&[before 2]m<<r 595&[before 3]r<<<q 596&m<<<w 597&m<<t 598&[before 3]w<<<v 599&w<<<x 600&w<<s 601* compare 602<1 l 603<1 q 604<3 r 605<2 u 606<3 m 607<3 v 608<3 w 609<3 x 610<2 s 611<2 t 612<1 n 613 614** test: case bits 615@ rules 616&w<x # tailored CE getting case bits 617 =uv=uV=Uv=UV # 2 chars -> 1 CE 618&ae=ch=cH=Ch=CH # 2 chars -> 2 CEs 619&rst=yz=yZ=Yz=YZ # 2 chars -> 3 CEs 620% caseFirst=lower 621* compare 622<1 ae 623= ch 624<3 cH 625<3 Ch 626<3 CH 627<1 rst 628= yz 629<3 yZ 630<3 Yz 631<3 YZ 632<1 w 633<1 x 634= uv 635<3 uV 636= Uv # mixed case on single CE cannot distinguish variations 637<3 UV 638 639** test: tertiary CEs, tertiary, caseLevel=off, caseFirst=lower 640@ rules 641&\u0001<<<t<<<T # tertiary CEs 642% caseFirst=lower 643* compare 644<1 aa 645<3 aat 646<3 aaT 647<3 aA 648<3 aAt 649<3 ata 650<3 aTa 651 652** test: tertiary CEs, tertiary, caseLevel=off, caseFirst=upper 653% caseFirst=upper 654* compare 655<1 aA 656<3 aAt 657<3 aa 658<3 aat 659<3 aaT 660<3 ata 661<3 aTa 662 663** test: reset on expansion, ICU tickets 9415 & 9593 664@ rules 665&æ<x # tailor the last primary CE so that x sorts between ae and af 666&æb=bæ # copy all reset CEs to make bæ sort the same 667&각<h # copy/tailor 3 CEs to make h sort before the next Hangul syllable 갂 668&⒀<<y # copy/tailor 4 CEs to make y sort with only a secondary difference 669&l·=z # handle the pre-context for · when fetching reset CEs 670 <<u # copy/tailor 2 CEs 671 672* compare 673<1 ae 674<2 æ 675<1 x 676<1 af 677 678* compare 679<1 aeb 680<2 æb 681= bæ 682 683* compare 684<1 각 685<1 h 686<1 갂 687<1 갃 688 689* compare 690<1 · # by itself: primary CE 691<1 l 692<2 l· # l+middle dot has only a secondary difference from l 693= z 694<2 u 695 696* compare 697<1 (13) 698<3 ⒀ # DUCET sets special tertiary weights in all CEs 699<2 y 700<1 (13[ 701 702% alternate=shifted 703* compare 704<1 (13) 705= 13 706<3 ⒀ 707= y # alternate=shifted removes the tailoring difference on the last CE 708<1 14 709 710** test: contraction inside extension, ICU ticket 9378 711@ rules 712&а<<х/й # all letters are Cyrillic 713* compare 714<1 ай 715<2 х 716 717** test: no duplicate tailored CEs for different reset positions with same CEs, ICU ticket 10104 718@ rules 719&t<x &ᵀ<y # same primary weights 720&q<u &[before 1]ꝗ<v # q and ꝗ are primary adjacent 721* compare 722<1 q 723<1 u 724<1 v 725<1 ꝗ 726<1 t 727<3 ᵀ 728<1 y 729<1 x 730 731# Principle: Each rule builds on the state of preceding rules and ignores following rules. 732 733** test: later rule does not affect earlier reset position, ICU ticket 10105 734@ rules 735&a < u < v < w &ov < x &b < v 736* compare 737<1 oa 738<1 ou 739<1 x # CE(o) followed by CE between u and w 740<1 ow 741<1 ob 742<1 ov 743 744** test: later rule does not affect earlier extension (1), ICU ticket 10105 745@ rules 746&a=x/b &v=b 747% strength=secondary 748* compare 749<1 B 750<1 c 751<1 v 752= b 753* compare 754<1 AB 755= x 756<1 ac 757<1 av 758= ab 759 760** test: later rule does not affect earlier extension (2), ICU ticket 10105 761@ rules 762&a <<< c / e &g <<< e / l 763% strength=secondary 764* compare 765<1 AE 766= c 767<2 æ 768<1 agl 769= ae 770 771** test: later rule does not affect earlier extension (3), ICU ticket 10105 772@ rules 773&a = b / c &d = c / e 774% strength=secondary 775* compare 776<1 AC # C is still only tertiary different from the original c 777= b 778<1 ade 779= ac 780 781** test: extension contains tailored character, ICU ticket 10105 782@ rules 783&a=e &b=u/e 784* compare 785<1 a 786= e 787<1 ba 788= be 789= u 790 791** test: add simple mappings for characters with root context 792@ rules 793&z=· # middle dot has a prefix mapping in the CLDR root 794&n=и # и (U+0438) has contractions in the root 795* compare 796<1 l 797<2 l· # root mapping for l|· still works 798<1 z 799= · 800* compare 801<1 n 802= и 803<1 И 804<1 и\u0306 # root mapping for й=и\u0306 still works 805= й 806<3 Й 807 808** test: add context mappings around characters with root context 809@ rules 810&z=·h # middle dot has a prefix mapping in the CLDR root 811&n=ә|и # и (U+0438) has contractions in the root 812* compare 813<1 l 814<2 l· # root mapping for l|· still works 815<1 z 816= ·h 817* compare 818<1 и 819<3 И 820<1 и\u0306 # root mapping for й=и\u0306 still works 821= й 822* compare 823<1 әn 824= әи 825<1 әo 826 827** test: many secondary CEs at the top of their range 828@ rules 829&[last primary ignorable]<<*\u2801-\u28ff 830* compare 831<2 \u0308 832<2 \u2801 833<2 \u2802 834<2 \u2803 835<2 \u2804 836<2 \u28fd 837<2 \u28fe 838<2 \u28ff 839<1 \x20 840 841** test: many tertiary CEs at the top of their range 842@ rules 843&[last secondary ignorable]<<<*a-z 844* compare 845<3 a 846<3 b 847<3 c 848<3 d 849# e..w 850<3 x 851<3 y 852<3 z 853<2 \u0308 854 855** test: tailor contraction together with nearly equivalent prefix, ICU ticket 10101 856@ rules 857&a=p|x &b=px &c=op 858* compare 859<1 b 860= px 861<3 B 862<1 c 863= op 864<3 C 865* compare 866<1 ca 867= opx # first contraction op, then prefix p|x 868<3 cA 869<3 Ca 870 871** test: reset position with prefix (pre-context), ICU ticket 10102 872@ rules 873&a=p|x &px=y 874* compare 875<1 pa 876= px 877= y 878<3 pA 879<1 q 880<1 x 881 882** test: prefix+contraction together (1), ICU ticket 10071 883@ rules 884&x=a|bc 885* compare 886<1 ab 887<1 Abc 888<1 abd 889<1 ac 890<1 aw 891<1 ax 892= abc 893<3 aX 894<3 Ax 895<1 b 896<1 bb 897<1 bc 898<3 bC 899<3 Bc 900<1 bd 901 902** test: prefix+contraction together (2), ICU ticket 10071 903@ rules 904&w=bc &x=a|b 905* compare 906<1 w 907= bc 908<3 W 909* compare 910<1 aw 911<1 ax 912= ab 913<3 aX 914<1 axb 915<1 axc 916= abc # prefix match a|b takes precedence over contraction match bc 917<3 abC 918<1 abd 919<1 ay 920 921** test: prefix+contraction together (3), ICU ticket 10071 922@ rules 923&x=a|b &w=bc # reverse order of rules as previous test, order should not matter here 924* compare # same "compare" sequences as previous test 925<1 w 926= bc 927<3 W 928* compare 929<1 aw 930<1 ax 931= ab 932<3 aX 933<1 axb 934<1 axc 935= abc # prefix match a|b takes precedence over contraction match bc 936<3 abC 937<1 abd 938<1 ay 939 940** test: no mapping p|c, falls back to contraction ch, CLDR ticket 5962 941@ rules 942&d=ch &v=p|ci 943* compare 944<1 pc 945<3 pC 946<1 pcH 947<1 pcI 948<1 pd 949= pch # no-prefix contraction ch matches 950<3 pD 951<1 pv 952= pci # prefix+contraction p|ci matches 953<3 pV 954 955** test: tailor in & around compact ranges of root primaries 956# The Ogham characters U+1681..U+169A are in simple ascending order of primary CEs 957# which should be reliably encoded as one range in the root elements data. 958@ rules 959&[before 1]ᚁ<a 960&ᚁ<b 961&[before 1]ᚂ<c 962&ᚂ<d 963&[before 1]ᚚ<y 964&ᚚ<z 965&[before 2]ᚁ<<r 966&ᚁ<<s 967&[before 3]ᚚ<<<t 968&ᚚ<<<u 969* compare 970<1 ᣵ # U+18F5 last Canadian Aboriginal 971<1 a 972<1 r 973<2 ᚁ 974<2 s 975<1 b 976<1 c 977<1 ᚂ 978<1 d 979<1 ᚃ 980<1 ᚙ 981<1 y 982<1 t 983<3 ᚚ 984<3 u 985<1 z 986<1 ᚠ # U+16A0 first Runic 987 988** test: suppressContractions 989@ rules 990&z<ch<әж [suppressContractions [·cә]] 991* compare 992<1 ch 993<3 cH # ch was suppressed 994<1 l 995<1 l· # primary difference, not secondary, because l|· was suppressed 996<1 ә 997<2 ә\u0308 # secondary difference, not primary, because contractions for ә were suppressed 998<1 әж 999<3 әЖ 1000 1001** test: Hangul & Jamo 1002@ rules 1003&L=\u1100 # first Jamo L 1004&V=\u1161 # first Jamo V 1005&T=\u11A8 # first Jamo T 1006&\uAC01<<*\u4E00-\u4EFF # first Hangul LVT syllable & lots of secondary diffs 1007* compare 1008<1 Lv 1009<3 LV 1010= \u1100\u1161 1011= \uAC00 1012<1 LVt 1013<3 LVT 1014= \u1100\u1161\u11A8 1015= \uAC00\u11A8 1016= \uAC01 1017<2 LVT\u0308 1018<2 \u4E00 1019<2 \u4E01 1020<2 \u4E80 1021<2 \u4EFF 1022<2 LV\u0308T 1023<1 \uAC02 1024 1025** test: adjust special reset positions according to previous rules, CLDR ticket 6070 1026@ rules 1027&[last variable]<x 1028[maxVariable space] # has effect only after building, no effect on following rules 1029&[last variable]<y 1030&[before 1][first regular]<z 1031* compare 1032<1 ? # some punctuation 1033<1 x 1034<1 y 1035<1 z 1036<1 $ # some symbol 1037 1038@ rules 1039&[last primary ignorable]<<x<<<y 1040&[last primary ignorable]<<z 1041* compare 1042<2 \u0358 1043<2 x 1044<3 y 1045<2 z 1046<1 \x20 1047 1048@ rules 1049&[last secondary ignorable]<<<x 1050&[last secondary ignorable]<<<y 1051* compare 1052<3 x 1053<3 y 1054<2 \u0358 1055 1056@ rules 1057&[before 2][first variable]<<z 1058&[before 2][first variable]<<y 1059&[before 3][first variable]<<<x 1060&[before 3][first variable]<<<w 1061&[before 1][first variable]<v 1062&[before 2][first variable]<<u 1063&[before 3][first variable]<<<t 1064&[before 2]\uFDD1\xA0<<s # FractionalUCA.txt: FDD1 00A0, SPACE first primary 1065* compare 1066<2 \u0358 1067<1 s 1068<2 \uFDD1\xA0 1069<1 t 1070<3 u 1071<2 v 1072<1 w 1073<3 x 1074<3 y 1075<2 z 1076<2 \t 1077 1078@ rules 1079&[before 2][first regular]<<z 1080&[before 3][first regular]<<<y 1081&[before 1][first regular]<x 1082&[before 3][first regular]<<<w 1083&[before 2]\uFDD1\u263A<<v # FractionalUCA.txt: FDD1 263A, SYMBOL first primary 1084&[before 3][first regular]<<<u 1085&[before 1][first regular]<p # primary before the boundary: becomes variable 1086&[before 3][first regular]<<<t # not affected by p 1087&[last variable]<q # after p! 1088* compare 1089<1 ? 1090<1 p 1091<1 q 1092<1 t 1093<3 u 1094<3 v 1095<1 w 1096<3 x 1097<1 y 1098<3 z 1099<1 $ 1100 1101# check that p & q are indeed variable 1102% alternate=shifted 1103* compare 1104= ? 1105= p 1106= q 1107<1 t 1108<3 u 1109<3 v 1110<1 w 1111<3 x 1112<1 y 1113<3 z 1114<1 $ 1115 1116@ rules 1117&[before 2][first trailing]<<z 1118&[before 1][first trailing]<y 1119&[before 3][first trailing]<<<x 1120* compare 1121<1 \u4E00 # first Han, first implicit 1122<1 \uFDD1\uFDD0 # FractionalUCA.txt: unassigned first primary 1123# Note: The root collator currently does not map any characters to the trailing first boundary primary. 1124<1 x 1125<3 y 1126<1 z 1127<2 \uFFFD # The root collator currently maps U+FFFD to the first real trailing primary. 1128 1129@ rules 1130&[before 2][first primary ignorable]<<z 1131&[before 2][first primary ignorable]<<y 1132&[before 3][first primary ignorable]<<<x 1133&[before 3][first primary ignorable]<<<w 1134* compare 1135= \x01 1136<2 w 1137<3 x 1138<3 y 1139<2 z 1140<2 \u0301 1141 1142@ rules 1143&[before 3][first secondary ignorable]<<<y 1144&[before 3][first secondary ignorable]<<<x 1145* compare 1146= \x01 1147<3 x 1148<3 y 1149<2 \u0301 1150 1151** test: canonical closure 1152@ rules 1153&X=A &U= 1154* compare 1155<1 U 1156=  1157= A\u0302 1158<2 Ú # U with acute 1159= U\u0301 1160= Ấ # A with circumflex & acute 1161= Â\u0301 1162= A\u0302\u0301 1163<1 X 1164= A 1165<2 X\u030A # with ring above 1166= Å 1167= A\u030A 1168= \u212B # Angstrom sign 1169 1170@ rules 1171&x=\u5140\u55C0 1172* compare 1173<1 x 1174= \u5140\u55C0 1175= \u5140\uFA0D 1176= \uFA0C\u55C0 1177= \uFA0C\uFA0D # CJK compatibility characters 1178<3 X 1179 1180# canonical closure on prefix rules, ICU ticket 9444 1181@ rules 1182&x=ä|ŝ 1183* compare 1184<1 äs # not tailored 1185<1 äx 1186= äŝ 1187= a\u0308s\u0302 1188= a\u0308ŝ 1189= äs\u0302 1190<3 äX 1191 1192** test: conjoining Jamo map to expansions 1193@ rules 1194&gg=\u1101 # Jamo Lead consonant GG 1195&nj=\u11AC # Jamo Trail consonant NJ 1196* compare 1197<1 gg\u1161nj 1198= \u1101\u1161\u11AC 1199= \uAE4C\u11AC 1200= \uAE51 1201<3 gg\u1161nJ 1202<1 \u1100\u1100 1203 1204** test: canonical tail closure, ICU ticket 5913 1205@ rules 1206&a<â 1207* compare 1208<1 a 1209<1 â # tailored 1210= a\u0302 1211<2 a\u0323\u0302 # discontiguous contraction 1212= ạ\u0302 # equivalent 1213= ậ # equivalent 1214<1 b 1215 1216@ rules 1217&a<ạ 1218* compare 1219<1 a 1220<1 ạ # tailored 1221= a\u0323 1222<2 a\u0323\u0302 # contiguous contraction plus extra diacritic 1223= ạ\u0302 # equivalent 1224= ậ # equivalent 1225<1 b 1226 1227# Tail closure should work even if there is a prefix and/or contraction. 1228@ rules 1229&a<\u5140|câ 1230# In order to find discontiguous contractions for \u5140|câ 1231# there must exist a mapping for \u5140|ca, regardless of what it maps to. 1232# (This follows from the UCA spec.) 1233&x=\u5140|ca 1234* compare 1235<1 \u5140a 1236= \uFA0Ca 1237<1 \u5140câ # tailored 1238= \uFA0Ccâ 1239= \u5140ca\u0302 1240= \uFA0Cca\u0302 1241<2 \u5140ca\u0323\u0302 # discontiguous contraction 1242= \uFA0Cca\u0323\u0302 1243= \u5140cạ\u0302 1244= \uFA0Ccạ\u0302 1245= \u5140cậ 1246= \uFA0Ccậ 1247<1 \u5140b 1248= \uFA0Cb 1249<1 \u5140x 1250= \u5140ca 1251 1252# Double-check that without the extra mapping there will be no discontiguous match. 1253@ rules 1254&a<\u5140|câ 1255* compare 1256<1 \u5140a 1257= \uFA0Ca 1258<1 \u5140câ # tailored 1259= \uFA0Ccâ 1260= \u5140ca\u0302 1261= \uFA0Cca\u0302 1262<1 \u5140b 1263= \uFA0Cb 1264<1 \u5140ca\u0323\u0302 # no discontiguous contraction 1265= \uFA0Cca\u0323\u0302 1266= \u5140cạ\u0302 1267= \uFA0Ccạ\u0302 1268= \u5140cậ 1269= \uFA0Ccậ 1270 1271@ rules 1272&a<cạ 1273* compare 1274<1 a 1275<1 cạ # tailored 1276= ca\u0323 1277<2 ca\u0323\u0302 # contiguous contraction plus extra diacritic 1278= cạ\u0302 # equivalent 1279= cậ # equivalent 1280<1 b 1281 1282# ᾢ = U+1FA2 GREEK SMALL LETTER OMEGA WITH PSILI AND VARIA AND YPOGEGRAMMENI 1283# = 03C9 0313 0300 0345 1284# ccc = 0, 230, 230, 240 1285@ rules 1286&δ=αῳ 1287# In order to find discontiguous contractions for αῳ 1288# there must exist a mapping for αω, regardless of what it maps to. 1289# (This follows from the UCA spec.) 1290&ε=αω 1291* compare 1292<1 δ 1293= αῳ 1294= αω\u0345 1295<2 αω\u0313\u0300\u0345 # discontiguous contraction 1296= αὠ\u0300\u0345 1297= αὢ\u0345 1298= αᾢ 1299<2 αω\u0300\u0313\u0345 1300= αὼ\u0313\u0345 1301= αῲ\u0313 # not FCD 1302<1 ε 1303= αω 1304 1305# Double-check that without the extra mapping there will be no discontiguous match. 1306@ rules 1307&δ=αῳ 1308* compare 1309<1 αω\u0313\u0300\u0345 # no discontiguous contraction 1310= αὠ\u0300\u0345 1311= αὢ\u0345 1312= αᾢ 1313<2 αω\u0300\u0313\u0345 1314= αὼ\u0313\u0345 1315= αῲ\u0313 # not FCD 1316<1 δ 1317= αῳ 1318= αω\u0345 1319 1320# Add U+0315 COMBINING COMMA ABOVE RIGHT which has ccc=232. 1321# Tests code paths where the tailored string has a combining mark 1322# that does not occur in any composite's decomposition. 1323@ rules 1324&δ=αὼ\u0315 1325* compare 1326<1 αω\u0313\u0300\u0315 # Not tailored: The grave accent blocks the comma above. 1327= αὠ\u0300\u0315 1328= αὢ\u0315 1329<1 δ 1330= αὼ\u0315 1331= αω\u0300\u0315 1332<2 αω\u0300\u0315\u0345 1333= αὼ\u0315\u0345 1334= αῲ\u0315 # not FCD 1335 1336** test: danish a+a vs. a-umlaut, ICU ticket 9319 1337@ rules 1338&z<aa 1339* compare 1340<1 z 1341<1 aa 1342<2 aa\u0308 1343= aä 1344 1345** test: Jamo L with and in prefix 1346# Useful for the Korean "searchjl" tailoring (instead of contractions of pairs of Jamo L). 1347@ rules 1348# Jamo Lead consonant G after G or GG 1349&[last primary ignorable]<<\u1100|\u1100=\u1101|\u1100 1350# Jamo Lead consonant GG sorts like G+G 1351&\u1100\u1100=\u1101 1352# Note: Making G|GG and GG|GG sort the same as G|G+G 1353# would require the ability to reset on G|G+G, 1354# or we could make G-after-G equal to some secondary-CE character, 1355# and reset on a pair of those. 1356# (It does not matter much if there are at most two G in a row in real text.) 1357* compare 1358<1 \u1100 1359<2 \u1100\u1100 # only one primary from a sequence of G lead consonants 1360= \u1101 1361<2 \u1100\u1100\u1100 1362= \u1101\u1100 1363# but not = \u1100\u1101, see above 1364<1 \u1100\u1161 1365= \uAC00 1366<2 \u1100\u1100\u1161 1367= \u1100\uAC00 # prefix match from the L of the LV syllable 1368= \u1101\u1161 1369= \uAE4C 1370 1371** test: proposed Korean "searchjl" tailoring with prefixes, CLDR ticket 6546 1372@ rules 1373# Low secondary CEs for Jamo V & T. 1374# Note: T should sort before V for proper syllable order. 1375&\u0332 # COMBINING LOW LINE (first primary ignorable) 1376<<\u1161<<\u1162 1377 1378# Korean Jamo lead consonant search rules, part 2: 1379# Make modern compound L jamo primary equivalent to non-compound forms. 1380 1381# Secondary CEs for Jamo L-after-L, greater than Jamo V & T. 1382&\u0313 # COMBINING COMMA ABOVE (second primary ignorable) 1383=\u1100|\u1100 1384=\u1103|\u1103 1385=\u1107|\u1107 1386=\u1109|\u1109 1387=\u110C|\u110C 1388 1389# Compound L Jamo map to equivalent expansions of primary+secondary CE. 1390&\u1100\u0313=\u1101<<<\u3132 # HANGUL CHOSEONG SSANGKIYEOK, HANGUL LETTER SSANGKIYEOK 1391&\u1103\u0313=\u1104<<<\u3138 # HANGUL CHOSEONG SSANGTIKEUT, HANGUL LETTER SSANGTIKEUT 1392&\u1107\u0313=\u1108<<<\u3143 # HANGUL CHOSEONG SSANGPIEUP, HANGUL LETTER SSANGPIEUP 1393&\u1109\u0313=\u110A<<<\u3146 # HANGUL CHOSEONG SSANGSIOS, HANGUL LETTER SSANGSIOS 1394&\u110C\u0313=\u110D<<<\u3149 # HANGUL CHOSEONG SSANGCIEUC, HANGUL LETTER SSANGCIEUC 1395 1396* compare 1397<1 \u1100\u1161 1398= \uAC00 1399<2 \u1100\u1162 1400= \uAC1C 1401<2 \u1100\u1100\u1161 1402= \u1100\uAC00 1403= \u1101\u1161 1404= \uAE4C 1405<3 \u3132\u1161 1406 1407** test: Hangul syllables in prefix & in the interior of a contraction 1408@ rules 1409&x=\u1100\u1161|a\u1102\u1162z 1410* compare 1411<1 \u1100\u1161x 1412= \u1100\u1161a\u1102\u1162z 1413= \u1100\u1161a\uB0B4z 1414= \uAC00a\u1102\u1162z 1415= \uAC00a\uB0B4z 1416 1417** test: digits are unsafe-backwards when numeric=on 1418@ root 1419% numeric=on 1420* compare 1421# If digits are not unsafe, then numeric collation sees "1"=="01" and "b">"a". 1422# We need to back up before the identical prefix "1" and compare the full numbers. 1423<1 11b 1424<1 101a 1425 1426** test: simple locale data test 1427@ locale de 1428* compare 1429<1 a 1430<2 ä 1431<1 ae 1432<2 æ 1433 1434@ locale de-u-co-phonebk 1435* compare 1436<1 a 1437<1 ae 1438<2 ä 1439<2 æ 1440 1441# The following test cases were moved here from ICU 52's DataDrivenCollationTest.txt. 1442 1443** test: DataDrivenCollationTest/TestMorePinyin 1444# Testing the primary strength. 1445@ locale zh 1446% strength=primary 1447* compare 1448< lā 1449= lĀ 1450= Lā 1451= LĀ 1452< lān 1453= lĀn 1454< lē 1455= lĒ 1456= Lē 1457= LĒ 1458< lēn 1459= lĒn 1460 1461** test: DataDrivenCollationTest/TestLithuanian 1462# Lithuanian sort order. 1463@ locale lt 1464* compare 1465< cz 1466< č 1467< d 1468< iz 1469< j 1470< sz 1471< š 1472< t 1473< zz 1474< ž 1475 1476** test: DataDrivenCollationTest/TestLatvian 1477# Latvian sort order. 1478@ locale lv 1479* compare 1480< cz 1481< č 1482< d 1483< gz 1484< ģ 1485< h 1486< iz 1487< j 1488< kz 1489< ķ 1490< l 1491< lz 1492< ļ 1493< m 1494< nz 1495< ņ 1496< o 1497< rz 1498< ŗ 1499< s 1500< sz 1501< š 1502< t 1503< zz 1504< ž 1505 1506** test: DataDrivenCollationTest/TestEstonian 1507# Estonian sort order. 1508@ locale et 1509* compare 1510< sy 1511< š 1512< šy 1513< z 1514< zy 1515< ž 1516< v 1517< va 1518< w 1519< õ 1520< õy 1521< ä 1522< äy 1523< ö 1524< öy 1525< ü 1526< üy 1527< x 1528 1529** test: DataDrivenCollationTest/TestAlbanian 1530# Albanian sort order. 1531@ locale sq 1532* compare 1533< cz 1534< ç 1535< d 1536< dz 1537< dh 1538< e 1539< ez 1540< ë 1541< f 1542< gz 1543< gj 1544< h 1545< lz 1546< ll 1547< m 1548< nz 1549< nj 1550< o 1551< rz 1552< rr 1553< s 1554< sz 1555< sh 1556< t 1557< tz 1558< th 1559< u 1560< xz 1561< xh 1562< y 1563< zz 1564< zh 1565 1566** test: DataDrivenCollationTest/TestSimplifiedChineseOrder 1567# Sorted file has different order. 1568@ root 1569# normalization=on turned on & off automatically. 1570* compare 1571< \u5F20 1572< \u5F20\u4E00\u8E3F 1573 1574** test: DataDrivenCollationTest/TestTibetanNormalizedIterativeCrash 1575# This pretty much crashes. 1576@ root 1577* compare 1578< \u0f71\u0f72\u0f80\u0f71\u0f72 1579< \u0f80 1580 1581** test: DataDrivenCollationTest/TestThaiPartialSortKeyProblems 1582# These are examples of strings that caused trouble in partial sort key testing. 1583@ locale th-TH 1584* compare 1585< \u0E01\u0E01\u0E38\u0E18\u0E20\u0E31\u0E13\u0E11\u0E4C 1586< \u0E01\u0E01\u0E38\u0E2A\u0E31\u0E19\u0E42\u0E18 1587* compare 1588< \u0E01\u0E07\u0E01\u0E32\u0E23 1589< \u0E01\u0E07\u0E42\u0E01\u0E49 1590* compare 1591< \u0E01\u0E23\u0E19\u0E17\u0E32 1592< \u0E01\u0E23\u0E19\u0E19\u0E40\u0E0A\u0E49\u0E32 1593* compare 1594< \u0E01\u0E23\u0E30\u0E40\u0E08\u0E35\u0E22\u0E27 1595< \u0E01\u0E23\u0E30\u0E40\u0E08\u0E35\u0E4A\u0E22\u0E27 1596* compare 1597< \u0E01\u0E23\u0E23\u0E40\u0E0A\u0E2D 1598< \u0E01\u0E23\u0E23\u0E40\u0E0A\u0E49\u0E32 1599 1600** test: DataDrivenCollationTest/TestJavaStyleRule 1601# java.text allows rules to start as '<<<x<<<y...' 1602# we emulate this by assuming a &[first tertiary ignorable] in this case. 1603@ rules 1604&\u0001=equal<<<z<<x<<<w &[first tertiary ignorable]=a &[first primary ignorable]=b 1605* compare 1606= a 1607= equal 1608< z 1609< x 1610= b # x had become the new first primary ignorable 1611< w 1612 1613** test: DataDrivenCollationTest/TestShiftedIgnorable 1614# The UCA states that primary ignorables should be completely 1615# ignorable when following a shifted code point. 1616@ root 1617% alternate=shifted 1618% strength=quaternary 1619* compare 1620< a\u0020b 1621= a\u0020\u0300b 1622= a\u0020\u0301b 1623< a_b 1624= a_\u0300b 1625= a_\u0301b 1626< A\u0020b 1627= A\u0020\u0300b 1628= A\u0020\u0301b 1629< A_b 1630= A_\u0300b 1631= A_\u0301b 1632< a\u0301b 1633< A\u0301b 1634< a\u0300b 1635< A\u0300b 1636 1637** test: DataDrivenCollationTest/TestNShiftedIgnorable 1638# The UCA states that primary ignorables should be completely 1639# ignorable when following a shifted code point. 1640@ root 1641% alternate=non-ignorable 1642% strength=tertiary 1643* compare 1644< a\u0020b 1645< A\u0020b 1646< a\u0020\u0301b 1647< A\u0020\u0301b 1648< a\u0020\u0300b 1649< A\u0020\u0300b 1650< a_b 1651< A_b 1652< a_\u0301b 1653< A_\u0301b 1654< a_\u0300b 1655< A_\u0300b 1656< a\u0301b 1657< A\u0301b 1658< a\u0300b 1659< A\u0300b 1660 1661** test: DataDrivenCollationTest/TestSafeSurrogates 1662# It turned out that surrogates were not skipped properly 1663# when iterating backwards if they were in the middle of a 1664# contraction. This test assures that this is fixed. 1665@ rules 1666&a < x\ud800\udc00b 1667* compare 1668< a 1669< x\ud800\udc00b 1670 1671** test: DataDrivenCollationTest/da_TestPrimary 1672# This test goes through primary strength cases 1673@ locale da 1674% strength=primary 1675* compare 1676< Lvi 1677< Lwi 1678* compare 1679< L\u00e4vi 1680< L\u00f6wi 1681* compare 1682< L\u00fcbeck 1683= Lybeck 1684 1685** test: DataDrivenCollationTest/da_TestTertiary 1686# This test goes through tertiary strength cases 1687@ locale da 1688% strength=tertiary 1689* compare 1690< Luc 1691< luck 1692* compare 1693< luck 1694< L\u00fcbeck 1695* compare 1696< lybeck 1697< L\u00fcbeck 1698* compare 1699< L\u00e4vi 1700< L\u00f6we 1701* compare 1702< L\u00f6ww 1703< mast 1704 1705* compare 1706< A/S 1707< ANDRE 1708< ANDR\u00c9 1709< ANDREAS 1710< AS 1711< CA 1712< \u00c7A 1713< CB 1714< \u00c7C 1715< D.S.B. 1716< DA 1717< \u00d0A 1718< DB 1719< \u00d0C 1720< DSB 1721< DSC 1722< EKSTRA_ARBEJDE 1723< EKSTRABUD0 1724< H\u00d8ST 1725< HAAG 1726< H\u00c5NDBOG 1727< HAANDV\u00c6RKSBANKEN 1728< Karl 1729< karl 1730< NIELS\u0020J\u00d8RGEN 1731< NIELS-J\u00d8RGEN 1732< NIELSEN 1733< R\u00c9E,\u0020A 1734< REE,\u0020B 1735< R\u00c9E,\u0020L 1736< REE,\u0020V 1737< SCHYTT,\u0020B 1738< SCHYTT,\u0020H 1739< SCH\u00dcTT,\u0020H 1740< SCHYTT,\u0020L 1741< SCH\u00dcTT,\u0020M 1742< SS 1743< \u00df 1744< SSA 1745< STORE\u0020VILDMOSE 1746< STOREK\u00c6R0 1747< STORM\u0020PETERSEN 1748< STORMLY 1749< THORVALD 1750< THORVARDUR 1751< \u00feORVAR\u00d0UR 1752< THYGESEN 1753< VESTERG\u00c5RD,\u0020A 1754< VESTERGAARD,\u0020A 1755< VESTERG\u00c5RD,\u0020B 1756< \u00c6BLE 1757< \u00c4BLE 1758< \u00d8BERG 1759< \u00d6BERG 1760 1761* compare 1762< andere 1763< chaque 1764< chemin 1765< cote 1766< cot\u00e9 1767< c\u00f4te 1768< c\u00f4t\u00e9 1769< \u010du\u010d\u0113t 1770< Czech 1771< hi\u0161a 1772< irdisch 1773< lie 1774< lire 1775< llama 1776< l\u00f5ug 1777< l\u00f2za 1778< lu\u010d 1779< luck 1780< L\u00fcbeck 1781< lye 1782< l\u00e4vi 1783< L\u00f6wen 1784< m\u00e0\u0161ta 1785< m\u00eer 1786< myndig 1787< M\u00e4nner 1788< m\u00f6chten 1789< pi\u00f1a 1790< pint 1791< pylon 1792< \u0161\u00e0ran 1793< savoir 1794< \u0160erb\u016bra 1795< Sietla 1796< \u015blub 1797< subtle 1798< symbol 1799< s\u00e4mtlich 1800< verkehrt 1801< vox 1802< v\u00e4ga 1803< waffle 1804< wood 1805< yen 1806< yuan 1807< yucca 1808< \u017eal 1809< \u017eena 1810< \u017den\u0113va 1811< zoo0 1812< Zviedrija 1813< Z\u00fcrich 1814< zysk0 1815< \u00e4ndere 1816 1817** test: DataDrivenCollationTest/hi_TestNewRules 1818# This test goes through new rules and tests against old rules 1819@ locale hi 1820* compare 1821< कॐ 1822< कं 1823< कँ 1824< कः 1825 1826** test: DataDrivenCollationTest/ro_TestNewRules 1827# This test goes through new rules and tests against old rules 1828@ locale ro 1829* compare 1830< xAx 1831< xă 1832< xĂ 1833< Xă 1834< XĂ 1835< xăx 1836< xĂx 1837< xâ 1838< x 1839< Xâ 1840< X 1841< xâx 1842< xÂx 1843< xb 1844< xIx 1845< xî 1846< xÎ 1847< Xî 1848< XÎ 1849< xîx 1850< xÎx 1851< xj 1852< xSx 1853< xș 1854= xş 1855< xȘ 1856= xŞ 1857< Xș 1858= Xş 1859< XȘ 1860= XŞ 1861< xșx 1862= xşx 1863< xȘx 1864= xŞx 1865< xT 1866< xTx 1867< xț 1868= xţ 1869< xȚ 1870= xŢ 1871< Xț 1872= Xţ 1873< XȚ 1874= XŢ 1875< xțx 1876= xţx 1877< xȚx 1878= xŢx 1879< xU 1880 1881** test: DataDrivenCollationTest/testOffsets 1882# This tests cases where forwards and backwards iteration get different offsets 1883@ locale en 1884% strength=tertiary 1885* compare 1886< a\uD800\uDC00\uDC00 1887< b\uD800\uDC00\uDC00 1888* compare 1889< \u0301A\u0301\u0301 1890< \u0301B\u0301\u0301 1891* compare 1892< abcd\r\u0301 1893< abce\r\u0301 1894# TODO: test offsets in new CollationTest 1895 1896# End of test cases moved here from ICU 52's DataDrivenCollationTest.txt. 1897 1898** test: was ICU 52 cmsccoll/TestRedundantRules 1899@ rules 1900& a < b < c < d& [before 1] c < m 1901* compare 1902<1 a 1903<1 b 1904<1 m 1905<1 c 1906<1 d 1907 1908@ rules 1909& a < b <<< c << d <<< e& [before 3] e <<< x 1910* compare 1911<1 a 1912<1 b 1913<3 c 1914<2 d 1915<3 x 1916<3 e 1917 1918@ rules 1919& a < b <<< c << d <<< e <<< f < g& [before 1] g < x 1920* compare 1921<1 a 1922<1 b 1923<3 c 1924<2 d 1925<3 e 1926<3 f 1927<1 x 1928<1 g 1929 1930@ rules 1931& a <<< b << c < d& a < m 1932* compare 1933<1 a 1934<3 b 1935<2 c 1936<1 m 1937<1 d 1938 1939@ rules 1940&a<b<<b\u0301 &z<b 1941* compare 1942<1 a 1943<1 b\u0301 1944<1 z 1945<1 b 1946 1947@ rules 1948&z<m<<<q<<<m 1949* compare 1950<1 z 1951<1 q 1952<3 m 1953 1954@ rules 1955&z<<<m<q<<<m 1956* compare 1957<1 z 1958<1 q 1959<3 m 1960 1961@ rules 1962& a < b < c < d& r < c 1963* compare 1964<1 a 1965<1 b 1966<1 d 1967<1 r 1968<1 c 1969 1970@ rules 1971& a < b < c < d& c < m 1972* compare 1973<1 a 1974<1 b 1975<1 c 1976<1 m 1977<1 d 1978 1979@ rules 1980& a < b < c < d& a < m 1981* compare 1982<1 a 1983<1 m 1984<1 b 1985<1 c 1986<1 d 1987 1988** test: was ICU 52 cmsccoll/TestExpansionSyntax 1989# The following two rules should sort the particular list of strings the same. 1990@ rules 1991&AE <<< a << b <<< c &d <<< f 1992* compare 1993<1 AE 1994<3 a 1995<2 b 1996<3 c 1997<1 d 1998<3 f 1999 2000@ rules 2001&A <<< a / E << b / E <<< c /E &d <<< f 2002* compare 2003<1 AE 2004<3 a 2005<2 b 2006<3 c 2007<1 d 2008<3 f 2009 2010# The following two rules should sort the particular list of strings the same. 2011@ rules 2012&AE <<< a <<< b << c << d < e < f <<< g 2013* compare 2014<1 AE 2015<3 a 2016<3 b 2017<2 c 2018<2 d 2019<1 e 2020<1 f 2021<3 g 2022 2023@ rules 2024&A <<< a / E <<< b / E << c / E << d / E < e < f <<< g 2025* compare 2026<1 AE 2027<3 a 2028<3 b 2029<2 c 2030<2 d 2031<1 e 2032<1 f 2033<3 g 2034 2035# The following two rules should sort the particular list of strings the same. 2036@ rules 2037&AE <<< B <<< C / D <<< F 2038* compare 2039<1 AE 2040<3 B 2041<3 F 2042<1 AED 2043<3 C 2044 2045@ rules 2046&A <<< B / E <<< C / ED <<< F / E 2047* compare 2048<1 AE 2049<3 B 2050<3 F 2051<1 AED 2052<3 C 2053 2054** test: never reorder trailing primaries 2055@ root 2056% reorder Zzzz Grek 2057* compare 2058<1 L 2059<1 字 2060<1 Ω 2061<1 \uFFFD 2062<1 \uFFFF 2063 2064** test: fall back to mappings with shorter prefixes, not immediately to ones with no prefixes 2065@ rules 2066&u=ab|cd 2067&v=b|ce 2068* compare 2069<1 abc 2070<1 abcc 2071<1 abcf 2072<1 abcd 2073= abu 2074<1 abce 2075= abv 2076 2077# With the following rules, there is only one prefix per composite ĉ or ç, 2078# but both prefixes apply to just c in NFD form. 2079# We would get different results for composed vs. NFD input 2080# if we fell back directly from longest-prefix mappings to no-prefix mappings. 2081@ rules 2082&x=op|ĉ 2083&y=p|ç 2084* compare 2085<1 opc 2086<2 opć 2087<1 opcz 2088<1 opd 2089<1 opĉ 2090= opc\u0302 2091= opx 2092<1 opç 2093= opc\u0327 2094= opy 2095 2096# The mapping is used which has the longest matching prefix for which 2097# there is also a suffix match, with the longest suffix match among several for that prefix. 2098@ rules 2099&❶=d 2100&❷=de 2101&❸=def 2102&①=c|d 2103&②=c|de 2104&③=c|def 2105&④=bc|d 2106&⑤=bc|de 2107&⑥=bc|def 2108&⑦=abc|d 2109&⑧=abc|de 2110&⑨=abc|def 2111* compare 2112<1 9aadzz 2113= 9aa❶zz 2114<1 9aadez 2115= 9aa❷z 2116<1 9aadef 2117= 9aa❸ 2118<1 9acdzz 2119= 9ac①zz 2120<1 9acdez 2121= 9ac②z 2122<1 9acdef 2123= 9ac③ 2124<1 9bcdzz 2125= 9bc④zz 2126<1 9bcdez 2127= 9bc⑤z 2128<1 9bcdef 2129= 9bc⑥ 2130<1 abcdzz 2131= abc⑦zz 2132<1 abcdez 2133= abc⑧z 2134<1 abcdef 2135= abc⑨ 2136 2137** test: prefix + discontiguous contraction with missing prefix contraction 2138# Unfortunate terminology: The first "prefix" here is the pre-context, 2139# the second "prefix" refers to the contraction/relation string that is 2140# one shorter than the one being tested. 2141@ rules 2142&x=p|e 2143&y=p|ê 2144&z=op|ê 2145# No mapping for op|e: 2146# Discontiguous contraction matching should not match op|ê in opệ 2147# because it would have to skip the dot below and extend a match on op|e by the circumflex, 2148# but there is no match on op|e. 2149* compare 2150<1 oPe 2151<1 ope 2152= opx 2153<1 opệ 2154= opy\u0323 # y not z 2155<1 opê 2156= opz 2157 2158# We cannot test for fallback by whether the contraction default CE32 2159# is for another contraction. With the following rules, there is no mapping for op|e, 2160# and the fallback to prefix p has no contractions. 2161@ rules 2162&x=p|e 2163&z=op|ê 2164* compare 2165<1 oPe 2166<1 ope 2167= opx 2168<2 opệ 2169= opx\u0323\u0302 # x not z 2170<1 opê 2171= opz 2172 2173# One more variation: Fallback to the simple code point, no shorter non-empty prefix. 2174@ rules 2175&x=e 2176&z=op|ê 2177* compare 2178<1 ope 2179= opx 2180<3 oPe 2181= oPx 2182<2 opệ 2183= opx\u0323\u0302 # x not z 2184<1 opê 2185= opz 2186 2187** test: maxVariable via rules 2188@ rules 2189[maxVariable space][alternate shifted] 2190* compare 2191= \u0020 2192= \u000A 2193<1 . 2194<1 ° # degree sign 2195<1 $ 2196<1 0 2197 2198** test: maxVariable via setting 2199@ root 2200% maxVariable=currency 2201% alternate=shifted 2202* compare 2203= \u0020 2204= \u000A 2205= . 2206= ° # degree sign 2207= $ 2208<1 0 2209 2210** test: ICU4J CollationMiscTest/TestContractionClosure (ää) 2211# This tests canonical closure, but it also tests that CollationFastLatin 2212# bails out properly for contractions with combining marks. 2213# For that we need pairs of strings that remain in the Latin fastpath 2214# long enough, hence the extra "= b" lines. 2215@ rules 2216&b=\u00e4\u00e4 2217* compare 2218<1 b 2219= \u00e4\u00e4 2220= b 2221= a\u0308a\u0308 2222= b 2223= \u00e4a\u0308 2224= b 2225= a\u0308\u00e4 2226 2227** test: ICU4J CollationMiscTest/TestContractionClosure (Å) 2228@ rules 2229&b=\u00C5 2230* compare 2231<1 b 2232= \u00C5 2233= b 2234= A\u030A 2235= b 2236= \u212B 2237 2238** test: reset-before on already-tailored characters, ICU ticket 10108 2239@ rules 2240&a<w<<x &[before 2]x<<y 2241* compare 2242<1 a 2243<1 w 2244<2 y 2245<2 x 2246 2247@ rules 2248&a<<w<<<x &[before 2]x<<y 2249* compare 2250<1 a 2251<2 y 2252<2 w 2253<3 x 2254 2255@ rules 2256&a<w<x &[before 2]x<<y 2257* compare 2258<1 a 2259<1 w 2260<1 y 2261<2 x 2262 2263@ rules 2264&a<w<<<x &[before 2]x<<y 2265* compare 2266<1 a 2267<1 y 2268<2 w 2269<3 x 2270 2271** test: numeric collation with other settings, ICU ticket 9092 2272@ root 2273% strength=identical 2274% caseFirst=upper 2275% numeric=on 2276* compare 2277<1 100\u0020a 2278<1 101 2279 2280** test: collation type fallback from unsupported type, ICU ticket 10149 2281@ locale fr-CA-u-co-phonebk 2282# Expect the same result as with fr-CA, using backwards-secondary order. 2283# That is, we should fall back from the unsupported collation type 2284# to the locale's default collation type. 2285* compare 2286<1 cote 2287<2 côte 2288<2 coté 2289<2 côté 2290 2291** test: @ is equivalent to [backwards 2], ICU ticket 9956 2292@ rules 2293&b<a @ &v<<w 2294* compare 2295<1 b 2296<1 a 2297<1 cote 2298<2 côte 2299<2 coté 2300<2 côté 2301<1 v 2302<2 w 2303<1 x 2304 2305** test: shifted+reordering, ICU ticket 9507 2306@ root 2307% reorder Grek punct space 2308% alternate=shifted 2309% strength=quaternary 2310# Which primaries are "variable" should be determined without script reordering, 2311# and then primaries should be reordered whether they are shifted to quaternary or not. 2312* compare 2313<4 ( # punctuation 2314<4 ) 2315<4 \u0020 # space 2316<1 ` # symbol 2317<1 ^ 2318<1 $ # currency symbol 2319<1 € 2320<1 0 # numbers 2321<1 ε # Greek 2322<1 e # Latin 2323<1 e(e 2324<4 e)e 2325<4 e\u0020e 2326<4 ee 2327<3 e(E 2328<4 e)E 2329<4 e\u0020E 2330<4 eE 2331 2332** test: "uppercase first" could sort a string before its prefix, ICU ticket 9351 2333@ rules 2334&\u0001<<<b<<<B 2335% caseFirst=upper 2336* compare 2337<1 aaa 2338<3 aaaB 2339 2340** test: secondary+case ignores secondary ignorables, ICU ticket 9355 2341@ rules 2342&\u0001<<<b<<<B 2343% strength=secondary 2344% caseLevel=on 2345* compare 2346<1 a 2347= ab 2348= aB 2349 2350** test: custom collation rules involving tail of a contraction in Malayalam, ICU ticket 6328 2351@ rules 2352&[before 2] ൌ << ൗ # U+0D57 << U+0D4C == 0D46+0D57 2353* compare 2354<1 ൗx 2355<2 ൌx 2356<1 ൗy 2357<2 ൌy 2358 2359** test: quoted apostrophe in compact syntax, ICU ticket 8204 2360@ rules 2361&q<<*a''c 2362* compare 2363<1 d 2364<1 p 2365<1 q 2366<2 a 2367<2 \u0027 2368<2 c 2369<1 r 2370 2371# ICU ticket #8260 "Support all collation-related keywords in Collator.getInstance()" 2372** test: locale -u- with collation keywords, ICU ticket 8260 2373@ locale de-u-kv-sPace-ka-shifTed-kn-kk-falsE-kf-Upper-kc-tRue-ks-leVel4 2374* compare 2375<4 \u0020 # space is shifted, strength=quaternary 2376<1 ! # punctuation is regular 2377<1 2 2378<1 12 # numeric sorting 2379<1 B 2380<c b # uppercase first on case level 2381<1 x\u0301\u0308 2382<2 x\u0308\u0301 # normalization off 2383 2384** test: locale @ with collation keywords, ICU ticket 8260 2385@ locale fr@colbAckwards=yes;ColStrength=Quaternary;kv=currencY;colalternate=shifted 2386* compare 2387<4 $ # currency symbols are shifted, strength=quaternary 2388<1 àla 2389<2 alà # backwards secondary level 2390 2391** test: locale -u- with script reordering, ICU ticket 8260 2392@ locale el-u-kr-kana-SYMBOL-Grek-hani-cyrl-latn-digit-armn-deva-ethi-thai 2393* compare 2394<1 \u0020 2395<1 あ 2396<1 ☂ 2397<1 Ω 2398<1 丂 2399<1 ж 2400<1 L 2401<1 4 2402<1 Ձ 2403<1 अ 2404<1 ሄ 2405<1 ฉ 2406 2407** test: locale @collation=type should be case-insensitive 2408@ locale de@coLLation=PhoneBook 2409* compare 2410<1 ae 2411<2 ä 2412<3 Ä 2413 2414** test: import root search rules plus German phonebook rules, ICU ticket 8962 2415@ locale de-u-co-search 2416* compare 2417<1 = 2418<1 ≠ 2419<1 a 2420<1 ae 2421<2 ä 2422 2423# Once more, but with runtime builder. 2424@ rules 2425[import und-u-co-search][import de-u-co-phonebk] 2426* compare 2427<1 = 2428<1 ≠ 2429<1 a 2430<1 ae 2431<2 ä 2432 2433# Once again, with import from "root" not "und" (as in a proper language tag). 2434@ rules 2435[import root-u-co-search][import de-u-co-phonebk] 2436* compare 2437<1 = 2438<1 ≠ 2439<1 a 2440<1 ae 2441<2 ä 2442 2443** test: import rules from a language with non-Latin native script, and reset the reordering, ICU ticket 10998 2444# Greek should sort Greek first. 2445@ rules 2446[import el] 2447* compare 2448<1 4 2449<1 Ω 2450<1 L 2451 2452# Import Greek, and then reset the reordering. 2453@ rules 2454[import el][reorder Zzzz] 2455* compare 2456<1 4 2457<1 L 2458<1 Ω 2459 2460# "others" is a synonym for Zzzz. 2461@ rules 2462[import el][reorder others] 2463* compare 2464<1 4 2465<1 L 2466<1 Ω 2467 2468** test: regression test for CollationFastLatinBuilder, ICU ticket 11388 2469@ rules 2470&x<<aa<<<Aa<<<AA 2471% strength=secondary 2472* compare 2473<1 AA 2474<2 Aẩ 2475<2 aą 2476* compare 2477<1 AA 2478<2 aą 2479 2480** test: tailor tertiary-after a common tertiary where there is a lower one 2481# Assume that Hiragana small A has a below-common tertiary, and Hiragana A has a common one. 2482# See ICU ticket 11448 & CLDR ticket 7222. 2483@ rules 2484&あ<<<x<<<y<<<z 2485* compare 2486<1 ぁ 2487<3 あ 2488<3 x 2489<3 y 2490<3 z 2491<3 ァ 2492<1 い 2493 2494** test: tailor tertiary-after a below-common tertiary 2495@ rules 2496&ぁ<<<x<<<y<<<z 2497* compare 2498<1 ぁ 2499<3 x 2500<3 y 2501<3 z 2502<3 あ 2503<3 ァ 2504<1 い 2505 2506** test: tailor tertiary-before a common tertiary where there is a lower one 2507@ rules 2508&[before 3]あ<<<x<<<y<<<z 2509* compare 2510<1 ぁ 2511<3 x 2512<3 y 2513<3 z 2514<3 あ 2515<3 ァ 2516<1 い 2517 2518** test: tailor tertiary-before a below-common tertiary 2519@ rules 2520&[before 3]ぁ<<<x<<<y<<<z 2521* compare 2522<1 x 2523<3 y 2524<3 z 2525<3 ぁ 2526<3 あ 2527<3 ァ 2528<1 い 2529 2530** test: reorder single scripts not groups, ICU ticket 11449 2531@ root 2532% reorder Goth Latn 2533* compare 2534<1 4 2535<1 �� # Gothic 2536<1 L 2537<1 Ω 2538# Before ICU 55, the following reordered together with Gothic. 2539<1 �� # Old Italic 2540<1 �� # Shavian 2541