1NOTE implicit vs. explicit repetitions : 2009-02-02 2 3# Glenn Fowler <gsf@research.att.com> 4# conforming matches (column 4) must match one of the following BREs 5# NOMATCH 6# (0,.)\((\(.\),\(.\))(?,?)(\2,\3)\)* 7# (0,.)\((\(.\),\(.\))(\2,\3)(?,?)\)* 8# i.e., each 3-tuple has two identical elements and one (?,?) 9 10E ((..)|(.)) NULL NOMATCH 11E ((..)|(.))((..)|(.)) NULL NOMATCH 12E ((..)|(.))((..)|(.))((..)|(.)) NULL NOMATCH 13 14E ((..)|(.)){1} NULL NOMATCH 15E ((..)|(.)){2} NULL NOMATCH 16E ((..)|(.)){3} NULL NOMATCH 17 18E ((..)|(.))* NULL (0,0) 19 20E ((..)|(.)) a (0,1)(0,1)(?,?)(0,1) 21E ((..)|(.))((..)|(.)) a NOMATCH 22E ((..)|(.))((..)|(.))((..)|(.)) a NOMATCH 23 24E ((..)|(.)){1} a (0,1)(0,1)(?,?)(0,1) 25E ((..)|(.)){2} a NOMATCH 26E ((..)|(.)){3} a NOMATCH 27 28E ((..)|(.))* a (0,1)(0,1)(?,?)(0,1) 29 30E ((..)|(.)) aa (0,2)(0,2)(0,2)(?,?) 31E ((..)|(.))((..)|(.)) aa (0,2)(0,1)(?,?)(0,1)(1,2)(?,?)(1,2) 32E ((..)|(.))((..)|(.))((..)|(.)) aa NOMATCH 33 34E ((..)|(.)){1} aa (0,2)(0,2)(0,2)(?,?) 35E ((..)|(.)){2} aa (0,2)(1,2)(?,?)(1,2) 36E ((..)|(.)){3} aa NOMATCH 37 38E ((..)|(.))* aa (0,2)(0,2)(0,2)(?,?) 39 40E ((..)|(.)) aaa (0,2)(0,2)(0,2)(?,?) 41E ((..)|(.))((..)|(.)) aaa (0,3)(0,2)(0,2)(?,?)(2,3)(?,?)(2,3) 42E ((..)|(.))((..)|(.))((..)|(.)) aaa (0,3)(0,1)(?,?)(0,1)(1,2)(?,?)(1,2)(2,3)(?,?)(2,3) 43 44E ((..)|(.)){1} aaa (0,2)(0,2)(0,2)(?,?) 45#E ((..)|(.)){2} aaa (0,3)(2,3)(?,?)(2,3) 46E ((..)|(.)){2} aaa (0,3)(2,3)(0,2)(2,3) RE2/Go 47E ((..)|(.)){3} aaa (0,3)(2,3)(?,?)(2,3) 48 49#E ((..)|(.))* aaa (0,3)(2,3)(?,?)(2,3) 50E ((..)|(.))* aaa (0,3)(2,3)(0,2)(2,3) RE2/Go 51 52E ((..)|(.)) aaaa (0,2)(0,2)(0,2)(?,?) 53E ((..)|(.))((..)|(.)) aaaa (0,4)(0,2)(0,2)(?,?)(2,4)(2,4)(?,?) 54E ((..)|(.))((..)|(.))((..)|(.)) aaaa (0,4)(0,2)(0,2)(?,?)(2,3)(?,?)(2,3)(3,4)(?,?)(3,4) 55 56E ((..)|(.)){1} aaaa (0,2)(0,2)(0,2)(?,?) 57E ((..)|(.)){2} aaaa (0,4)(2,4)(2,4)(?,?) 58#E ((..)|(.)){3} aaaa (0,4)(3,4)(?,?)(3,4) 59E ((..)|(.)){3} aaaa (0,4)(3,4)(0,2)(3,4) RE2/Go 60 61E ((..)|(.))* aaaa (0,4)(2,4)(2,4)(?,?) 62 63E ((..)|(.)) aaaaa (0,2)(0,2)(0,2)(?,?) 64E ((..)|(.))((..)|(.)) aaaaa (0,4)(0,2)(0,2)(?,?)(2,4)(2,4)(?,?) 65E ((..)|(.))((..)|(.))((..)|(.)) aaaaa (0,5)(0,2)(0,2)(?,?)(2,4)(2,4)(?,?)(4,5)(?,?)(4,5) 66 67E ((..)|(.)){1} aaaaa (0,2)(0,2)(0,2)(?,?) 68E ((..)|(.)){2} aaaaa (0,4)(2,4)(2,4)(?,?) 69#E ((..)|(.)){3} aaaaa (0,5)(4,5)(?,?)(4,5) 70E ((..)|(.)){3} aaaaa (0,5)(4,5)(2,4)(4,5) RE2/Go 71 72#E ((..)|(.))* aaaaa (0,5)(4,5)(?,?)(4,5) 73E ((..)|(.))* aaaaa (0,5)(4,5)(2,4)(4,5) RE2/Go 74 75E ((..)|(.)) aaaaaa (0,2)(0,2)(0,2)(?,?) 76E ((..)|(.))((..)|(.)) aaaaaa (0,4)(0,2)(0,2)(?,?)(2,4)(2,4)(?,?) 77E ((..)|(.))((..)|(.))((..)|(.)) aaaaaa (0,6)(0,2)(0,2)(?,?)(2,4)(2,4)(?,?)(4,6)(4,6)(?,?) 78 79E ((..)|(.)){1} aaaaaa (0,2)(0,2)(0,2)(?,?) 80E ((..)|(.)){2} aaaaaa (0,4)(2,4)(2,4)(?,?) 81E ((..)|(.)){3} aaaaaa (0,6)(4,6)(4,6)(?,?) 82 83E ((..)|(.))* aaaaaa (0,6)(4,6)(4,6)(?,?) 84 85NOTE additional repetition tests graciously provided by Chris Kuklewicz www.haskell.org 2009-02-02 86 87# These test a bug in OS X / FreeBSD / NetBSD, and libtree. 88# Linux/GLIBC gets the {8,} and {8,8} wrong. 89 90:HA#100:E X(.?){0,}Y X1234567Y (0,9)(7,8) 91:HA#101:E X(.?){1,}Y X1234567Y (0,9)(7,8) 92:HA#102:E X(.?){2,}Y X1234567Y (0,9)(7,8) 93:HA#103:E X(.?){3,}Y X1234567Y (0,9)(7,8) 94:HA#104:E X(.?){4,}Y X1234567Y (0,9)(7,8) 95:HA#105:E X(.?){5,}Y X1234567Y (0,9)(7,8) 96:HA#106:E X(.?){6,}Y X1234567Y (0,9)(7,8) 97:HA#107:E X(.?){7,}Y X1234567Y (0,9)(7,8) 98:HA#108:E X(.?){8,}Y X1234567Y (0,9)(8,8) 99#:HA#110:E X(.?){0,8}Y X1234567Y (0,9)(7,8) 100:HA#110:E X(.?){0,8}Y X1234567Y (0,9)(8,8) RE2/Go 101#:HA#111:E X(.?){1,8}Y X1234567Y (0,9)(7,8) 102:HA#111:E X(.?){1,8}Y X1234567Y (0,9)(8,8) RE2/Go 103#:HA#112:E X(.?){2,8}Y X1234567Y (0,9)(7,8) 104:HA#112:E X(.?){2,8}Y X1234567Y (0,9)(8,8) RE2/Go 105#:HA#113:E X(.?){3,8}Y X1234567Y (0,9)(7,8) 106:HA#113:E X(.?){3,8}Y X1234567Y (0,9)(8,8) RE2/Go 107#:HA#114:E X(.?){4,8}Y X1234567Y (0,9)(7,8) 108:HA#114:E X(.?){4,8}Y X1234567Y (0,9)(8,8) RE2/Go 109#:HA#115:E X(.?){5,8}Y X1234567Y (0,9)(7,8) 110:HA#115:E X(.?){5,8}Y X1234567Y (0,9)(8,8) RE2/Go 111#:HA#116:E X(.?){6,8}Y X1234567Y (0,9)(7,8) 112:HA#116:E X(.?){6,8}Y X1234567Y (0,9)(8,8) RE2/Go 113#:HA#117:E X(.?){7,8}Y X1234567Y (0,9)(7,8) 114:HA#117:E X(.?){7,8}Y X1234567Y (0,9)(8,8) RE2/Go 115:HA#118:E X(.?){8,8}Y X1234567Y (0,9)(8,8) 116 117# These test a fixed bug in my regex-tdfa that did not keep the expanded 118# form properly grouped, so right association did the wrong thing with 119# these ambiguous patterns (crafted just to test my code when I became 120# suspicious of my implementation). The first subexpression should use 121# "ab" then "a" then "bcd". 122 123# OS X / FreeBSD / NetBSD badly fail many of these, with impossible 124# results like (0,6)(4,5)(6,6). 125 126:HA#260:E (a|ab|c|bcd){0,}(d*) ababcd (0,1)(0,1)(1,1) 127:HA#261:E (a|ab|c|bcd){1,}(d*) ababcd (0,1)(0,1)(1,1) 128:HA#262:E (a|ab|c|bcd){2,}(d*) ababcd (0,6)(3,6)(6,6) 129:HA#263:E (a|ab|c|bcd){3,}(d*) ababcd (0,6)(3,6)(6,6) 130:HA#264:E (a|ab|c|bcd){4,}(d*) ababcd NOMATCH 131:HA#265:E (a|ab|c|bcd){0,10}(d*) ababcd (0,1)(0,1)(1,1) 132:HA#266:E (a|ab|c|bcd){1,10}(d*) ababcd (0,1)(0,1)(1,1) 133:HA#267:E (a|ab|c|bcd){2,10}(d*) ababcd (0,6)(3,6)(6,6) 134:HA#268:E (a|ab|c|bcd){3,10}(d*) ababcd (0,6)(3,6)(6,6) 135:HA#269:E (a|ab|c|bcd){4,10}(d*) ababcd NOMATCH 136:HA#270:E (a|ab|c|bcd)*(d*) ababcd (0,1)(0,1)(1,1) 137:HA#271:E (a|ab|c|bcd)+(d*) ababcd (0,1)(0,1)(1,1) 138 139# The above worked on Linux/GLIBC but the following often fail. 140# They also trip up OS X / FreeBSD / NetBSD: 141 142#:HA#280:E (ab|a|c|bcd){0,}(d*) ababcd (0,6)(3,6)(6,6) 143:HA#280:E (ab|a|c|bcd){0,}(d*) ababcd (0,6)(4,5)(5,6) RE2/Go 144#:HA#281:E (ab|a|c|bcd){1,}(d*) ababcd (0,6)(3,6)(6,6) 145:HA#281:E (ab|a|c|bcd){1,}(d*) ababcd (0,6)(4,5)(5,6) RE2/Go 146#:HA#282:E (ab|a|c|bcd){2,}(d*) ababcd (0,6)(3,6)(6,6) 147:HA#282:E (ab|a|c|bcd){2,}(d*) ababcd (0,6)(4,5)(5,6) RE2/Go 148#:HA#283:E (ab|a|c|bcd){3,}(d*) ababcd (0,6)(3,6)(6,6) 149:HA#283:E (ab|a|c|bcd){3,}(d*) ababcd (0,6)(4,5)(5,6) RE2/Go 150:HA#284:E (ab|a|c|bcd){4,}(d*) ababcd NOMATCH 151#:HA#285:E (ab|a|c|bcd){0,10}(d*) ababcd (0,6)(3,6)(6,6) 152:HA#285:E (ab|a|c|bcd){0,10}(d*) ababcd (0,6)(4,5)(5,6) RE2/Go 153#:HA#286:E (ab|a|c|bcd){1,10}(d*) ababcd (0,6)(3,6)(6,6) 154:HA#286:E (ab|a|c|bcd){1,10}(d*) ababcd (0,6)(4,5)(5,6) RE2/Go 155#:HA#287:E (ab|a|c|bcd){2,10}(d*) ababcd (0,6)(3,6)(6,6) 156:HA#287:E (ab|a|c|bcd){2,10}(d*) ababcd (0,6)(4,5)(5,6) RE2/Go 157#:HA#288:E (ab|a|c|bcd){3,10}(d*) ababcd (0,6)(3,6)(6,6) 158:HA#288:E (ab|a|c|bcd){3,10}(d*) ababcd (0,6)(4,5)(5,6) RE2/Go 159:HA#289:E (ab|a|c|bcd){4,10}(d*) ababcd NOMATCH 160#:HA#290:E (ab|a|c|bcd)*(d*) ababcd (0,6)(3,6)(6,6) 161:HA#290:E (ab|a|c|bcd)*(d*) ababcd (0,6)(4,5)(5,6) RE2/Go 162#:HA#291:E (ab|a|c|bcd)+(d*) ababcd (0,6)(3,6)(6,6) 163:HA#291:E (ab|a|c|bcd)+(d*) ababcd (0,6)(4,5)(5,6) RE2/Go 164