Update ICU to 54.1 step 1

source/test/testdata/regextst.txt

-# Copyright (c) 2001-2013 International Business Machines
+# Copyright (c) 2001-2014 International Business Machines
 # Corporation and others. All Rights Reserved.
 #
 #  file:
 #
 #   ICU regular expression test cases.
 #
 #   format:   one test case per line,
 #               <test case>    =  <pattern>   <flags>  <match string>  [# comment]
 #               <pattern>      =  "<regular expression pattern>"
 #               <match string> =  "<tagged string>"
@@ skipping 500 matching lines @@
 "ab(?:(c)|(d))\1"              "<0>ab<1>c</1>c</0>e"
 "ab(?:(c)|(d))\1"            i "abde"
 "ab(?:(c)|(d))\1"            i "<0>ab<1>c</1>c</0>e"
 
 # Case Insensitive
 "aBc"                    i      "<0>ABC</0>"
 "a[^bc]d"                i      "ABD"
 '((((((((((a))))))))))\10' i    "<0><1><2><3><4><5><6><7><8><9><10>A</10></9></8></7></6></5></4></3></2></1>A</0>"
 
 "(?:(?i)a)b"                    "<0>Ab</0>"
-"ab(?i)cd"                       "<0>abCd</0>"
+"ab(?i)cd"                          "<0>abCd</0>"
 "ab$cd"                         "abcd"
 
+"ssl"                      i    "abc<0>ßl</0>xyz"
+"ssl"                      i    "abc<0>ẞl</0>xyz"
+"FIND"                     i    "can <0>find</0> ?"  # fi ligature, \ufb01
+"find"                      i    "can <0>FIND</0> ?"
+"ῧ"                        i    "xxx<0>ῧ</0>xxx"    # Composed char (match string) decomposes when case-folded (pattern)
+
 # White space handling
 "a b"                           "ab"
 "abc "                          "abc"
 "abc "                          "<0>abc </0>"
 "ab[cd e]z"                     "<0>ab z</0>"
 "ab\ c"                         "<0>ab c</0> "
 "ab c"                          "<0>ab c</0> "
 "ab c"                        x "ab c "
 "ab\ c"                       x "<0>ab c</0> "
 
@@ skipping 630 matching lines @@
 
 "(?<=a{1,5})bc"                   "aaaa<0>bc</0>def"
 "(?<=(?:aa){3,20})bc"             "aaaaaa<0>bc</0>def"
 "(?<!abc {1,100}|def {1,100}|ghi {1,100})jkl"      "def jkl"
 "(?<!abc {1,100}|def {1,100}|ghi {1,100})jkl"      "rst <0>jkl</0>"
 "(?<=a{11})bc"                   "aaaaaaaaaaa<0>bc</0>"
 "(?<=a{11})bc"                   "aaaaaaaaaabc"
 "(?<=a{1,})bc"           E       "aaaa<0>bc</0>def"   # U_REGEX_LOOK_BEHIND_LIMIT error.
 "(?<=(?:){11})bc"                "<0>bc</0>"          # Empty (?:) expression.
 
+# Bug 10835
+#   Match Start Set not being correctly computed for case insensitive patterns.
+#   (Test here is to dump the compiled pattern & manually check the start set.)
 
-# Bug 11369
-#   Incorrect optimization of patterns with a zero length quantifier {0}
+"(private|secret|confidential|classified|restricted)"  i   "hmm, <0><1>Classified</1></0> stuff"
+"(private|secret|confidential|classified|restricted)"      "hmm, Classified stuff"
 
-"(.|b)(|b){0}\$(?#xxx){3}(?>\D*)"   "AAAAABBBBBCCCCCDDDDEEEEE"
-"(|b)ab(c)"                     "<0><1></1>ab<2>c</2></0>"
-"(|b){0}a{3}(D*)"               "<0>aaa<2></2></0>"
-"(|b){0,1}a{3}(D*)"             "<0><1></1>aaa<2></2></0>"
-"((|b){0})a{3}(D*)"             "<0><1></1>aaa<3></3></0>"
+# Bug 10844
 
-# Bug 11370
-#   Max match length computation of look-behind expression gives result that is too big to fit in the
-#   in the 24 bit operand portion of the compiled code. Expressions should fail to compile
-#   (Look-behind match length must be bounded. This case is treated as unbounded, an error.)
+"^([\w\d:]+)$"                  "<0><1>DiesIst1Beispiel:text</1></0>"
+"^([\w\d:]+)$"           i      "<0><1>DiesIst1Beispiel:text</1></0>"
+"^(\w+\d\w+:\w+)$"              "<0><1>DiesIst1Beispiel:text</1></0>"
+"^(\w+\d\w+:\w+)$"       i      "<0><1>DiesIst1Beispiel:text</1></0>"
 
-"(?<!(0123456789a){10000000})x"         E  "no match"
-"(?<!\\ubeaf(\\ubeaf{11000}){11000})"   E  "no match"
+# Bug 11049
+#   Edge cases in find() when pattern match begins with set of code points
+#   and the match begins at the end of the string.
 
+"A|B|C"                         "hello <0>A</0>"
+"A|B|C"                         "hello \U00011234"
+"A|B|\U00012345"                "hello <0>\U00012345</0>"
+"A|B|\U00010000"                "hello \ud800"
 
 #  Random debugging, Temporary
 #
-#"^(?:a?b?)*$"                    "a--"
 
 "This is a string with (?:one |two |three )endings"   "<0>This is a string with two endings</0>"
-"((?:a|b|c)whoop-dee-do) | [jkl]|zed"             "x"
-"astring|another[bcd]|alpha|a|[a]"    "x"
 
 
 #
 #  Regexps from http://www.regexlib.com
 #
 "^[a-zA-Z]{1,2}[0-9][0-9A-Za-z]{0,1} {0,1}[0-9][A-Za-z]{2}$"   G "<0>G1 1AA</0>"
 "^[a-zA-Z]{1,2}[0-9][0-9A-Za-z]{0,1} {0,1}[0-9][A-Za-z]{2}$"   G "<0>EH10 2QQ</0>"
 "^[a-zA-Z]{1,2}[0-9][0-9A-Za-z]{0,1} {0,1}[0-9][A-Za-z]{2}$"   G "<0>SW1 1ZZ</0>"
 "^[a-zA-Z]{1,2}[0-9][0-9A-Za-z]{0,1} {0,1}[0-9][A-Za-z]{2}$"     "G111 1AA"
 "^[a-zA-Z]{1,2}[0-9][0-9A-Za-z]{0,1} {0,1}[0-9][A-Za-z]{2}$"     "X10 WW"
@@ skipping 1267 matching lines @@
 "(\w+)\s+\1"     "may day"
 "(\w+)\s+\1"     "gogo"
 "(\w+)\s+\1"     "1212"
 "^[a-zA-Z0-9\-\.]+\.(com|org|net|mil|edu|COM|ORG|NET|MIL|EDU)$"   G "<0>3SquareBand.com</0>"
 "^[a-zA-Z0-9\-\.]+\.(com|org|net|mil|edu|COM|ORG|NET|MIL|EDU)$"   G "<0>asp.net</0>"
 "^[a-zA-Z0-9\-\.]+\.(com|org|net|mil|edu|COM|ORG|NET|MIL|EDU)$"   G "<0>army.mil</0>"
 "^[a-zA-Z0-9\-\.]+\.(com|org|net|mil|edu|COM|ORG|NET|MIL|EDU)$"     "$SquareBand.com"
 "^[a-zA-Z0-9\-\.]+\.(com|org|net|mil|edu|COM|ORG|NET|MIL|EDU)$"     "asp/dot.net"
 "^[a-zA-Z0-9\-\.]+\.(com|org|net|mil|edu|COM|ORG|NET|MIL|EDU)$"     "army.military"