1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/js/src/tests/ecma_3/RegExp/regress-85721.js Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,243 @@
1.4 +// |reftest| random -- bogus perf test (bug 467263)
1.5 +/* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
1.6 +/* This Source Code Form is subject to the terms of the Mozilla Public
1.7 + * License, v. 2.0. If a copy of the MPL was not distributed with this
1.8 + * file, You can obtain one at http://mozilla.org/MPL/2.0/. */
1.9 +
1.10 +/*
1.11 + *
1.12 + * Date: 14 Feb 2002
1.13 + * SUMMARY: Performance: Regexp performance degraded from 4.7
1.14 + * See http://bugzilla.mozilla.org/show_bug.cgi?id=85721
1.15 + *
1.16 + * Adjust this testcase if necessary. The FAST constant defines
1.17 + * an upper bound in milliseconds for any execution to take.
1.18 + *
1.19 + */
1.20 +//-----------------------------------------------------------------------------
1.21 +var BUGNUMBER = 85721;
1.22 +var summary = 'Performance: execution of regular expression';
1.23 +var FAST = 100; // execution should be 100 ms or less to pass the test
1.24 +var MSG_FAST = 'Execution took less than ' + FAST + ' ms';
1.25 +var MSG_SLOW = 'Execution took ';
1.26 +var MSG_MS = ' ms';
1.27 +var str = '';
1.28 +var re = '';
1.29 +var status = '';
1.30 +var actual = '';
1.31 +var expect= '';
1.32 +
1.33 +printBugNumber(BUGNUMBER);
1.34 +printStatus (summary);
1.35 +
1.36 +
1.37 +function elapsedTime(startTime)
1.38 +{
1.39 + return new Date() - startTime;
1.40 +}
1.41 +
1.42 +
1.43 +function isThisFast(ms)
1.44 +{
1.45 + if (ms <= FAST)
1.46 + return MSG_FAST;
1.47 + return MSG_SLOW + ms + MSG_MS;
1.48 +}
1.49 +
1.50 +
1.51 +
1.52 +/*
1.53 + * The first regexp. We'll test for performance (Section 1) and accuracy (Section 2).
1.54 + */
1.55 +str='<sql:connection id="conn1"> <sql:url>www.m.com</sql:url> <sql:driver>drive.class</sql:driver>\n<sql:userId>foo</sql:userId> <sql:password>goo</sql:password> </sql:connection>';
1.56 +re = /<sql:connection id="([^\r\n]*?)">\s*<sql:url>\s*([^\r\n]*?)\s*<\/sql:url>\s*<sql:driver>\s*([^\r\n]*?)\s*<\/sql:driver>\s*(\s*<sql:userId>\s*([^\r\n]*?)\s*<\/sql:userId>\s*)?\s*(\s*<sql:password>\s*([^\r\n]*?)\s*<\/sql:password>\s*)?\s*<\/sql:connection>/;
1.57 +expect = Array("<sql:connection id=\"conn1\"> <sql:url>www.m.com</sql:url> <sql:driver>drive.class</sql:driver>\n<sql:userId>foo</sql:userId> <sql:password>goo</sql:password> </sql:connection>","conn1","www.m.com","drive.class","<sql:userId>foo</sql:userId> ","foo","<sql:password>goo</sql:password> ","goo");
1.58 +
1.59 +/*
1.60 + * Check performance -
1.61 + */
1.62 +status = inSection(1);
1.63 +var start = new Date();
1.64 +var result = re.exec(str);
1.65 +actual = elapsedTime(start);
1.66 +reportCompare(isThisFast(FAST), isThisFast(actual), status);
1.67 +
1.68 +/*
1.69 + * Check accuracy -
1.70 + */
1.71 +status = inSection(2);
1.72 +testRegExp([status], [re], [str], [result], [expect]);
1.73 +
1.74 +
1.75 +
1.76 +/*
1.77 + * The second regexp (HUGE!). We'll test for performance (Section 3) and accuracy (Section 4).
1.78 + * It comes from the O'Reilly book "Mastering Regular Expressions" by Jeffrey Friedl, Appendix B
1.79 + */
1.80 +
1.81 +//# Some things for avoiding backslashitis later on.
1.82 +$esc = '\\\\';
1.83 +$Period = '\.';
1.84 +$space = '\040'; $tab = '\t';
1.85 +$OpenBR = '\\['; $CloseBR = '\\]';
1.86 +$OpenParen = '\\('; $CloseParen = '\\)';
1.87 +$NonASCII = '\x80-\xff'; $ctrl = '\000-\037';
1.88 +$CRlist = '\n\015'; //# note: this should really be only \015.
1.89 +// Items 19, 20, 21
1.90 +$qtext = '[^' + $esc + $NonASCII + $CRlist + '\"]'; // # for within "..."
1.91 +$dtext = '[^' + $esc + $NonASCII + $CRlist + $OpenBR + $CloseBR + ']'; // # for within [...]
1.92 +$quoted_pair = $esc + '[^' + $NonASCII + ']'; // # an escaped character
1.93 +
1.94 +//##############################################################################
1.95 +//# Items 22 and 23, comment.
1.96 +//# Impossible to do properly with a regex, I make do by allowing at most one level of nesting.
1.97 +$ctext = '[^' + $esc + $NonASCII + $CRlist + '()]';
1.98 +
1.99 +//# $Cnested matches one non-nested comment.
1.100 +//# It is unrolled, with normal of $ctext, special of $quoted_pair.
1.101 +$Cnested =
1.102 + $OpenParen + // # (
1.103 + $ctext + '*' + // # normal*
1.104 + '(?:' + $quoted_pair + $ctext + '*)*' + // # (special normal*)*
1.105 + $CloseParen; // # )
1.106 +
1.107 +
1.108 +//# $comment allows one level of nested parentheses
1.109 +//# It is unrolled, with normal of $ctext, special of ($quoted_pair|$Cnested)
1.110 +$comment =
1.111 + $OpenParen + // # (
1.112 + $ctext + '*' + // # normal*
1.113 + '(?:' + // # (
1.114 + '(?:' + $quoted_pair + '|' + $Cnested + ')' + // # special
1.115 + $ctext + '*' + // # normal*
1.116 + ')*' + // # )*
1.117 + $CloseParen; // # )
1.118 +
1.119 +
1.120 +//##############################################################################
1.121 +//# $X is optional whitespace/comments.
1.122 +$X =
1.123 + '[' + $space + $tab + ']*' + // # Nab whitespace.
1.124 + '(?:' + $comment + '[' + $space + $tab + ']*)*'; // # If comment found, allow more spaces.
1.125 +
1.126 +
1.127 +//# Item 10: atom
1.128 +$atom_char = '[^(' + $space + '<>\@,;:\".' + $esc + $OpenBR + $CloseBR + $ctrl + $NonASCII + ']';
1.129 +$atom =
1.130 + $atom_char + '+' + // # some number of atom characters...
1.131 + '(?!' + $atom_char + ')'; // # ..not followed by something that could be part of an atom
1.132 +
1.133 +// # Item 11: doublequoted string, unrolled.
1.134 +$quoted_str =
1.135 + '\"' + // # "
1.136 + $qtext + '*' + // # normal
1.137 + '(?:' + $quoted_pair + $qtext + '*)*' + // # ( special normal* )*
1.138 + '\"'; // # "
1.139 +
1.140 +//# Item 7: word is an atom or quoted string
1.141 +$word =
1.142 + '(?:' +
1.143 + $atom + // # Atom
1.144 + '|' + // # or
1.145 + $quoted_str + // # Quoted string
1.146 + ')'
1.147 +
1.148 +//# Item 12: domain-ref is just an atom
1.149 + $domain_ref = $atom;
1.150 +
1.151 +//# Item 13: domain-literal is like a quoted string, but [...] instead of "..."
1.152 +$domain_lit =
1.153 + $OpenBR + // # [
1.154 + '(?:' + $dtext + '|' + $quoted_pair + ')*' + // # stuff
1.155 + $CloseBR; // # ]
1.156 +
1.157 +// # Item 9: sub-domain is a domain-ref or domain-literal
1.158 +$sub_domain =
1.159 + '(?:' +
1.160 + $domain_ref +
1.161 + '|' +
1.162 + $domain_lit +
1.163 + ')' +
1.164 + $X; // # optional trailing comments
1.165 +
1.166 +// # Item 6: domain is a list of subdomains separated by dots.
1.167 +$domain =
1.168 + $sub_domain +
1.169 + '(?:' +
1.170 + $Period + $X + $sub_domain +
1.171 + ')*';
1.172 +
1.173 +//# Item 8: a route. A bunch of "@ $domain" separated by commas, followed by a colon.
1.174 +$route =
1.175 + '\@' + $X + $domain +
1.176 + '(?:,' + $X + '\@' + $X + $domain + ')*' + // # additional domains
1.177 + ':' +
1.178 + $X; // # optional trailing comments
1.179 +
1.180 +//# Item 6: local-part is a bunch of $word separated by periods
1.181 +$local_part =
1.182 + $word + $X
1.183 + '(?:' +
1.184 + $Period + $X + $word + $X + // # additional words
1.185 + ')*';
1.186 +
1.187 +// # Item 2: addr-spec is local@domain
1.188 +$addr_spec =
1.189 + $local_part + '\@' + $X + $domain;
1.190 +
1.191 +//# Item 4: route-addr is <route? addr-spec>
1.192 +$route_addr =
1.193 + '<' + $X + // # <
1.194 + '(?:' + $route + ')?' + // # optional route
1.195 + $addr_spec + // # address spec
1.196 + '>'; // # >
1.197 +
1.198 +//# Item 3: phrase........
1.199 +$phrase_ctrl = '\000-\010\012-\037'; // # like ctrl, but without tab
1.200 +
1.201 +//# Like atom-char, but without listing space, and uses phrase_ctrl.
1.202 +//# Since the class is negated, this matches the same as atom-char plus space and tab
1.203 +$phrase_char =
1.204 + '[^()<>\@,;:\".' + $esc + $OpenBR + $CloseBR + $NonASCII + $phrase_ctrl + ']';
1.205 +
1.206 +// # We've worked it so that $word, $comment, and $quoted_str to not consume trailing $X
1.207 +// # because we take care of it manually.
1.208 +$phrase =
1.209 + $word + // # leading word
1.210 + $phrase_char + '*' + // # "normal" atoms and/or spaces
1.211 + '(?:' +
1.212 + '(?:' + $comment + '|' + $quoted_str + ')' + // # "special" comment or quoted string
1.213 + $phrase_char + '*' + // # more "normal"
1.214 + ')*';
1.215 +
1.216 +// ## Item #1: mailbox is an addr_spec or a phrase/route_addr
1.217 +$mailbox =
1.218 + $X + // # optional leading comment
1.219 + '(?:' +
1.220 + $phrase + $route_addr + // # name and address
1.221 + '|' + // # or
1.222 + $addr_spec + // # address
1.223 + ')';
1.224 +
1.225 +
1.226 +//###########################################################################
1.227 +
1.228 +
1.229 +re = new RegExp($mailbox, "g");
1.230 +str = 'Jeffy<"That Tall Guy"@ora.com (this address is no longer active)>';
1.231 +expect = Array('Jeffy<"That Tall Guy"@ora.com (this address is no longer active)>');
1.232 +
1.233 +/*
1.234 + * Check performance -
1.235 + */
1.236 +status = inSection(3);
1.237 +var start = new Date();
1.238 +var result = re.exec(str);
1.239 +actual = elapsedTime(start);
1.240 +reportCompare(isThisFast(FAST), isThisFast(actual), status);
1.241 +
1.242 +/*
1.243 + * Check accuracy -
1.244 + */
1.245 +status = inSection(4);
1.246 +testRegExp([status], [re], [str], [result], [expect]);
