136670652c
This CL adds support for disjunctions and some quantification in EXPERIMENTAL regexp patterns. It is implemented using a new bytecode format and an NFA-based breadth-first interpreter. R=jgruber@chromium.org Bug: v8:10765 Change-Id: Idd49a3bbc9a9fcc2be80d822c9d84a638e53e777 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2370634 Commit-Queue: Martin Bidlingmaier <mbid@google.com> Reviewed-by: Dominik Inführ <dinfuehr@chromium.org> Reviewed-by: Jakob Gruber <jgruber@chromium.org> Reviewed-by: Clemens Backes <clemensb@chromium.org> Cr-Commit-Position: refs/heads/master@{#69621}
103 lines
3.8 KiB
JavaScript
103 lines
3.8 KiB
JavaScript
// Copyright 2019 the V8 project authors. All rights reserved.
|
|
// Use of this source code is governed by a BSD-style license that can be
|
|
// found in the LICENSE file.
|
|
|
|
// Tier-up behavior differs between slow and fast paths in
|
|
// RegExp.prototype.replace with a function as an argument.
|
|
// Flags: --regexp-tier-up --regexp-tier-up-ticks=5
|
|
// Flags: --allow-natives-syntax --no-force-slow-path --no-regexp-interpret-all
|
|
// Flags: --no-enable-experimental-regexp-engine
|
|
|
|
const kLatin1 = true;
|
|
const kUnicode = false;
|
|
|
|
function CheckRegexpNotYetCompiled(regexp) {
|
|
assertFalse(%RegexpHasBytecode(regexp, kLatin1) &&
|
|
%RegexpHasNativeCode(regexp, kLatin1));
|
|
assertFalse(%RegexpHasBytecode(regexp, kUnicode) &&
|
|
%RegexpHasNativeCode(regexp, kUnicode));
|
|
}
|
|
|
|
// Testing RegExp.test method which calls into Runtime_RegExpExec.
|
|
let re = new RegExp('^.$');
|
|
CheckRegexpNotYetCompiled(re);
|
|
|
|
// Testing first five executions of regexp with one-byte string subject.
|
|
for (var i = 0; i < 5; i++) {
|
|
re.test("a");
|
|
assertTrue(%RegexpHasBytecode(re, kLatin1));
|
|
assertTrue(!%RegexpHasBytecode(re, kUnicode) &&
|
|
!%RegexpHasNativeCode(re, kUnicode));
|
|
}
|
|
// Testing the tier-up to native code.
|
|
re.test("a");
|
|
assertTrue(!%RegexpHasBytecode(re, kLatin1) &&
|
|
%RegexpHasNativeCode(re,kLatin1));
|
|
assertTrue(!%RegexpHasBytecode(re, kUnicode) &&
|
|
!%RegexpHasNativeCode(re,kUnicode));
|
|
re.test("a");
|
|
assertTrue(!%RegexpHasBytecode(re, kLatin1) &&
|
|
%RegexpHasNativeCode(re,kLatin1));
|
|
assertTrue(!%RegexpHasBytecode(re, kUnicode) &&
|
|
!%RegexpHasNativeCode(re,kUnicode));
|
|
// Testing that the regexp will compile to native code for two-byte string
|
|
// subject as well, because we have a single tick counter for both string
|
|
// representations.
|
|
re.test("π");
|
|
assertTrue(!%RegexpHasBytecode(re, kLatin1) &&
|
|
%RegexpHasNativeCode(re,kLatin1));
|
|
assertTrue(!%RegexpHasBytecode(re, kUnicode) &&
|
|
%RegexpHasNativeCode(re,kUnicode));
|
|
|
|
// Testing String.replace method for non-global regexps.
|
|
var subject = "a1111";
|
|
re = /\w1/;
|
|
CheckRegexpNotYetCompiled(re);
|
|
|
|
for (var i = 0; i < 5; i++) {
|
|
subject.replace(re, "x");
|
|
assertTrue(%RegexpHasBytecode(re, kLatin1));
|
|
assertTrue(!%RegexpHasBytecode(re, kUnicode) &&
|
|
!%RegexpHasNativeCode(re, kUnicode));
|
|
}
|
|
|
|
subject.replace(re, "x");
|
|
assertTrue(!%RegexpHasBytecode(re, kLatin1) &&
|
|
%RegexpHasNativeCode(re, kLatin1));
|
|
assertTrue(!%RegexpHasBytecode(re, kUnicode) &&
|
|
!%RegexpHasNativeCode(re, kUnicode));
|
|
|
|
// Testing String.replace method for global regexps.
|
|
let re_g = /\w11111/g;
|
|
CheckRegexpNotYetCompiled(re_g);
|
|
// This regexp will not match, so it will only execute the bytecode once,
|
|
// each time the replace method is invoked, without tiering-up and
|
|
// recompiling to native code.
|
|
for (var i = 0; i < 5; i++) {
|
|
subject.replace(re_g, "x");
|
|
assertTrue(%RegexpHasBytecode(re_g, kLatin1));
|
|
assertTrue(!%RegexpHasBytecode(re_g, kUnicode) &&
|
|
!%RegexpHasNativeCode(re_g, kUnicode));
|
|
}
|
|
|
|
// This regexp will match, so it will execute five times, and tier-up.
|
|
re_g = /\w/g;
|
|
CheckRegexpNotYetCompiled(re_g);
|
|
subject.replace(re_g, "x");
|
|
assertTrue(!%RegexpHasBytecode(re_g, kLatin1) &&
|
|
%RegexpHasNativeCode(re_g, kLatin1));
|
|
assertTrue(!%RegexpHasBytecode(re_g, kUnicode) &&
|
|
!%RegexpHasNativeCode(re_g, kUnicode));
|
|
|
|
// Testing String.replace method for global regexps with a function as a
|
|
// parameter. This will tier-up eagerly and compile to native code right
|
|
// away, even though the regexp is only executed once.
|
|
function f() { return "x"; }
|
|
re_g = /\w2/g;
|
|
CheckRegexpNotYetCompiled(re_g);
|
|
subject.replace(re_g, f);
|
|
assertTrue(!%RegexpHasBytecode(re_g, kLatin1) &&
|
|
%RegexpHasNativeCode(re_g, kLatin1));
|
|
assertTrue(!%RegexpHasBytecode(re_g, kUnicode) &&
|
|
!%RegexpHasNativeCode(re_g, kUnicode));
|