
org.sonarsource.analyzer.commons.regex.finders.ImpossibleBoundaryFinder Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of sonar-regex-parsing Show documentation
Show all versions of sonar-regex-parsing Show documentation
Logic useful to read and analyze regular expressions
The newest version!
/*
* SonarSource Analyzers Regex Parsing Commons
* Copyright (C) 2009-2024 SonarSource SA
* mailto:info AT sonarsource DOT com
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 3 of the License, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public License
* along with this program; if not, write to the Free Software Foundation,
* Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
*/
package org.sonarsource.analyzer.commons.regex.finders;
import java.util.Collections;
import java.util.HashSet;
import java.util.Set;
import java.util.regex.Pattern;
import org.sonarsource.analyzer.commons.regex.RegexIssueReporter;
import org.sonarsource.analyzer.commons.regex.RegexParseResult;
import org.sonarsource.analyzer.commons.regex.ast.AutomatonState;
import org.sonarsource.analyzer.commons.regex.ast.BoundaryTree;
import org.sonarsource.analyzer.commons.regex.ast.DisjunctionTree;
import org.sonarsource.analyzer.commons.regex.ast.LookAroundTree;
import org.sonarsource.analyzer.commons.regex.ast.RegexBaseVisitor;
import org.sonarsource.analyzer.commons.regex.ast.RegexTree;
import org.sonarsource.analyzer.commons.regex.helpers.RegexReachabilityChecker;
public class ImpossibleBoundaryFinder extends RegexBaseVisitor {
private static final String MESSAGE = "Remove or replace this boundary that will never match because it appears %s mandatory input.";
private static final String SOFT_MESSAGE =
"Remove or replace this boundary that can only match if the previous part matched the empty string because it appears %s mandatory input.";
private final Set excluded = new HashSet<>();
private final RegexReachabilityChecker regexReachabilityChecker = new RegexReachabilityChecker(false);
private final RegexIssueReporter.ElementIssue regexElementIssueReporter;
private AutomatonState start;
private AutomatonState end;
public ImpossibleBoundaryFinder(RegexIssueReporter.ElementIssue regexElementIssueReporter) {
this.regexElementIssueReporter = regexElementIssueReporter;
}
@Override
public void visit(RegexParseResult regexParseResult) {
regexReachabilityChecker.clearCache();
start = regexParseResult.getStartState();
end = regexParseResult.getFinalState();
super.visit(regexParseResult);
}
@Override
public void visitLookAround(LookAroundTree tree) {
// Inside a lookaround we consider the end/start of the lookahead/behind respectively as if it were the end/start
// of the regex. This avoids false positives for cases like `(?=.*$)foo` or `foo(?<=^...)`.
if (tree.getDirection() == LookAroundTree.Direction.BEHIND) {
AutomatonState oldStart = start;
start = tree.getElement();
super.visitLookAround(tree);
start = oldStart;
} else {
AutomatonState oldEnd = end;
// Set end to the lookaround's end-of-lookaround state
end = tree.getElement().continuation();
super.visitLookAround(tree);
end = oldEnd;
}
}
@Override
public void visitDisjunction(DisjunctionTree tree) {
BoundaryInDisjunctionFinder boundaryInDisjunctionFinder = new BoundaryInDisjunctionFinder();
boundaryInDisjunctionFinder.visit(tree);
excluded.addAll(boundaryInDisjunctionFinder.foundBoundaries());
super.visitDisjunction(tree);
}
@Override
public void visitBoundary(BoundaryTree boundaryTree) {
switch (boundaryTree.type()) {
case LINE_START:
if (!boundaryTree.activeFlags().contains(Pattern.MULTILINE)) {
checkStartBoundary(boundaryTree);
}
break;
case INPUT_START:
checkStartBoundary(boundaryTree);
break;
case LINE_END:
if (!boundaryTree.activeFlags().contains(Pattern.MULTILINE)) {
checkEndBoundary(boundaryTree);
}
break;
case INPUT_END:
case INPUT_END_FINAL_TERMINATOR:
checkEndBoundary(boundaryTree);
break;
default:
// Do nothing
}
}
private void checkStartBoundary(BoundaryTree boundaryTree) {
if (!RegexReachabilityChecker.canReachWithoutConsumingInput(start, boundaryTree)) {
regexElementIssueReporter.report(boundaryTree, String.format(MESSAGE, "after"), null, Collections.emptyList());
} else if (!excluded.contains(boundaryTree) && probablyShouldConsumeInput(start, boundaryTree)) {
regexElementIssueReporter.report(boundaryTree, String.format(SOFT_MESSAGE, "after"), null, Collections.emptyList());
}
}
private void checkEndBoundary(BoundaryTree boundaryTree) {
if (!RegexReachabilityChecker.canReachWithoutConsumingInput(boundaryTree, end)) {
regexElementIssueReporter.report(boundaryTree, String.format(MESSAGE, "before"), null, Collections.emptyList());
} else if (!excluded.contains(boundaryTree) && probablyShouldConsumeInput(boundaryTree, end)) {
regexElementIssueReporter.report(boundaryTree, String.format(SOFT_MESSAGE, "before"), null, Collections.emptyList());
}
}
private boolean probablyShouldConsumeInput(AutomatonState start, AutomatonState goal) {
return regexReachabilityChecker.canReachWithConsumingInput(start, goal, new HashSet<>());
}
private static class BoundaryInDisjunctionFinder extends RegexBaseVisitor {
private final Set foundBoundaries = new HashSet<>();
@Override
public void visitBoundary(BoundaryTree boundaryTree) {
foundBoundaries.add(boundaryTree);
}
public Set foundBoundaries() {
return new HashSet<>(foundBoundaries);
}
}
}
© 2015 - 2025 Weber Informatics LLC | Privacy Policy