NoWhitespaceAfterCheck.java

///////////////////////////////////////////////////////////////////////////////////////////////
// checkstyle: Checks Java source code and other text files for adherence to a set of rules.
// Copyright (C) 2001-2022 the original author or authors.
//
// This library 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 2.1 of the License, or (at your option) any later version.
//
// This library 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 library; if not, write to the Free Software
// Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
///////////////////////////////////////////////////////////////////////////////////////////////

package com.puppycrawl.tools.checkstyle.checks.whitespace;

import java.util.Optional;

import com.puppycrawl.tools.checkstyle.StatelessCheck;
import com.puppycrawl.tools.checkstyle.api.AbstractCheck;
import com.puppycrawl.tools.checkstyle.api.DetailAST;
import com.puppycrawl.tools.checkstyle.api.TokenTypes;
import com.puppycrawl.tools.checkstyle.utils.CommonUtil;

/**
 * <p>
 * Checks that there is no whitespace after a token.
 * More specifically, it checks that it is not followed by whitespace,
 * or (if linebreaks are allowed) all characters on the line after are
 * whitespace. To forbid linebreaks after a token, set property
 * {@code allowLineBreaks} to {@code false}.
 * </p>
 * <p>
 * The check processes
 * <a href="https://checkstyle.org/apidocs/com/puppycrawl/tools/checkstyle/api/TokenTypes.html#ARRAY_DECLARATOR">
 * ARRAY_DECLARATOR</a> and
 * <a href="https://checkstyle.org/apidocs/com/puppycrawl/tools/checkstyle/api/TokenTypes.html#INDEX_OP">
 * INDEX_OP</a> tokens specially from other tokens. Actually it is checked that
 * there is no whitespace before these tokens, not after them. Space after the
 * <a href="https://checkstyle.org/apidocs/com/puppycrawl/tools/checkstyle/api/TokenTypes.html#ANNOTATIONS">
 * ANNOTATIONS</a> before
 * <a href="https://checkstyle.org/apidocs/com/puppycrawl/tools/checkstyle/api/TokenTypes.html#ARRAY_DECLARATOR">
 * ARRAY_DECLARATOR</a> and
 * <a href="https://checkstyle.org/apidocs/com/puppycrawl/tools/checkstyle/api/TokenTypes.html#INDEX_OP">
 * INDEX_OP</a> will be ignored.
 * </p>
 * <p>
 * If the annotation is between the type and the array, the check will skip validation for spaces
 * </p>
 * <pre>
 * public void foo(final char @NotNull [] param) {} // No violation
 * </pre>
 * <ul>
 * <li>
 * Property {@code allowLineBreaks} - Control whether whitespace is allowed
 * if the token is at a linebreak.
 * Type is {@code boolean}.
 * Default value is {@code true}.
 * </li>
 * <li>
 * Property {@code tokens} - tokens to check
 * Type is {@code java.lang.String[]}.
 * Validation type is {@code tokenSet}.
 * Default value is:
 * <a href="https://checkstyle.org/apidocs/com/puppycrawl/tools/checkstyle/api/TokenTypes.html#ARRAY_INIT">
 * ARRAY_INIT</a>,
 * <a href="https://checkstyle.org/apidocs/com/puppycrawl/tools/checkstyle/api/TokenTypes.html#AT">
 * AT</a>,
 * <a href="https://checkstyle.org/apidocs/com/puppycrawl/tools/checkstyle/api/TokenTypes.html#INC">
 * INC</a>,
 * <a href="https://checkstyle.org/apidocs/com/puppycrawl/tools/checkstyle/api/TokenTypes.html#DEC">
 * DEC</a>,
 * <a href="https://checkstyle.org/apidocs/com/puppycrawl/tools/checkstyle/api/TokenTypes.html#UNARY_MINUS">
 * UNARY_MINUS</a>,
 * <a href="https://checkstyle.org/apidocs/com/puppycrawl/tools/checkstyle/api/TokenTypes.html#UNARY_PLUS">
 * UNARY_PLUS</a>,
 * <a href="https://checkstyle.org/apidocs/com/puppycrawl/tools/checkstyle/api/TokenTypes.html#BNOT">
 * BNOT</a>,
 * <a href="https://checkstyle.org/apidocs/com/puppycrawl/tools/checkstyle/api/TokenTypes.html#LNOT">
 * LNOT</a>,
 * <a href="https://checkstyle.org/apidocs/com/puppycrawl/tools/checkstyle/api/TokenTypes.html#DOT">
 * DOT</a>,
 * <a href="https://checkstyle.org/apidocs/com/puppycrawl/tools/checkstyle/api/TokenTypes.html#ARRAY_DECLARATOR">
 * ARRAY_DECLARATOR</a>,
 * <a href="https://checkstyle.org/apidocs/com/puppycrawl/tools/checkstyle/api/TokenTypes.html#INDEX_OP">
 * INDEX_OP</a>.
 * </li>
 * </ul>
 * <p>
 * To configure the check:
 * </p>
 * <pre>
 * &lt;module name=&quot;NoWhitespaceAfter&quot;/&gt;
 * </pre>
 * <p>Example:</p>
 * <pre>
 * class Test {
 *
 *   public void lineBreak(String x) {
 *     Integer.
 *         parseInt(x); // Ok
 *     Integer.parseInt(x); // Ok
 *   }
 *
 *   public void dotOperator(String s) {
 *     Integer.parseInt(s); // Ok
 *     Integer. parseInt(s); // violation, '.' is followed by whitespace
 *   }
 *
 *   public void arrayDec() {
 *     int[] arr = arr; // Ok
 *     int [] arr = arr; // violation, int is followed by whitespace
 *   }
 *
 *   public void bitwiseNot(int a) {
 *     a = ~ a; // violation '~' is followed by whitespace
 *     a = ~a; // Ok
 *   }
 * }
 * </pre>
 * <p>To configure the check to forbid linebreaks after a DOT token:
 * </p>
 * <pre>
 * &lt;module name=&quot;NoWhitespaceAfter&quot;&gt;
 *   &lt;property name=&quot;tokens&quot; value=&quot;DOT&quot;/&gt;
 *   &lt;property name=&quot;allowLineBreaks&quot; value=&quot;false&quot;/&gt;
 * &lt;/module&gt;
 * </pre>
 * <p>Example:</p>
 * <pre>
 * class Test {
 *
 *   public void lineBreak(String x) {
 *     Integer.
 *         parseInt(x); // violation, '.' is followed by whitespace
 *     Integer.parseInt(x); // Ok
 *   }
 *
 *   public void dotOperator(String s) {
 *     Integer.parseInt(s); // Ok
 *     Integer. parseInt(s); // violation, '.' is followed by whitespace
 *   }
 *
 *   public void arrayDec() {
 *     int[] arr = arr; // Ok
 *     int [] arr = arr; // Ok
 *   }
 *
 *   public void bitwiseNot(int a) {
 *     a = ~ a; // Ok
 *     a = ~a; // Ok
 *   }
 * }
 * </pre>
 * <p>
 * Parent is {@code com.puppycrawl.tools.checkstyle.TreeWalker}
 * </p>
 * <p>
 * Violation Message Keys:
 * </p>
 * <ul>
 * <li>
 * {@code ws.followed}
 * </li>
 * </ul>
 *
 * @since 3.0
 */
@StatelessCheck
public class NoWhitespaceAfterCheck extends AbstractCheck {

    /**
     * A key is pointing to the warning message text in "messages.properties"
     * file.
     */
    public static final String MSG_KEY = "ws.followed";

    /** Control whether whitespace is allowed if the token is at a linebreak. */
    private boolean allowLineBreaks = true;

    @Override
    public int[] getDefaultTokens() {
        return new int[] {
            TokenTypes.ARRAY_INIT,
            TokenTypes.AT,
            TokenTypes.INC,
            TokenTypes.DEC,
            TokenTypes.UNARY_MINUS,
            TokenTypes.UNARY_PLUS,
            TokenTypes.BNOT,
            TokenTypes.LNOT,
            TokenTypes.DOT,
            TokenTypes.ARRAY_DECLARATOR,
            TokenTypes.INDEX_OP,
        };
    }

    @Override
    public int[] getAcceptableTokens() {
        return new int[] {
            TokenTypes.ARRAY_INIT,
            TokenTypes.AT,
            TokenTypes.INC,
            TokenTypes.DEC,
            TokenTypes.UNARY_MINUS,
            TokenTypes.UNARY_PLUS,
            TokenTypes.BNOT,
            TokenTypes.LNOT,
            TokenTypes.DOT,
            TokenTypes.TYPECAST,
            TokenTypes.ARRAY_DECLARATOR,
            TokenTypes.INDEX_OP,
            TokenTypes.LITERAL_SYNCHRONIZED,
            TokenTypes.METHOD_REF,
        };
    }

    @Override
    public int[] getRequiredTokens() {
        return CommonUtil.EMPTY_INT_ARRAY;
    }

    /**
     * Setter to control whether whitespace is allowed if the token is at a linebreak.
     *
     * @param allowLineBreaks whether whitespace should be
     *     flagged at linebreaks.
     */
    public void setAllowLineBreaks(boolean allowLineBreaks) {
        this.allowLineBreaks = allowLineBreaks;
    }

    @Override
    public void visitToken(DetailAST ast) {
        if (shouldCheckWhitespaceAfter(ast)) {
            final DetailAST whitespaceFollowedAst = getWhitespaceFollowedNode(ast);
            final int whitespaceColumnNo = getPositionAfter(whitespaceFollowedAst);
            final int whitespaceLineNo = whitespaceFollowedAst.getLineNo();

            if (hasTrailingWhitespace(ast, whitespaceColumnNo, whitespaceLineNo)) {
                log(ast, MSG_KEY, whitespaceFollowedAst.getText());
            }
        }
    }

    /**
     * For a visited ast node returns node that should be checked
     * for not being followed by whitespace.
     *
     * @param ast
     *        , visited node.
     * @return node before ast.
     */
    private static DetailAST getWhitespaceFollowedNode(DetailAST ast) {
        final DetailAST whitespaceFollowedAst;
        switch (ast.getType()) {
            case TokenTypes.TYPECAST:
                whitespaceFollowedAst = ast.findFirstToken(TokenTypes.RPAREN);
                break;
            case TokenTypes.ARRAY_DECLARATOR:
                whitespaceFollowedAst = getArrayDeclaratorPreviousElement(ast);
                break;
            case TokenTypes.INDEX_OP:
                whitespaceFollowedAst = getIndexOpPreviousElement(ast);
                break;
            default:
                whitespaceFollowedAst = ast;
        }
        return whitespaceFollowedAst;
    }

    /**
     * Returns whether whitespace after a visited node should be checked. For example, whitespace
     * is not allowed between a type and an array declarator (returns true), except when there is
     * an annotation in between the type and array declarator (returns false).
     *
     * @param ast the visited node
     * @return true if whitespace after ast should be checked
     */
    private static boolean shouldCheckWhitespaceAfter(DetailAST ast) {
        boolean checkWhitespace = true;
        final DetailAST previousSibling = ast.getPreviousSibling();
        if (previousSibling != null && previousSibling.getType() == TokenTypes.ANNOTATIONS) {
            checkWhitespace = false;
        }
        return checkWhitespace;
    }

    /**
     * Gets position after token (place of possible redundant whitespace).
     *
     * @param ast Node representing token.
     * @return position after token.
     */
    private static int getPositionAfter(DetailAST ast) {
        final int after;
        // If target of possible redundant whitespace is in method definition.
        if (ast.getType() == TokenTypes.IDENT
                && ast.getNextSibling() != null
                && ast.getNextSibling().getType() == TokenTypes.LPAREN) {
            final DetailAST methodDef = ast.getParent();
            final DetailAST endOfParams = methodDef.findFirstToken(TokenTypes.RPAREN);
            after = endOfParams.getColumnNo() + 1;
        }
        else {
            after = ast.getColumnNo() + ast.getText().length();
        }
        return after;
    }

    /**
     * Checks if there is unwanted whitespace after the visited node.
     *
     * @param ast
     *        , visited node.
     * @param whitespaceColumnNo
     *        , column number of a possible whitespace.
     * @param whitespaceLineNo
     *        , line number of a possible whitespace.
     * @return true if whitespace found.
     */
    private boolean hasTrailingWhitespace(DetailAST ast,
        int whitespaceColumnNo, int whitespaceLineNo) {
        final boolean result;
        final int astLineNo = ast.getLineNo();
        final int[] line = getLineCodePoints(astLineNo - 1);
        if (astLineNo == whitespaceLineNo && whitespaceColumnNo < line.length) {
            result = CommonUtil.isCodePointWhitespace(line, whitespaceColumnNo);
        }
        else {
            result = !allowLineBreaks;
        }
        return result;
    }

    /**
     * Returns proper argument for getPositionAfter method, it is a token after
     * {@link TokenTypes#ARRAY_DECLARATOR ARRAY_DECLARATOR}, in can be {@link TokenTypes#RBRACK
     * RBRACK}, {@link TokenTypes#IDENT IDENT} or an array type definition (literal).
     *
     * @param ast
     *        , {@link TokenTypes#ARRAY_DECLARATOR ARRAY_DECLARATOR} node.
     * @return previous node by text order.
     * @throws IllegalStateException if an unexpected token type is encountered.
     */
    private static DetailAST getArrayDeclaratorPreviousElement(DetailAST ast) {
        final DetailAST previousElement;

        if (ast.getPreviousSibling() != null
                && ast.getPreviousSibling().getType() == TokenTypes.ARRAY_DECLARATOR) {
            // Covers higher dimension array declarations and initializations
            previousElement = getPreviousElementOfMultiDimArray(ast);
        }
        else {
            // first array index, is preceded with identifier or type
            final DetailAST parent = ast.getParent();
            switch (parent.getType()) {
                // generics
                case TokenTypes.TYPE_UPPER_BOUNDS:
                case TokenTypes.TYPE_LOWER_BOUNDS:
                    previousElement = ast.getPreviousSibling();
                    break;
                case TokenTypes.LITERAL_NEW:
                case TokenTypes.TYPE_ARGUMENT:
                case TokenTypes.DOT:
                    previousElement = getTypeLastNode(ast);
                    break;
                // mundane array declaration, can be either java style or C style
                case TokenTypes.TYPE:
                    previousElement = getPreviousNodeWithParentOfTypeAst(ast, parent);
                    break;
                // java 8 method reference
                case TokenTypes.METHOD_REF:
                    final DetailAST ident = getIdentLastToken(ast);
                    if (ident == null) {
                        // i.e. int[]::new
                        previousElement = ast.getParent().getFirstChild();
                    }
                    else {
                        previousElement = ident;
                    }
                    break;
                default:
                    throw new IllegalStateException("unexpected ast syntax " + parent);
            }
        }
        return previousElement;
    }

    /**
     * Gets the previous element of a second or higher dimension of an
     * array declaration or initialization.
     *
     * @param leftBracket the token to get previous element of
     * @return the previous element
     */
    private static DetailAST getPreviousElementOfMultiDimArray(DetailAST leftBracket) {
        final DetailAST previousRightBracket = leftBracket.getPreviousSibling().getLastChild();

        DetailAST ident = null;
        // This will get us past the type ident, to the actual identifier
        DetailAST parent = leftBracket.getParent().getParent();
        while (ident == null) {
            ident = parent.findFirstToken(TokenTypes.IDENT);
            parent = parent.getParent();
        }

        final DetailAST previousElement;
        if (ident.getColumnNo() > previousRightBracket.getColumnNo()
                && ident.getColumnNo() < leftBracket.getColumnNo()) {
            // C style and Java style ' int[] arr []' in same construct
            previousElement = ident;
        }
        else {
            // 'int[][] arr' or 'int arr[][]'
            previousElement = previousRightBracket;
        }
        return previousElement;
    }

    /**
     * Gets previous node for {@link TokenTypes#INDEX_OP INDEX_OP} token
     * for usage in getPositionAfter method, it is a simplified copy of
     * getArrayDeclaratorPreviousElement method.
     *
     * @param ast
     *        , {@link TokenTypes#INDEX_OP INDEX_OP} node.
     * @return previous node by text order.
     */
    private static DetailAST getIndexOpPreviousElement(DetailAST ast) {
        final DetailAST result;
        final DetailAST firstChild = ast.getFirstChild();
        if (firstChild.getType() == TokenTypes.INDEX_OP) {
            // second or higher array index
            result = firstChild.findFirstToken(TokenTypes.RBRACK);
        }
        else if (firstChild.getType() == TokenTypes.IDENT) {
            result = firstChild;
        }
        else {
            final DetailAST ident = getIdentLastToken(ast);
            if (ident == null) {
                final DetailAST rparen = ast.findFirstToken(TokenTypes.RPAREN);
                // construction like new int[]{1}[0]
                if (rparen == null) {
                    final DetailAST lastChild = firstChild.getLastChild();
                    result = lastChild.findFirstToken(TokenTypes.RCURLY);
                }
                // construction like ((byte[]) pixels)[0]
                else {
                    result = rparen;
                }
            }
            else {
                result = ident;
            }
        }
        return result;
    }

    /**
     * Searches parameter node for a type node.
     * Returns it or its last node if it has an extended structure.
     *
     * @param ast
     *        , subject node.
     * @return type node.
     */
    private static DetailAST getTypeLastNode(DetailAST ast) {
        final DetailAST typeLastNode;
        final DetailAST parent = ast.getParent();
        final boolean isPrecededByTypeArgs =
                parent.findFirstToken(TokenTypes.TYPE_ARGUMENTS) != null;
        final Optional<DetailAST> objectArrayType = Optional.ofNullable(getIdentLastToken(ast));

        if (isPrecededByTypeArgs) {
            typeLastNode = parent.findFirstToken(TokenTypes.TYPE_ARGUMENTS)
                    .findFirstToken(TokenTypes.GENERIC_END);
        }
        else if (objectArrayType.isPresent()) {
            typeLastNode = objectArrayType.get();
        }
        else {
            typeLastNode = parent.getFirstChild();
        }

        return typeLastNode;
    }

    /**
     * Finds previous node by text order for an array declarator,
     * which parent type is {@link TokenTypes#TYPE TYPE}.
     *
     * @param ast
     *        , array declarator node.
     * @param parent
     *        , its parent node.
     * @return previous node by text order.
     */
    private static DetailAST getPreviousNodeWithParentOfTypeAst(DetailAST ast, DetailAST parent) {
        final DetailAST previousElement;
        final DetailAST ident = getIdentLastToken(parent.getParent());
        final DetailAST lastTypeNode = getTypeLastNode(ast);
        // sometimes there are ident-less sentences
        // i.e. "(Object[]) null", but in casual case should be
        // checked whether ident or lastTypeNode has preceding position
        // determining if it is java style or C style

        if (ident == null || ident.getLineNo() > ast.getLineNo()) {
            previousElement = lastTypeNode;
        }
        else if (ident.getLineNo() < ast.getLineNo()) {
            previousElement = ident;
        }
        // ident and lastTypeNode lay on one line
        else {
            final int instanceOfSize = 13;
            // +2 because ast has `[]` after the ident
            if (ident.getColumnNo() >= ast.getColumnNo() + 2
                // +13 because ident (at most 1 character) is followed by
                // ' instanceof ' (12 characters)
                || lastTypeNode.getColumnNo() >= ident.getColumnNo() + instanceOfSize) {
                previousElement = lastTypeNode;
            }
            else {
                previousElement = ident;
            }
        }
        return previousElement;
    }

    /**
     * Gets leftmost token of identifier.
     *
     * @param ast
     *        , token possibly possessing an identifier.
     * @return leftmost token of identifier.
     */
    private static DetailAST getIdentLastToken(DetailAST ast) {
        final DetailAST result;
        final Optional<DetailAST> dot = getPrecedingDot(ast);
        // method call case
        if (dot.isEmpty() || ast.getFirstChild().getType() == TokenTypes.METHOD_CALL) {
            final DetailAST methodCall = ast.findFirstToken(TokenTypes.METHOD_CALL);
            if (methodCall == null) {
                result = ast.findFirstToken(TokenTypes.IDENT);
            }
            else {
                result = methodCall.findFirstToken(TokenTypes.RPAREN);
            }
        }
        // qualified name case
        else {
            result = dot.get().getFirstChild().getNextSibling();
        }
        return result;
    }

    /**
     * Gets the dot preceding a class member array index operation or class
     * reference.
     *
     * @param leftBracket the ast we are checking
     * @return dot preceding the left bracket
     */
    private static Optional<DetailAST> getPrecedingDot(DetailAST leftBracket) {
        final DetailAST referencedMemberDot = leftBracket.findFirstToken(TokenTypes.DOT);
        final Optional<DetailAST> result = Optional.ofNullable(referencedMemberDot);
        return result.or(() -> getReferencedClassDot(leftBracket));
    }

    /**
     * Gets the dot preceding a class reference.
     *
     * @param leftBracket the ast we are checking
     * @return dot preceding the left bracket
     */
    private static Optional<DetailAST> getReferencedClassDot(DetailAST leftBracket) {
        final DetailAST parent = leftBracket.getParent();
        Optional<DetailAST> classDot = Optional.empty();
        if (parent.getType() != TokenTypes.ASSIGN) {
            classDot = Optional.ofNullable(parent.findFirstToken(TokenTypes.DOT));
        }
        return classDot;
    }
}