001/////////////////////////////////////////////////////////////////////////////////////////////// 002// checkstyle: Checks Java source code and other text files for adherence to a set of rules. 003// Copyright (C) 2001-2025 the original author or authors. 004// 005// This library is free software; you can redistribute it and/or 006// modify it under the terms of the GNU Lesser General Public 007// License as published by the Free Software Foundation; either 008// version 2.1 of the License, or (at your option) any later version. 009// 010// This library is distributed in the hope that it will be useful, 011// but WITHOUT ANY WARRANTY; without even the implied warranty of 012// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU 013// Lesser General Public License for more details. 014// 015// You should have received a copy of the GNU Lesser General Public 016// License along with this library; if not, write to the Free Software 017// Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA 018/////////////////////////////////////////////////////////////////////////////////////////////// 019 020package com.puppycrawl.tools.checkstyle.meta; 021 022import java.util.Optional; 023import java.util.regex.Matcher; 024import java.util.regex.Pattern; 025 026import com.puppycrawl.tools.checkstyle.api.DetailNode; 027import com.puppycrawl.tools.checkstyle.api.JavadocCommentsTokenTypes; 028import com.puppycrawl.tools.checkstyle.utils.JavadocUtil; 029 030/** 031 * Class for scraping module metadata from the corresponding class' class-level javadoc. 032 */ 033public final class JavadocMetadataScraperUtil { 034 035 /** Regular expression for detecting ANTLR tokens(for e.g. CLASS_DEF). */ 036 private static final Pattern TOKEN_TEXT_PATTERN = Pattern.compile("([A-Z_]{2,})+"); 037 038 /** 039 * Private utility constructor. 040 */ 041 private JavadocMetadataScraperUtil() { 042 } 043 044 /** 045 * Performs a depth-first traversal of the subtree starting at {@code startNode} 046 * and ending at {@code endNode}, and constructs the concatenated text of all nodes 047 * in that range, ignoring {@code JavadocToken} texts. 048 * 049 * @param startNode the node where traversal begins (inclusive) 050 * @param endNode the node where traversal ends (inclusive) 051 * @return the constructed text from the specified subtree range 052 */ 053 public static String constructSubTreeText(DetailNode startNode, 054 DetailNode endNode) { 055 DetailNode curNode = startNode; 056 final StringBuilder result = new StringBuilder(1024); 057 058 while (curNode != null) { 059 if (isContentToWrite(curNode)) { 060 String childText = curNode.getText(); 061 062 if (isInsideCodeInlineTag(curNode)) { 063 childText = adjustCodeInlineTagChildToHtml(curNode); 064 } 065 066 result.append(childText); 067 } 068 069 DetailNode toVisit = curNode.getFirstChild(); 070 while (curNode != endNode && toVisit == null) { 071 toVisit = curNode.getNextSibling(); 072 curNode = curNode.getParent(); 073 } 074 075 curNode = toVisit; 076 } 077 return result.toString().trim(); 078 } 079 080 /** 081 * Checks whether the given node is inside a {@code @code} Javadoc inline tag. 082 * 083 * @param node the node to check 084 * @return true if the node is inside a {@code @code} inline tag, false otherwise 085 */ 086 private static boolean isInsideCodeInlineTag(DetailNode node) { 087 return node.getParent() != null 088 && node.getParent().getType() == JavadocCommentsTokenTypes.CODE_INLINE_TAG; 089 } 090 091 /** 092 * Checks whether selected Javadoc node is considered as something to write. 093 * 094 * @param detailNode javadoc node to check. 095 * @return whether javadoc node is something to write. 096 */ 097 private static boolean isContentToWrite(DetailNode detailNode) { 098 099 return detailNode.getType() != JavadocCommentsTokenTypes.LEADING_ASTERISK 100 && (detailNode.getType() == JavadocCommentsTokenTypes.TEXT 101 || !TOKEN_TEXT_PATTERN.matcher(detailNode.getText()).matches()); 102 } 103 104 /** 105 * Adjusts certain child of {@code @code} Javadoc inline tag to its analogous html format. 106 * 107 * @param codeChild {@code @code} child to convert. 108 * @return converted {@code @code} child element, otherwise just the original text. 109 */ 110 public static String adjustCodeInlineTagChildToHtml(DetailNode codeChild) { 111 112 return switch (codeChild.getType()) { 113 case JavadocCommentsTokenTypes.JAVADOC_INLINE_TAG_END -> "</code>"; 114 case JavadocCommentsTokenTypes.TAG_NAME -> ""; 115 case JavadocCommentsTokenTypes.JAVADOC_INLINE_TAG_START -> "<code>"; 116 default -> codeChild.getText().trim(); 117 }; 118 } 119 120 /** 121 * Returns the first child node of the given parent that matches the provided {@code tokenType}. 122 * 123 * @param node the parent node 124 * @param tokenType the token type to match 125 * @return an {@link Optional} containing the first matching child node, 126 * or an empty {@link Optional} if none is found 127 */ 128 private static Optional<DetailNode> getFirstChildOfType(DetailNode node, int tokenType) { 129 return JavadocUtil.getAllNodesOfType(node, tokenType).stream().findFirst(); 130 } 131 132 /** 133 * Checks whether the first child {@code JavadocTokenType.TEXT} node matches given pattern. 134 * 135 * @param ast parent javadoc node 136 * @param pattern pattern to match 137 * @return true if one of child text nodes matches pattern 138 */ 139 public static boolean isChildNodeTextMatches(DetailNode ast, Pattern pattern) { 140 return getFirstChildOfType(ast, JavadocCommentsTokenTypes.TEXT) 141 .map(DetailNode::getText) 142 .map(pattern::matcher) 143 .map(Matcher::matches) 144 .orElse(Boolean.FALSE); 145 } 146}