001///////////////////////////////////////////////////////////////////////////////////////////////
002// checkstyle: Checks Java source code and other text files for adherence to a set of rules.
003// Copyright (C) 2001-2026 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.site;
021
022import java.lang.reflect.Field;
023import java.util.Set;
024import java.util.regex.Pattern;
025
026import org.apache.maven.doxia.macro.MacroExecutionException;
027import org.apache.maven.doxia.sink.Sink;
028
029import com.puppycrawl.tools.checkstyle.PropertyType;
030import com.puppycrawl.tools.checkstyle.XdocsPropertyType;
031import com.puppycrawl.tools.checkstyle.api.DetailNode;
032import com.puppycrawl.tools.checkstyle.api.JavadocCommentsTokenTypes;
033import com.puppycrawl.tools.checkstyle.meta.JavadocMetadataScraperUtil;
034import com.puppycrawl.tools.checkstyle.utils.JavadocUtil;
035
036/**
037 * Utility class for parsing javadocs of modules.
038 */
039public final class ModuleJavadocParsingUtil {
040
041    /** New line escape character. */
042    public static final String NEWLINE = System.lineSeparator();
043    /** A newline with 4 spaces of indentation. */
044    public static final String INDENT_LEVEL_4 = SiteUtil.getNewlineAndIndentSpaces(4);
045    /** A newline with 6 spaces of indentation. */
046    public static final String INDENT_LEVEL_6 = SiteUtil.getNewlineAndIndentSpaces(6);
047    /** A newline with 8 spaces of indentation. */
048    public static final String INDENT_LEVEL_8 = SiteUtil.getNewlineAndIndentSpaces(8);
049    /** A newline with 10 spaces of indentation. */
050    public static final String INDENT_LEVEL_10 = SiteUtil.getNewlineAndIndentSpaces(10);
051    /** A newline with 12 spaces of indentation. */
052    public static final String INDENT_LEVEL_12 = SiteUtil.getNewlineAndIndentSpaces(12);
053    /** A newline with 14 spaces of indentation. */
054    public static final String INDENT_LEVEL_14 = SiteUtil.getNewlineAndIndentSpaces(14);
055    /** A newline with 16 spaces of indentation. */
056    public static final String INDENT_LEVEL_16 = SiteUtil.getNewlineAndIndentSpaces(16);
057    /** A newline with 18 spaces of indentation. */
058    public static final String INDENT_LEVEL_18 = SiteUtil.getNewlineAndIndentSpaces(18);
059    /** A newline with 20 spaces of indentation. */
060    public static final String INDENT_LEVEL_20 = SiteUtil.getNewlineAndIndentSpaces(20);
061    /** A set of all html tags that need to be considered as text formatting for this macro. */
062    public static final Set<String> HTML_TEXT_FORMAT_TAGS = Set.of("<code>", "<a", "</a>", "<b>",
063        "</b>", "<strong>", "</strong>", "<i>", "</i>", "<em>", "</em>", "<small>", "</small>",
064        "<ins>", "<sub>", "<sup>");
065    /** "Notes:" javadoc marking. */
066    public static final String NOTES = "Notes:";
067    /** "Notes:" line. */
068    public static final Pattern NOTES_LINE = Pattern.compile("\\s*" + NOTES + "$");
069    /** "Notes:" line with new line accounted. */
070    public static final Pattern NOTES_LINE_WITH_NEWLINE = Pattern.compile("\r?\n\\s?" + NOTES);
071
072    /**
073     * Private utility constructor.
074     */
075    private ModuleJavadocParsingUtil() {
076    }
077
078    /**
079     * Gets properties of the specified module.
080     *
081     * @param moduleName name of module.
082     * @return set of properties name if present, otherwise null.
083     * @throws MacroExecutionException if the module could not be retrieved.
084     */
085    public static Set<String> getPropertyNames(String moduleName)
086            throws MacroExecutionException {
087        final Object instance = SiteUtil.getModuleInstance(moduleName);
088        final Class<?> clss = instance.getClass();
089
090        return SiteUtil.getPropertiesForDocumentation(clss, instance);
091    }
092
093    /**
094     * Determines whether the given HTML node marks the start of the "Notes" section.
095     *
096     * @param htmlElement html element to check.
097     * @return true if the element starts the "Notes" section, false otherwise.
098     */
099    private static boolean isStartOfNotesSection(DetailNode htmlElement) {
100        boolean result = false;
101        if (htmlElement != null) {
102            final DetailNode htmlContentNode = JavadocUtil.findFirstToken(
103                htmlElement, JavadocCommentsTokenTypes.HTML_CONTENT);
104
105            result = htmlContentNode != null && JavadocMetadataScraperUtil.isChildNodeTextMatches(
106                htmlContentNode, NOTES_LINE);
107        }
108        return result;
109    }
110
111    /**
112     * Writes the given javadoc chunk into xdoc.
113     *
114     * @param javadocPortion javadoc text.
115     * @param sink sink of the macro.
116     */
117    public static void writeOutJavadocPortion(String javadocPortion, Sink sink) {
118        final String[] javadocPortionLinesSplit = javadocPortion.split(NEWLINE
119            .replace("\r", ""), -1);
120
121        sink.rawText(javadocPortionLinesSplit[0]);
122        String lastHtmlTag = javadocPortionLinesSplit[0];
123
124        for (int index = 1; index < javadocPortionLinesSplit.length; index++) {
125            final String currentLine = javadocPortionLinesSplit[index].trim();
126            final String processedLine;
127
128            if (currentLine.isEmpty()) {
129                processedLine = NEWLINE;
130            }
131            else if (currentLine.startsWith("<")
132                && !startsWithTextFormattingHtmlTag(currentLine)) {
133
134                processedLine = INDENT_LEVEL_8 + currentLine;
135                lastHtmlTag = currentLine;
136            }
137            else if (lastHtmlTag.contains("<pre")) {
138                final String currentLineWithPreservedIndent = javadocPortionLinesSplit[index]
139                    .substring(1);
140
141                processedLine = NEWLINE + currentLineWithPreservedIndent;
142            }
143            else {
144                processedLine = INDENT_LEVEL_10 + currentLine;
145            }
146
147            sink.rawText(processedLine);
148        }
149
150    }
151
152    /**
153     * Checks if given line starts with HTML text-formatting tag.
154     *
155     * @param line line to check on.
156     * @return whether given line starts with HTML text-formatting tag.
157     */
158    public static boolean startsWithTextFormattingHtmlTag(String line) {
159        boolean result = false;
160
161        for (String tag : HTML_TEXT_FORMAT_TAGS) {
162            if (line.startsWith(tag)) {
163                result = true;
164                break;
165            }
166        }
167
168        return result;
169    }
170
171    /**
172     * Gets the description of module from module javadoc.
173     *
174     * @param moduleJavadoc module javadoc.
175     * @return module description.
176     */
177    public static String getModuleDescription(DetailNode moduleJavadoc) {
178        final DetailNode descriptionEndNode = getDescriptionEndNode(moduleJavadoc);
179        String result = "";
180        if (descriptionEndNode != null) {
181            result = JavadocMetadataScraperUtil.constructSubTreeText(moduleJavadoc,
182                    descriptionEndNode);
183        }
184        return result;
185    }
186
187    /**
188     * Gets the {@code @since} version of module from module javadoc.
189     *
190     * @param moduleJavadoc module javadoc
191     * @return module {@code @since} version. For instance, {@code 8.0}
192     */
193    public static String getModuleSinceVersion(DetailNode moduleJavadoc) {
194        final DetailNode sinceTagNode = getModuleSinceVersionTagStartNode(moduleJavadoc);
195        String result = "";
196
197        if (sinceTagNode == null) {
198            result = "";
199        }
200        else if (sinceTagNode.getFirstChild() != null) {
201            result = JavadocMetadataScraperUtil.constructSubTreeText(sinceTagNode,
202                    sinceTagNode.getFirstChild()).replace("@since ", "");
203        }
204        return result;
205    }
206
207    /**
208     * Gets the end node of the description.
209     *
210     * @param moduleJavadoc javadoc of module.
211     * @return the end index.
212     */
213    public static DetailNode getDescriptionEndNode(DetailNode moduleJavadoc) {
214        final DetailNode descriptionEndNode;
215
216        final DetailNode notesStartingNode =
217            getNotesSectionStartNode(moduleJavadoc);
218
219        if (notesStartingNode != null) {
220            descriptionEndNode = notesStartingNode.getPreviousSibling();
221        }
222        else {
223            descriptionEndNode = getNodeBeforeJavadocTags(moduleJavadoc);
224        }
225
226        return descriptionEndNode;
227    }
228
229    /**
230     * Gets the start node of the Notes section.
231     *
232     * @param moduleJavadoc javadoc of module.
233     * @return start node.
234     */
235    public static DetailNode getNotesSectionStartNode(DetailNode moduleJavadoc) {
236        DetailNode notesStartNode = null;
237        if (moduleJavadoc != null) {
238            DetailNode node = moduleJavadoc.getFirstChild();
239
240            while (node != null) {
241                if (isNotesSectionNode(node)) {
242                    notesStartNode = node;
243                    break;
244                }
245                node = node.getNextSibling();
246            }
247        }
248
249        return notesStartNode;
250    }
251
252    /**
253     * Checks if the given node is the start of the Notes section.
254     *
255     * @param node the node to check.
256     * @return true if the node is the start of the Notes section, false otherwise.
257     */
258    private static boolean isNotesSectionNode(DetailNode node) {
259        boolean result = false;
260        if (node.getType() == JavadocCommentsTokenTypes.HTML_ELEMENT) {
261            if (JavadocUtil.isTag(node, "ul")) {
262                final DetailNode htmlContentNode = JavadocUtil.findFirstToken(
263                    node, JavadocCommentsTokenTypes.HTML_CONTENT);
264                result = htmlContentNode != null
265                        && isStartOfNotesSection(htmlContentNode.getFirstChild());
266            }
267            else {
268                result = (JavadocUtil.isTag(node, "p")
269                            || JavadocUtil.isTag(node, "li"))
270                            && isStartOfNotesSection(node);
271            }
272        }
273        return result;
274    }
275
276    /**
277     * Gets the node representing the start of the {@code @since} version tag
278     * in the module's Javadoc.
279     *
280     * @param moduleJavadoc the root Javadoc node of the module
281     * @return the {@code @since} tag start node, or {@code null} if not found
282     */
283    public static DetailNode getModuleSinceVersionTagStartNode(DetailNode moduleJavadoc) {
284        DetailNode result = null;
285
286        if (moduleJavadoc != null) {
287            result = JavadocUtil.getAllNodesOfType(
288                    moduleJavadoc, JavadocCommentsTokenTypes.JAVADOC_BLOCK_TAG).stream()
289                .filter(javadocTag -> {
290                    final DetailNode firstChild = javadocTag.getFirstChild();
291                    return firstChild != null
292                            && firstChild.getType()
293                                == JavadocCommentsTokenTypes.SINCE_BLOCK_TAG;
294                })
295                .findFirst()
296                .orElse(null);
297        }
298        return result;
299    }
300
301    /**
302     * Gets the node of module's javadoc whose next sibling is a node that defines a javadoc tag.
303     *
304     * @param moduleJavadoc the root Javadoc node of the module
305     * @return the node that precedes node defining javadoc tag if present,
306     *     otherwise just the last node of module's javadoc.
307     */
308    public static DetailNode getNodeBeforeJavadocTags(DetailNode moduleJavadoc) {
309        DetailNode nodeBeforeJavadocTags = null;
310
311        if (moduleJavadoc != null) {
312            nodeBeforeJavadocTags = moduleJavadoc.getFirstChild();
313
314            if (nodeBeforeJavadocTags != null) {
315                while (nodeBeforeJavadocTags.getNextSibling() != null
316                        && nodeBeforeJavadocTags.getNextSibling().getType()
317                            != JavadocCommentsTokenTypes.JAVADOC_BLOCK_TAG) {
318
319                    nodeBeforeJavadocTags = nodeBeforeJavadocTags.getNextSibling();
320                }
321            }
322        }
323
324        return nodeBeforeJavadocTags;
325    }
326
327    /**
328     * Gets the Notes section of module from module javadoc.
329     *
330     * @param moduleJavadoc module javadoc.
331     * @return Notes section of module.
332     */
333    public static String getModuleNotes(DetailNode moduleJavadoc) {
334        final String result;
335
336        final DetailNode notesStartNode = getNotesSectionStartNode(moduleJavadoc);
337
338        if (notesStartNode == null) {
339            result = "";
340        }
341        else {
342            final DetailNode notesEndNode = getNodeBeforeJavadocTags(moduleJavadoc);
343
344            if (notesEndNode == null) {
345                result = "";
346            }
347            else {
348                final String unprocessedNotes = JavadocMetadataScraperUtil.constructSubTreeText(
349                            notesStartNode, notesEndNode);
350                result = NOTES_LINE_WITH_NEWLINE.matcher(unprocessedNotes).replaceAll("");
351            }
352        }
353
354        return result;
355    }
356
357    /**
358     * Checks whether property is to contain tokens.
359     *
360     * @param propertyField property field.
361     * @return true if property is to contain tokens, false otherwise.
362     */
363    public static boolean isPropertySpecialTokenProp(Field propertyField) {
364        boolean result = false;
365
366        if (propertyField != null) {
367            final XdocsPropertyType fieldXdocAnnotation =
368                propertyField.getAnnotation(XdocsPropertyType.class);
369
370            result = fieldXdocAnnotation != null
371                && fieldXdocAnnotation.value() == PropertyType.TOKEN_ARRAY;
372        }
373
374        return result;
375    }
376
377}