001////////////////////////////////////////////////////////////////////////////////
002// checkstyle: Checks Java source code for adherence to a set of rules.
003// Copyright (C) 2001-2021 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.api;
021
022import java.io.BufferedReader;
023import java.io.File;
024import java.io.FileNotFoundException;
025import java.io.IOException;
026import java.io.InputStream;
027import java.io.InputStreamReader;
028import java.io.Reader;
029import java.io.StringReader;
030import java.nio.charset.Charset;
031import java.nio.charset.CharsetDecoder;
032import java.nio.charset.CodingErrorAction;
033import java.nio.charset.UnsupportedCharsetException;
034import java.nio.file.Files;
035import java.util.ArrayList;
036import java.util.Arrays;
037import java.util.List;
038import java.util.regex.Matcher;
039import java.util.regex.Pattern;
040
041import com.puppycrawl.tools.checkstyle.utils.CommonUtil;
042
043/**
044 * Represents the text contents of a file of arbitrary plain text type.
045 * <p>
046 * This class will be passed to instances of class FileSetCheck by
047 * Checker.
048 * </p>
049 *
050 */
051public final class FileText {
052
053    /**
054     * The number of characters to read in one go.
055     */
056    private static final int READ_BUFFER_SIZE = 1024;
057
058    /**
059     * Regular expression pattern matching all line terminators.
060     */
061    private static final Pattern LINE_TERMINATOR = Pattern.compile("\\n|\\r\\n?");
062
063    // For now, we always keep both full text and lines array.
064    // In the long run, however, the one passed at initialization might be
065    // enough, while the other could be lazily created when requested.
066    // This would save memory but cost CPU cycles.
067
068    /**
069     * The name of the file.
070     * {@code null} if no file name is available for whatever reason.
071     */
072    private final File file;
073
074    /**
075     * The charset used to read the file.
076     * {@code null} if the file was reconstructed from a list of lines.
077     */
078    private final Charset charset;
079
080    /**
081     * The lines of the file, without terminators.
082     */
083    private final String[] lines;
084
085    /**
086     * The full text contents of the file.
087     *
088     * <p>Field is not final to ease reaching full test coverage.
089     *
090     * @noinspection FieldMayBeFinal
091     */
092    private String fullText;
093
094    /**
095     * The first position of each line within the full text.
096     */
097    private int[] lineBreaks;
098
099    /**
100     * Copy constructor.
101     *
102     * @param fileText to make copy of
103     */
104    public FileText(FileText fileText) {
105        file = fileText.file;
106        charset = fileText.charset;
107        fullText = fileText.fullText;
108        lines = fileText.lines.clone();
109        if (fileText.lineBreaks == null) {
110            lineBreaks = null;
111        }
112        else {
113            lineBreaks = fileText.lineBreaks.clone();
114        }
115    }
116
117    /**
118     * Compatibility constructor.
119     *
120     * <p>This constructor reconstructs the text of the file by joining
121     * lines with linefeed characters. This process does not restore
122     * the original line terminators and should therefore be avoided.
123     *
124     * @param file the name of the file
125     * @param lines the lines of the text, without terminators
126     * @throws NullPointerException if the lines array is null
127     */
128    public FileText(File file, List<String> lines) {
129        final StringBuilder buf = new StringBuilder(1024);
130        for (final String line : lines) {
131            buf.append(line).append('\n');
132        }
133
134        this.file = file;
135        charset = null;
136        fullText = buf.toString();
137        this.lines = lines.toArray(CommonUtil.EMPTY_STRING_ARRAY);
138    }
139
140    /**
141     * Creates a new file text representation.
142     *
143     * <p>The file will be read using the specified encoding, replacing
144     * malformed input and unmappable characters with the default
145     * replacement character.
146     *
147     * @param file the name of the file
148     * @param charsetName the encoding to use when reading the file
149     * @throws NullPointerException if the text is null
150     * @throws IllegalStateException if the charset is not supported.
151     * @throws IOException if the file could not be read
152     */
153    public FileText(File file, String charsetName) throws IOException {
154        this.file = file;
155
156        // We use our own decoder, to be sure we have complete control
157        // about replacements.
158        final CharsetDecoder decoder;
159        try {
160            charset = Charset.forName(charsetName);
161            decoder = charset.newDecoder();
162            decoder.onMalformedInput(CodingErrorAction.REPLACE);
163            decoder.onUnmappableCharacter(CodingErrorAction.REPLACE);
164        }
165        catch (final UnsupportedCharsetException ex) {
166            final String message = "Unsupported charset: " + charsetName;
167            throw new IllegalStateException(message, ex);
168        }
169
170        fullText = readFile(file, decoder);
171
172        // Use the BufferedReader to break down the lines as this
173        // is about 30% faster than using the
174        // LINE_TERMINATOR.split(fullText, -1) method
175        try (BufferedReader reader = new BufferedReader(new StringReader(fullText))) {
176            final ArrayList<String> textLines = new ArrayList<>();
177            while (true) {
178                final String line = reader.readLine();
179                if (line == null) {
180                    break;
181                }
182                textLines.add(line);
183            }
184            lines = textLines.toArray(CommonUtil.EMPTY_STRING_ARRAY);
185        }
186    }
187
188    /**
189     * Reads file using specific decoder and returns all its content as a String.
190     *
191     * @param inputFile File to read
192     * @param decoder Charset decoder
193     * @return File's text
194     * @throws IOException Unable to open or read the file
195     * @throws FileNotFoundException when inputFile does not exists
196     */
197    private static String readFile(final File inputFile, final CharsetDecoder decoder)
198            throws IOException {
199        if (!inputFile.exists()) {
200            throw new FileNotFoundException(inputFile.getPath() + " (No such file or directory)");
201        }
202        final StringBuilder buf = new StringBuilder(1024);
203        final InputStream stream = Files.newInputStream(inputFile.toPath());
204        try (Reader reader = new InputStreamReader(stream, decoder)) {
205            final char[] chars = new char[READ_BUFFER_SIZE];
206            while (true) {
207                final int len = reader.read(chars);
208                if (len == -1) {
209                    break;
210                }
211                buf.append(chars, 0, len);
212            }
213        }
214        return buf.toString();
215    }
216
217    /**
218     * Retrieves a line of the text by its number.
219     * The returned line will not contain a trailing terminator.
220     *
221     * @param lineNo the number of the line to get, starting at zero
222     * @return the line with the given number
223     */
224    public String get(final int lineNo) {
225        return lines[lineNo];
226    }
227
228    /**
229     * Get the name of the file.
230     *
231     * @return an object containing the name of the file
232     */
233    public File getFile() {
234        return file;
235    }
236
237    /**
238     * Get the character set which was used to read the file.
239     * Will be {@code null} for a file reconstructed from its lines.
240     *
241     * @return the charset used when the file was read
242     */
243    public Charset getCharset() {
244        return charset;
245    }
246
247    /**
248     * Retrieve the full text of the file.
249     *
250     * @return the full text of the file
251     */
252    public CharSequence getFullText() {
253        return fullText;
254    }
255
256    /**
257     * Returns an array of all lines.
258     * {@code text.toLinesArray()} is equivalent to
259     * {@code text.toArray(new String[text.size()])}.
260     *
261     * @return an array of all lines of the text
262     */
263    public String[] toLinesArray() {
264        return lines.clone();
265    }
266
267    /**
268     * Determine line and column numbers in full text.
269     *
270     * @param pos the character position in the full text
271     * @return the line and column numbers of this character
272     */
273    public LineColumn lineColumn(int pos) {
274        final int[] lineBreakPositions = findLineBreaks();
275        int lineNo = Arrays.binarySearch(lineBreakPositions, pos);
276        if (lineNo < 0) {
277            // we have: lineNo = -(insertion point) - 1
278            // we want: lineNo =  (insertion point) - 1
279            lineNo = -lineNo - 2;
280        }
281        final int startOfLine = lineBreakPositions[lineNo];
282        final int columnNo = pos - startOfLine;
283        // now we have lineNo and columnNo, both starting at zero.
284        return new LineColumn(lineNo + 1, columnNo);
285    }
286
287    /**
288     * Find positions of line breaks in the full text.
289     *
290     * @return an array giving the first positions of each line.
291     */
292    private int[] findLineBreaks() {
293        if (lineBreaks == null) {
294            final int[] lineBreakPositions = new int[size() + 1];
295            lineBreakPositions[0] = 0;
296            int lineNo = 1;
297            final Matcher matcher = LINE_TERMINATOR.matcher(fullText);
298            while (matcher.find()) {
299                lineBreakPositions[lineNo] = matcher.end();
300                lineNo++;
301            }
302            if (lineNo < lineBreakPositions.length) {
303                lineBreakPositions[lineNo] = fullText.length();
304            }
305            lineBreaks = lineBreakPositions;
306        }
307        return lineBreaks;
308    }
309
310    /**
311     * Counts the lines of the text.
312     *
313     * @return the number of lines in the text
314     */
315    public int size() {
316        return lines.length;
317    }
318
319}