001////////////////////////////////////////////////////////////////////////////////
002// checkstyle: Checks Java source code for adherence to a set of rules.
003// Copyright (C) 2001-2020 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     * @param fileText to make copy of
102     */
103    public FileText(FileText fileText) {
104        file = fileText.file;
105        charset = fileText.charset;
106        fullText = fileText.fullText;
107        lines = fileText.lines.clone();
108        if (fileText.lineBreaks == null) {
109            lineBreaks = null;
110        }
111        else {
112            lineBreaks = fileText.lineBreaks.clone();
113        }
114    }
115
116    /**
117     * Compatibility constructor.
118     *
119     * <p>This constructor reconstructs the text of the file by joining
120     * lines with linefeed characters. This process does not restore
121     * the original line terminators and should therefore be avoided.
122     *
123     * @param file the name of the file
124     * @param lines the lines of the text, without terminators
125     * @throws NullPointerException if the lines array is null
126     */
127    public FileText(File file, List<String> lines) {
128        final StringBuilder buf = new StringBuilder(1024);
129        for (final String line : lines) {
130            buf.append(line).append('\n');
131        }
132
133        this.file = file;
134        charset = null;
135        fullText = buf.toString();
136        this.lines = lines.toArray(CommonUtil.EMPTY_STRING_ARRAY);
137    }
138
139    /**
140     * Creates a new file text representation.
141     *
142     * <p>The file will be read using the specified encoding, replacing
143     * malformed input and unmappable characters with the default
144     * replacement character.
145     *
146     * @param file the name of the file
147     * @param charsetName the encoding to use when reading the file
148     * @throws NullPointerException if the text is null
149     * @throws IOException if the file could not be read
150     */
151    public FileText(File file, String charsetName) throws IOException {
152        this.file = file;
153
154        // We use our own decoder, to be sure we have complete control
155        // about replacements.
156        final CharsetDecoder decoder;
157        try {
158            charset = Charset.forName(charsetName);
159            decoder = charset.newDecoder();
160            decoder.onMalformedInput(CodingErrorAction.REPLACE);
161            decoder.onUnmappableCharacter(CodingErrorAction.REPLACE);
162        }
163        catch (final UnsupportedCharsetException ex) {
164            final String message = "Unsupported charset: " + charsetName;
165            throw new IllegalStateException(message, ex);
166        }
167
168        fullText = readFile(file, decoder);
169
170        // Use the BufferedReader to break down the lines as this
171        // is about 30% faster than using the
172        // LINE_TERMINATOR.split(fullText, -1) method
173        try (BufferedReader reader = new BufferedReader(new StringReader(fullText))) {
174            final ArrayList<String> textLines = new ArrayList<>();
175            while (true) {
176                final String line = reader.readLine();
177                if (line == null) {
178                    break;
179                }
180                textLines.add(line);
181            }
182            lines = textLines.toArray(CommonUtil.EMPTY_STRING_ARRAY);
183        }
184    }
185
186    /**
187     * Reads file using specific decoder and returns all its content as a String.
188     *
189     * @param inputFile File to read
190     * @param decoder Charset decoder
191     * @return File's text
192     * @throws IOException Unable to open or read the file
193     * @throws FileNotFoundException when inputFile does not exists
194     */
195    private static String readFile(final File inputFile, final CharsetDecoder decoder)
196            throws IOException {
197        if (!inputFile.exists()) {
198            throw new FileNotFoundException(inputFile.getPath() + " (No such file or directory)");
199        }
200        final StringBuilder buf = new StringBuilder(1024);
201        final InputStream stream = Files.newInputStream(inputFile.toPath());
202        try (Reader reader = new InputStreamReader(stream, decoder)) {
203            final char[] chars = new char[READ_BUFFER_SIZE];
204            while (true) {
205                final int len = reader.read(chars);
206                if (len == -1) {
207                    break;
208                }
209                buf.append(chars, 0, len);
210            }
211        }
212        return buf.toString();
213    }
214
215    /**
216     * Retrieves a line of the text by its number.
217     * The returned line will not contain a trailing terminator.
218     * @param lineNo the number of the line to get, starting at zero
219     * @return the line with the given number
220     */
221    public String get(final int lineNo) {
222        return lines[lineNo];
223    }
224
225    /**
226     * Get the name of the file.
227     *
228     * @return an object containing the name of the file
229     */
230    public File getFile() {
231        return file;
232    }
233
234    /**
235     * Get the character set which was used to read the file.
236     * Will be {@code null} for a file reconstructed from its lines.
237     *
238     * @return the charset used when the file was read
239     */
240    public Charset getCharset() {
241        return charset;
242    }
243
244    /**
245     * Retrieve the full text of the file.
246     *
247     * @return the full text of the file
248     */
249    public CharSequence getFullText() {
250        return fullText;
251    }
252
253    /**
254     * Returns an array of all lines.
255     * {@code text.toLinesArray()} is equivalent to
256     * {@code text.toArray(new String[text.size()])}.
257     *
258     * @return an array of all lines of the text
259     */
260    public String[] toLinesArray() {
261        return lines.clone();
262    }
263
264    /**
265     * Determine line and column numbers in full text.
266     *
267     * @param pos the character position in the full text
268     * @return the line and column numbers of this character
269     */
270    public LineColumn lineColumn(int pos) {
271        final int[] lineBreakPositions = findLineBreaks();
272        int lineNo = Arrays.binarySearch(lineBreakPositions, pos);
273        if (lineNo < 0) {
274            // we have: lineNo = -(insertion point) - 1
275            // we want: lineNo =  (insertion point) - 1
276            lineNo = -lineNo - 2;
277        }
278        final int startOfLine = lineBreakPositions[lineNo];
279        final int columnNo = pos - startOfLine;
280        // now we have lineNo and columnNo, both starting at zero.
281        return new LineColumn(lineNo + 1, columnNo);
282    }
283
284    /**
285     * Find positions of line breaks in the full text.
286     * @return an array giving the first positions of each line.
287     */
288    private int[] findLineBreaks() {
289        if (lineBreaks == null) {
290            final int[] lineBreakPositions = new int[size() + 1];
291            lineBreakPositions[0] = 0;
292            int lineNo = 1;
293            final Matcher matcher = LINE_TERMINATOR.matcher(fullText);
294            while (matcher.find()) {
295                lineBreakPositions[lineNo] = matcher.end();
296                lineNo++;
297            }
298            if (lineNo < lineBreakPositions.length) {
299                lineBreakPositions[lineNo] = fullText.length();
300            }
301            lineBreaks = lineBreakPositions;
302        }
303        return lineBreaks;
304    }
305
306    /**
307     * Counts the lines of the text.
308     *
309     * @return the number of lines in the text
310     */
311    public int size() {
312        return lines.length;
313    }
314
315}