001////////////////////////////////////////////////////////////////////////////////
002// checkstyle: Checks Java source code for adherence to a set of rules.
003// Copyright (C) 2001-2019 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 full text contents of the file.
082     */
083    private final String fullText;
084
085    /**
086     * The lines of the file, without terminators.
087     */
088    private final String[] lines;
089
090    /**
091     * The first position of each line within the full text.
092     */
093    private int[] lineBreaks;
094
095    /**
096     * Creates a new file text representation.
097     *
098     * <p>The file will be read using the specified encoding, replacing
099     * malformed input and unmappable characters with the default
100     * replacement character.
101     *
102     * @param file the name of the file
103     * @param charsetName the encoding to use when reading the file
104     * @throws NullPointerException if the text is null
105     * @throws IOException if the file could not be read
106     */
107    public FileText(File file, String charsetName) throws IOException {
108        this.file = file;
109
110        // We use our own decoder, to be sure we have complete control
111        // about replacements.
112        final CharsetDecoder decoder;
113        try {
114            charset = Charset.forName(charsetName);
115            decoder = charset.newDecoder();
116            decoder.onMalformedInput(CodingErrorAction.REPLACE);
117            decoder.onUnmappableCharacter(CodingErrorAction.REPLACE);
118        }
119        catch (final UnsupportedCharsetException ex) {
120            final String message = "Unsupported charset: " + charsetName;
121            throw new IllegalStateException(message, ex);
122        }
123
124        fullText = readFile(file, decoder);
125
126        // Use the BufferedReader to break down the lines as this
127        // is about 30% faster than using the
128        // LINE_TERMINATOR.split(fullText, -1) method
129        try (BufferedReader reader = new BufferedReader(new StringReader(fullText))) {
130            final ArrayList<String> textLines = new ArrayList<>();
131            while (true) {
132                final String line = reader.readLine();
133                if (line == null) {
134                    break;
135                }
136                textLines.add(line);
137            }
138            lines = textLines.toArray(CommonUtil.EMPTY_STRING_ARRAY);
139        }
140    }
141
142    /**
143     * Copy constructor.
144     * @param fileText to make copy of
145     */
146    public FileText(FileText fileText) {
147        file = fileText.file;
148        charset = fileText.charset;
149        fullText = fileText.fullText;
150        lines = fileText.lines.clone();
151        if (fileText.lineBreaks == null) {
152            lineBreaks = null;
153        }
154        else {
155            lineBreaks = fileText.lineBreaks.clone();
156        }
157    }
158
159    /**
160     * Compatibility constructor.
161     *
162     * <p>This constructor reconstructs the text of the file by joining
163     * lines with linefeed characters. This process does not restore
164     * the original line terminators and should therefore be avoided.
165     *
166     * @param file the name of the file
167     * @param lines the lines of the text, without terminators
168     * @throws NullPointerException if the lines array is null
169     */
170    public FileText(File file, List<String> lines) {
171        final StringBuilder buf = new StringBuilder(1024);
172        for (final String line : lines) {
173            buf.append(line).append('\n');
174        }
175
176        this.file = file;
177        charset = null;
178        fullText = buf.toString();
179        this.lines = lines.toArray(CommonUtil.EMPTY_STRING_ARRAY);
180    }
181
182    /**
183     * Reads file using specific decoder and returns all its content as a String.
184     * @param inputFile File to read
185     * @param decoder Charset decoder
186     * @return File's text
187     * @throws IOException Unable to open or read the file
188     */
189    private static String readFile(final File inputFile, final CharsetDecoder decoder)
190            throws IOException {
191        if (!inputFile.exists()) {
192            throw new FileNotFoundException(inputFile.getPath() + " (No such file or directory)");
193        }
194        final StringBuilder buf = new StringBuilder(1024);
195        final InputStream stream = Files.newInputStream(inputFile.toPath());
196        try (Reader reader = new InputStreamReader(stream, decoder)) {
197            final char[] chars = new char[READ_BUFFER_SIZE];
198            while (true) {
199                final int len = reader.read(chars);
200                if (len == -1) {
201                    break;
202                }
203                buf.append(chars, 0, len);
204            }
205        }
206        return buf.toString();
207    }
208
209    /**
210     * Get the name of the file.
211     * @return an object containing the name of the file
212     */
213    public File getFile() {
214        return file;
215    }
216
217    /**
218     * Get the character set which was used to read the file.
219     * Will be {@code null} for a file reconstructed from its lines.
220     * @return the charset used when the file was read
221     */
222    public Charset getCharset() {
223        return charset;
224    }
225
226    /**
227     * Retrieve the full text of the file.
228     * @return the full text of the file
229     */
230    public CharSequence getFullText() {
231        return fullText;
232    }
233
234    /**
235     * Returns an array of all lines.
236     * {@code text.toLinesArray()} is equivalent to
237     * {@code text.toArray(new String[text.size()])}.
238     * @return an array of all lines of the text
239     */
240    public String[] toLinesArray() {
241        return lines.clone();
242    }
243
244    /**
245     * Find positions of line breaks in the full text.
246     * @return an array giving the first positions of each line.
247     */
248    private int[] findLineBreaks() {
249        if (lineBreaks == null) {
250            final int[] lineBreakPositions = new int[size() + 1];
251            lineBreakPositions[0] = 0;
252            int lineNo = 1;
253            final Matcher matcher = LINE_TERMINATOR.matcher(fullText);
254            while (matcher.find()) {
255                lineBreakPositions[lineNo] = matcher.end();
256                lineNo++;
257            }
258            if (lineNo < lineBreakPositions.length) {
259                lineBreakPositions[lineNo] = fullText.length();
260            }
261            lineBreaks = lineBreakPositions;
262        }
263        return lineBreaks;
264    }
265
266    /**
267     * Determine line and column numbers in full text.
268     * @param pos the character position in the full text
269     * @return the line and column numbers of this character
270     */
271    public LineColumn lineColumn(int pos) {
272        final int[] lineBreakPositions = findLineBreaks();
273        int lineNo = Arrays.binarySearch(lineBreakPositions, pos);
274        if (lineNo < 0) {
275            // we have: lineNo = -(insertion point) - 1
276            // we want: lineNo =  (insertion point) - 1
277            lineNo = -lineNo - 2;
278        }
279        final int startOfLine = lineBreakPositions[lineNo];
280        final int columnNo = pos - startOfLine;
281        // now we have lineNo and columnNo, both starting at zero.
282        return new LineColumn(lineNo + 1, columnNo);
283    }
284
285    /**
286     * Retrieves a line of the text by its number.
287     * The returned line will not contain a trailing terminator.
288     * @param lineNo the number of the line to get, starting at zero
289     * @return the line with the given number
290     */
291    public String get(final int lineNo) {
292        return lines[lineNo];
293    }
294
295    /**
296     * Counts the lines of the text.
297     * @return the number of lines in the text
298     */
299    public int size() {
300        return lines.length;
301    }
302
303}