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 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     * @noinspection FieldMayBeFinal
090     */
091    private String fullText;
092
093    /**
094     * The first position of each line within the full text.
095     */
096    private int[] lineBreaks;
097
098    /**
099     * Creates a new file text representation.
100     *
101     * <p>The file will be read using the specified encoding, replacing
102     * malformed input and unmappable characters with the default
103     * replacement character.
104     *
105     * @param file the name of the file
106     * @param charsetName the encoding to use when reading the file
107     * @throws NullPointerException if the text is null
108     * @throws IOException if the file could not be read
109     */
110    public FileText(File file, String charsetName) throws IOException {
111        this.file = file;
112
113        // We use our own decoder, to be sure we have complete control
114        // about replacements.
115        final CharsetDecoder decoder;
116        try {
117            charset = Charset.forName(charsetName);
118            decoder = charset.newDecoder();
119            decoder.onMalformedInput(CodingErrorAction.REPLACE);
120            decoder.onUnmappableCharacter(CodingErrorAction.REPLACE);
121        }
122        catch (final UnsupportedCharsetException ex) {
123            final String message = "Unsupported charset: " + charsetName;
124            throw new IllegalStateException(message, ex);
125        }
126
127        fullText = readFile(file, decoder);
128
129        // Use the BufferedReader to break down the lines as this
130        // is about 30% faster than using the
131        // LINE_TERMINATOR.split(fullText, -1) method
132        try (BufferedReader reader = new BufferedReader(new StringReader(fullText))) {
133            final ArrayList<String> textLines = new ArrayList<>();
134            while (true) {
135                final String line = reader.readLine();
136                if (line == null) {
137                    break;
138                }
139                textLines.add(line);
140            }
141            lines = textLines.toArray(CommonUtil.EMPTY_STRING_ARRAY);
142        }
143    }
144
145    /**
146     * Copy constructor.
147     * @param fileText to make copy of
148     */
149    public FileText(FileText fileText) {
150        file = fileText.file;
151        charset = fileText.charset;
152        fullText = fileText.fullText;
153        lines = fileText.lines.clone();
154        if (fileText.lineBreaks == null) {
155            lineBreaks = null;
156        }
157        else {
158            lineBreaks = fileText.lineBreaks.clone();
159        }
160    }
161
162    /**
163     * Compatibility constructor.
164     *
165     * <p>This constructor reconstructs the text of the file by joining
166     * lines with linefeed characters. This process does not restore
167     * the original line terminators and should therefore be avoided.
168     *
169     * @param file the name of the file
170     * @param lines the lines of the text, without terminators
171     * @throws NullPointerException if the lines array is null
172     */
173    public FileText(File file, List<String> lines) {
174        final StringBuilder buf = new StringBuilder(1024);
175        for (final String line : lines) {
176            buf.append(line).append('\n');
177        }
178
179        this.file = file;
180        charset = null;
181        fullText = buf.toString();
182        this.lines = lines.toArray(CommonUtil.EMPTY_STRING_ARRAY);
183    }
184
185    /**
186     * Reads file using specific decoder and returns all its content as a String.
187     * @param inputFile File to read
188     * @param decoder Charset decoder
189     * @return File's text
190     * @throws IOException Unable to open or read the file
191     */
192    private static String readFile(final File inputFile, final CharsetDecoder decoder)
193            throws IOException {
194        if (!inputFile.exists()) {
195            throw new FileNotFoundException(inputFile.getPath() + " (No such file or directory)");
196        }
197        final StringBuilder buf = new StringBuilder(1024);
198        final InputStream stream = Files.newInputStream(inputFile.toPath());
199        try (Reader reader = new InputStreamReader(stream, decoder)) {
200            final char[] chars = new char[READ_BUFFER_SIZE];
201            while (true) {
202                final int len = reader.read(chars);
203                if (len == -1) {
204                    break;
205                }
206                buf.append(chars, 0, len);
207            }
208        }
209        return buf.toString();
210    }
211
212    /**
213     * Get the name of the file.
214     * @return an object containing the name of the file
215     */
216    public File getFile() {
217        return file;
218    }
219
220    /**
221     * Get the character set which was used to read the file.
222     * Will be {@code null} for a file reconstructed from its lines.
223     * @return the charset used when the file was read
224     */
225    public Charset getCharset() {
226        return charset;
227    }
228
229    /**
230     * Retrieve the full text of the file.
231     * @return the full text of the file
232     */
233    public CharSequence getFullText() {
234        return fullText;
235    }
236
237    /**
238     * Returns an array of all lines.
239     * {@code text.toLinesArray()} is equivalent to
240     * {@code text.toArray(new String[text.size()])}.
241     * @return an array of all lines of the text
242     */
243    public String[] toLinesArray() {
244        return lines.clone();
245    }
246
247    /**
248     * Find positions of line breaks in the full text.
249     * @return an array giving the first positions of each line.
250     */
251    private int[] findLineBreaks() {
252        if (lineBreaks == null) {
253            final int[] lineBreakPositions = new int[size() + 1];
254            lineBreakPositions[0] = 0;
255            int lineNo = 1;
256            final Matcher matcher = LINE_TERMINATOR.matcher(fullText);
257            while (matcher.find()) {
258                lineBreakPositions[lineNo] = matcher.end();
259                lineNo++;
260            }
261            if (lineNo < lineBreakPositions.length) {
262                lineBreakPositions[lineNo] = fullText.length();
263            }
264            lineBreaks = lineBreakPositions;
265        }
266        return lineBreaks;
267    }
268
269    /**
270     * Determine line and column numbers in full text.
271     * @param pos the character position in the full text
272     * @return the line and column numbers of this character
273     */
274    public LineColumn lineColumn(int pos) {
275        final int[] lineBreakPositions = findLineBreaks();
276        int lineNo = Arrays.binarySearch(lineBreakPositions, pos);
277        if (lineNo < 0) {
278            // we have: lineNo = -(insertion point) - 1
279            // we want: lineNo =  (insertion point) - 1
280            lineNo = -lineNo - 2;
281        }
282        final int startOfLine = lineBreakPositions[lineNo];
283        final int columnNo = pos - startOfLine;
284        // now we have lineNo and columnNo, both starting at zero.
285        return new LineColumn(lineNo + 1, columnNo);
286    }
287
288    /**
289     * Retrieves a line of the text by its number.
290     * The returned line will not contain a trailing terminator.
291     * @param lineNo the number of the line to get, starting at zero
292     * @return the line with the given number
293     */
294    public String get(final int lineNo) {
295        return lines[lineNo];
296    }
297
298    /**
299     * Counts the lines of the text.
300     * @return the number of lines in the text
301     */
302    public int size() {
303        return lines.length;
304    }
305
306}