| 1 | /* |
| 2 | * Copyright 2006-2007 the original author or authors. |
| 3 | * |
| 4 | * Licensed under the Apache License, Version 2.0 (the "License"); |
| 5 | * you may not use this file except in compliance with the License. |
| 6 | * You may obtain a copy of the License at |
| 7 | * |
| 8 | * http://www.apache.org/licenses/LICENSE-2.0 |
| 9 | * |
| 10 | * Unless required by applicable law or agreed to in writing, software |
| 11 | * distributed under the License is distributed on an "AS IS" BASIS, |
| 12 | * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| 13 | * See the License for the specific language governing permissions and |
| 14 | * limitations under the License. |
| 15 | */ |
| 16 | |
| 17 | package org.springframework.batch.item.file; |
| 18 | |
| 19 | import java.io.BufferedReader; |
| 20 | import java.io.IOException; |
| 21 | import java.nio.charset.Charset; |
| 22 | |
| 23 | import org.apache.commons.logging.Log; |
| 24 | import org.apache.commons.logging.LogFactory; |
| 25 | import org.springframework.batch.item.ItemReader; |
| 26 | import org.springframework.batch.item.ReaderNotOpenException; |
| 27 | import org.springframework.batch.item.file.separator.RecordSeparatorPolicy; |
| 28 | import org.springframework.batch.item.file.separator.SimpleRecordSeparatorPolicy; |
| 29 | import org.springframework.batch.item.support.AbstractItemCountingItemStreamItemReader; |
| 30 | import org.springframework.beans.factory.InitializingBean; |
| 31 | import org.springframework.core.io.Resource; |
| 32 | import org.springframework.util.Assert; |
| 33 | import org.springframework.util.ClassUtils; |
| 34 | import org.springframework.util.StringUtils; |
| 35 | |
| 36 | /** |
| 37 | * Restartable {@link ItemReader} that reads lines from input {@link #setResource(Resource)}. Line is defined by the |
| 38 | * {@link #setRecordSeparatorPolicy(RecordSeparatorPolicy)} and mapped to item using {@link #setLineMapper(LineMapper)}. |
| 39 | * If an exception is thrown during line mapping it is rethrown as {@link FlatFileParseException} adding information |
| 40 | * about the problematic line and its line number. |
| 41 | * |
| 42 | * @author Robert Kasanicky |
| 43 | */ |
| 44 | public class FlatFileItemReader<T> extends AbstractItemCountingItemStreamItemReader<T> implements |
| 45 | ResourceAwareItemReaderItemStream<T>, InitializingBean { |
| 46 | |
| 47 | private static final Log logger = LogFactory.getLog(FlatFileItemReader.class); |
| 48 | |
| 49 | // default encoding for input files |
| 50 | public static final String DEFAULT_CHARSET = Charset.defaultCharset().name(); |
| 51 | |
| 52 | private RecordSeparatorPolicy recordSeparatorPolicy = new SimpleRecordSeparatorPolicy(); |
| 53 | |
| 54 | private Resource resource; |
| 55 | |
| 56 | private BufferedReader reader; |
| 57 | |
| 58 | private int lineCount = 0; |
| 59 | |
| 60 | private String[] comments = new String[] { "#" }; |
| 61 | |
| 62 | private boolean noInput = false; |
| 63 | |
| 64 | private String encoding = DEFAULT_CHARSET; |
| 65 | |
| 66 | private LineMapper<T> lineMapper; |
| 67 | |
| 68 | private int linesToSkip = 0; |
| 69 | |
| 70 | private LineCallbackHandler skippedLinesCallback; |
| 71 | |
| 72 | private boolean strict = true; |
| 73 | |
| 74 | private BufferedReaderFactory bufferedReaderFactory = new DefaultBufferedReaderFactory(); |
| 75 | |
| 76 | public FlatFileItemReader() { |
| 77 | setName(ClassUtils.getShortName(FlatFileItemReader.class)); |
| 78 | } |
| 79 | |
| 80 | /** |
| 81 | * In strict mode the reader will throw an exception on |
| 82 | * {@link #open(org.springframework.batch.item.ExecutionContext)} if the input resource does not exist. |
| 83 | * @param strict <code>true</code> by default |
| 84 | */ |
| 85 | public void setStrict(boolean strict) { |
| 86 | this.strict = strict; |
| 87 | } |
| 88 | |
| 89 | /** |
| 90 | * @param skippedLinesCallback will be called for each one of the initial skipped lines before any items are read. |
| 91 | */ |
| 92 | public void setSkippedLinesCallback(LineCallbackHandler skippedLinesCallback) { |
| 93 | this.skippedLinesCallback = skippedLinesCallback; |
| 94 | } |
| 95 | |
| 96 | /** |
| 97 | * Public setter for the number of lines to skip at the start of a file. Can be used if the file contains a header |
| 98 | * without useful (column name) information, and without a comment delimiter at the beginning of the lines. |
| 99 | * |
| 100 | * @param linesToSkip the number of lines to skip |
| 101 | */ |
| 102 | public void setLinesToSkip(int linesToSkip) { |
| 103 | this.linesToSkip = linesToSkip; |
| 104 | } |
| 105 | |
| 106 | /** |
| 107 | * Setter for line mapper. This property is required to be set. |
| 108 | * @param lineMapper maps line to item |
| 109 | */ |
| 110 | public void setLineMapper(LineMapper<T> lineMapper) { |
| 111 | this.lineMapper = lineMapper; |
| 112 | } |
| 113 | |
| 114 | /** |
| 115 | * Setter for the encoding for this input source. Default value is {@link #DEFAULT_CHARSET}. |
| 116 | * |
| 117 | * @param encoding a properties object which possibly contains the encoding for this input file; |
| 118 | */ |
| 119 | public void setEncoding(String encoding) { |
| 120 | this.encoding = encoding; |
| 121 | } |
| 122 | |
| 123 | /** |
| 124 | * Factory for the {@link BufferedReader} that will be used to extract lines from the file. The default is fine for |
| 125 | * plain text files, but this is a useful strategy for binary files where the standard BufferedReaader from java.io |
| 126 | * is limiting. |
| 127 | * |
| 128 | * @param bufferedReaderFactory the bufferedReaderFactory to set |
| 129 | */ |
| 130 | public void setBufferedReaderFactory(BufferedReaderFactory bufferedReaderFactory) { |
| 131 | this.bufferedReaderFactory = bufferedReaderFactory; |
| 132 | } |
| 133 | |
| 134 | /** |
| 135 | * Setter for comment prefixes. Can be used to ignore header lines as well by using e.g. the first couple of column |
| 136 | * names as a prefix. |
| 137 | * |
| 138 | * @param comments an array of comment line prefixes. |
| 139 | */ |
| 140 | public void setComments(String[] comments) { |
| 141 | this.comments = new String[comments.length]; |
| 142 | System.arraycopy(comments, 0, this.comments, 0, comments.length); |
| 143 | } |
| 144 | |
| 145 | /** |
| 146 | * Public setter for the input resource. |
| 147 | */ |
| 148 | public void setResource(Resource resource) { |
| 149 | this.resource = resource; |
| 150 | } |
| 151 | |
| 152 | /** |
| 153 | * Public setter for the recordSeparatorPolicy. Used to determine where the line endings are and do things like |
| 154 | * continue over a line ending if inside a quoted string. |
| 155 | * |
| 156 | * @param recordSeparatorPolicy the recordSeparatorPolicy to set |
| 157 | */ |
| 158 | public void setRecordSeparatorPolicy(RecordSeparatorPolicy recordSeparatorPolicy) { |
| 159 | this.recordSeparatorPolicy = recordSeparatorPolicy; |
| 160 | } |
| 161 | |
| 162 | /** |
| 163 | * @return string corresponding to logical record according to |
| 164 | * {@link #setRecordSeparatorPolicy(RecordSeparatorPolicy)} (might span multiple lines in file). |
| 165 | */ |
| 166 | @Override |
| 167 | protected T doRead() throws Exception { |
| 168 | if (noInput) { |
| 169 | return null; |
| 170 | } |
| 171 | |
| 172 | String line = readLine(); |
| 173 | |
| 174 | if (line == null) { |
| 175 | return null; |
| 176 | } |
| 177 | else { |
| 178 | try { |
| 179 | return lineMapper.mapLine(line, lineCount); |
| 180 | } |
| 181 | catch (Exception ex) { |
| 182 | throw new FlatFileParseException("Parsing error at line: " + lineCount + " in resource=[" |
| 183 | + resource.getDescription() + "], input=[" + line + "]", ex, line, lineCount); |
| 184 | } |
| 185 | } |
| 186 | } |
| 187 | |
| 188 | /** |
| 189 | * @return next line (skip comments).getCurrentResource |
| 190 | */ |
| 191 | private String readLine() { |
| 192 | |
| 193 | if (reader == null) { |
| 194 | throw new ReaderNotOpenException("Reader must be open before it can be read."); |
| 195 | } |
| 196 | |
| 197 | String line = null; |
| 198 | |
| 199 | try { |
| 200 | line = this.reader.readLine(); |
| 201 | if (line == null) { |
| 202 | return null; |
| 203 | } |
| 204 | lineCount++; |
| 205 | while (isComment(line)) { |
| 206 | line = reader.readLine(); |
| 207 | if (line == null) { |
| 208 | return null; |
| 209 | } |
| 210 | lineCount++; |
| 211 | } |
| 212 | |
| 213 | line = applyRecordSeparatorPolicy(line); |
| 214 | } |
| 215 | catch (IOException e) { |
| 216 | // Prevent IOException from recurring indefinitely |
| 217 | // if client keeps catching and re-calling |
| 218 | noInput = true; |
| 219 | throw new NonTransientFlatFileException("Unable to read from resource: [" + resource + "]", e, line, |
| 220 | lineCount); |
| 221 | } |
| 222 | return line; |
| 223 | } |
| 224 | |
| 225 | private boolean isComment(String line) { |
| 226 | for (String prefix : comments) { |
| 227 | if (line.startsWith(prefix)) { |
| 228 | return true; |
| 229 | } |
| 230 | } |
| 231 | return false; |
| 232 | } |
| 233 | |
| 234 | @Override |
| 235 | protected void doClose() throws Exception { |
| 236 | lineCount = 0; |
| 237 | if (reader != null) { |
| 238 | reader.close(); |
| 239 | } |
| 240 | } |
| 241 | |
| 242 | @Override |
| 243 | protected void doOpen() throws Exception { |
| 244 | Assert.notNull(resource, "Input resource must be set"); |
| 245 | Assert.notNull(recordSeparatorPolicy, "RecordSeparatorPolicy must be set"); |
| 246 | |
| 247 | noInput = true; |
| 248 | if (!resource.exists()) { |
| 249 | if (strict) { |
| 250 | throw new IllegalStateException("Input resource must exist (reader is in 'strict' mode): " + resource); |
| 251 | } |
| 252 | logger.warn("Input resource does not exist " + resource.getDescription()); |
| 253 | return; |
| 254 | } |
| 255 | |
| 256 | if (!resource.isReadable()) { |
| 257 | if (strict) { |
| 258 | throw new IllegalStateException("Input resource must be readable (reader is in 'strict' mode): " |
| 259 | + resource); |
| 260 | } |
| 261 | logger.warn("Input resource is not readable " + resource.getDescription()); |
| 262 | return; |
| 263 | } |
| 264 | |
| 265 | reader = bufferedReaderFactory.create(resource, encoding); |
| 266 | for (int i = 0; i < linesToSkip; i++) { |
| 267 | String line = readLine(); |
| 268 | if (skippedLinesCallback != null) { |
| 269 | skippedLinesCallback.handleLine(line); |
| 270 | } |
| 271 | } |
| 272 | noInput = false; |
| 273 | } |
| 274 | |
| 275 | public void afterPropertiesSet() throws Exception { |
| 276 | Assert.notNull(lineMapper, "LineMapper is required"); |
| 277 | } |
| 278 | |
| 279 | @Override |
| 280 | protected void jumpToItem(int itemIndex) throws Exception { |
| 281 | for (int i = 0; i < itemIndex; i++) { |
| 282 | readLine(); |
| 283 | } |
| 284 | } |
| 285 | |
| 286 | private String applyRecordSeparatorPolicy(String line) throws IOException { |
| 287 | |
| 288 | String record = line; |
| 289 | while (line != null && !recordSeparatorPolicy.isEndOfRecord(record)) { |
| 290 | line = this.reader.readLine(); |
| 291 | if (line == null) { |
| 292 | if (StringUtils.hasText(record)) { |
| 293 | // A record was partially complete since it hasn't ended but |
| 294 | // the line is null |
| 295 | throw new FlatFileParseException("Unexpected end of file before record complete", record, lineCount); |
| 296 | } |
| 297 | else { |
| 298 | // Record has no text but it might still be post processed |
| 299 | // to something (skipping preProcess since that was already |
| 300 | // done) |
| 301 | break; |
| 302 | } |
| 303 | } |
| 304 | else { |
| 305 | lineCount++; |
| 306 | } |
| 307 | record = recordSeparatorPolicy.preProcess(record) + line; |
| 308 | } |
| 309 | |
| 310 | return recordSeparatorPolicy.postProcess(record); |
| 311 | |
| 312 | } |
| 313 | |
| 314 | } |