官网:https://www.csvreader.com/ C# 和 VB.Net版本
java 版本：https://sourceforge.net/projects/javacsv/ 或直接点此连接下载
https://nchc.dl.sourceforge.net/project/javacsv/JavaCsv/JavaCsv%202.1/javacsv2.1.zip
废话不多说，直接上使用demo！
import lombok.extern.slf4j.Slf4j;
import java.nio.charset.Charset;
import java.util.ArrayList;
import java.util.List;
/**
* @Author changle
* @Time 17/5/4.
* @Desc 查找遗漏记录
*/
@Slf4j
public class ImgFindTest {
public static final char COMMA = ',';
public static void main(String[] args){
findOutRow04();
}
static void findOutRow01(){
String fileOrigin = "/var/tmp/files/findOut/img-list01.csv";
String fileSuccess = "/var/tmp/files/findOut/img-list-20170503_01.success.csv";
String fileNull = "/var/tmp/files/findOut/img-list-oss-null20170503_01.success.csv";
String fileMissing = "/var/tmp/files/findOut/img-list-20170503_01-missing.csv";
findOutRow(fileOrigin, fileSuccess, fileNull, fileMissing);
}
static void findOutRow(String fileOrigin, String fileSuccess, String fileNull, String fileMissing){
try {
CsvReader readerOrigin = new CsvReader(fileOrigin, COMMA, Charset.forName("UTF-8"));
CsvReader readerSuccess = new CsvReader(fileSuccess, COMMA, Charset.forName("UTF-8"));
CsvReader readerNull = new CsvReader(fileNull, COMMA, Charset.forName("UTF-8"));
CsvWriter writer = new CsvWriter(fileMissing, COMMA, Charset.forName("UTF-8"));
boolean headerOk = readerNull.readHeaders();
boolean headerOk2 = readerSuccess.readHeaders();
boolean headerOk3 = readerOrigin.readHeaders();
String[] header ={"id","outer_id","main_image","oss_file_id"};
//构造写文件的文件头信息
writer.writeRecord(header);
int countAll=0;
//加载成功的记录id
List<String> idsSuccess = new ArrayList<>();
while(readerSuccess.readRecord()){
idsSuccess.add(readerSuccess.get(0));
}
//加载失败的记录id
List<String> idsNull = new ArrayList<>();
while(readerNull.readRecord()){
idsNull.add(readerNull.get(0));
}
//寻找丢失了的源文件的记录id
while(readerOrigin.readRecord()){
String id = readerOrigin.get(0);
if(idsSuccess.contains(id)||idsNull.contains(id)){
continue;
}else{
countAll++;
log.info("找不到的ID,{}",id);
writer.writeRecord(buildNewRecord(readerOrigin.getValues(),""));
}
}
log.info("共找到丢失id{}条",countAll);
readerOrigin.close();
readerSuccess.close();
readerNull.close();
writer.flush();
writer.close();
}catch(Exception e){
log.error("发生错误",e);
}
}
public static String[] buildNewRecord(String[] oldStrarr, String ossFileId){
return new String[]{oldStrarr[0],oldStrarr[1],oldStrarr[2],ossFileId};
}
}
如果懒得下载，或下载链接失效，请直接使用以下CsvReader.java 和 CsvWriter.java 源码：
CsvReader.java源码：
/*
* Java CSV is a stream based library for reading and writing
* CSV and other delimited data.
*
* Copyright (C) Bruce Dunwiddie bruce@csvreader.com
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with this library; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA
*/
import java.io.BufferedReader;
import java.io.File;
import java.io.FileInputStream;
import java.io.FileNotFoundException;
import java.io.IOException;
import java.io.InputStream;
import java.io.InputStreamReader;
import java.io.Reader;
import java.io.StringReader;
import java.nio.charset.Charset;
import java.text.NumberFormat;
import java.util.HashMap;
/**
* A stream based parser for parsing delimited text data from a file or a
* stream.
*/
public class CsvReader {
private Reader inputStream = null;
private String fileName = null;
// this holds all the values for switches that the user is allowed to set
private UserSettings userSettings = new UserSettings();
private Charset charset = null;
private boolean useCustomRecordDelimiter = false;
// this will be our working buffer to hold data chunks
// read in from the data file
private DataBuffer dataBuffer = new DataBuffer();
private ColumnBuffer columnBuffer = new ColumnBuffer();
private RawRecordBuffer rawBuffer = new RawRecordBuffer();
private boolean[] isQualified = null;
private String rawRecord = "";
private HeadersHolder headersHolder = new HeadersHolder();
// these are all more or less global loop variables
// to keep from needing to pass them all into various
// methods during parsing
private boolean startedColumn = false;
private boolean startedWithQualifier = false;
private boolean hasMoreData = true;
private char lastLetter = '\0';
private boolean hasReadNextLine = false;
private int columnsCount = 0;
private long currentRecord = 0;
private String[] values = new String[StaticSettings.INITIAL_COLUMN_COUNT];
private boolean initialized = false;
private boolean closed = false;
/**
* Double up the text qualifier to represent an occurance of the text
* qualifier.
*/
public static final int ESCAPE_MODE_DOUBLED = 1;
/**
* Use a backslash character before the text qualifier to represent an
* occurance of the text qualifier.
*/
public static final int ESCAPE_MODE_BACKSLASH = 2;
/**
* Creates a {@link com.csvreader.CsvReader CsvReader} object using a file
* as the data source.
*
* @param fileName
* The path to the file to use as the data source.
* @param delimiter
* The character to use as the column delimiter.
* @param charset
* The {@link Charset Charset} to use while
* parsing the data.
*/
public CsvReader(String fileName, char delimiter, Charset charset)
throws FileNotFoundException {
if (fileName == null) {
throw new IllegalArgumentException(
"Parameter fileName can not be null.");
}
if (charset == null) {
throw new IllegalArgumentException(
"Parameter charset can not be null.");
}
if (!new File(fileName).exists()) {
throw new FileNotFoundException("File " + fileName
+ " does not exist.");
}
this.fileName = fileName;
this.userSettings.Delimiter = delimiter;
this.charset = charset;
isQualified = new boolean[values.length];
}
/**
* Creates a {@link com.csvreader.CsvReader CsvReader} object using a file
* as the data source. Uses ISO-8859-1 as the
* {@link Charset Charset}.
*
* @param fileName
* The path to the file to use as the data source.
* @param delimiter
* The character to use as the column delimiter.
*/
public CsvReader(String fileName, char delimiter)
throws FileNotFoundException {
this(fileName, delimiter, Charset.forName("ISO-8859-1"));
}
/**
* Creates a {@link com.csvreader.CsvReader CsvReader} object using a file
* as the data source. Uses a comma as the column delimiter and
* ISO-8859-1 as the {@link Charset Charset}.
*
* @param fileName
* The path to the file to use as the data source.
*/
public CsvReader(String fileName) throws FileNotFoundException {
this(fileName, Letters.COMMA);
}
/**
* Constructs a {@link com.csvreader.CsvReader CsvReader} object using a
* {@link Reader Reader} object as the data source.
*
* @param inputStream
* The stream to use as the data source.
* @param delimiter
* The character to use as the column delimiter.
*/
public CsvReader(Reader inputStream, char delimiter) {
if (inputStream == null) {
throw new IllegalArgumentException(
"Parameter inputStream can not be null.");
}
this.inputStream = inputStream;
this.userSettings.Delimiter = delimiter;
initialized = true;
isQualified = new boolean[values.length];
}
/**
* Constructs a {@link com.csvreader.CsvReader CsvReader} object using a
* {@link Reader Reader} object as the data source. Uses a
* comma as the column delimiter.
*
* @param inputStream
* The stream to use as the data source.
*/
public CsvReader(Reader inputStream) {
this(inputStream, Letters.COMMA);
}
/**
* Constructs a {@link com.csvreader.CsvReader CsvReader} object using an
* {@link InputStream InputStream} object as the data source.
*
* @param inputStream
* The stream to use as the data source.
* @param delimiter
* The character to use as the column delimiter.
* @param charset
* The {@link Charset Charset} to use while
* parsing the data.
*/
public CsvReader(InputStream inputStream, char delimiter, Charset charset) {
this(new InputStreamReader(inputStream, charset), delimiter);
}
/**
* Constructs a {@link com.csvreader.CsvReader CsvReader} object using an
* {@link InputStream InputStream} object as the data
* source. Uses a comma as the column delimiter.
*
* @param inputStream
* The stream to use as the data source.
* @param charset
* The {@link Charset Charset} to use while
* parsing the data.
*/
public CsvReader(InputStream inputStream, Charset charset) {
this(new InputStreamReader(inputStream, charset));
}
public boolean getCaptureRawRecord() {
return userSettings.CaptureRawRecord;
}
public void setCaptureRawRecord(boolean captureRawRecord) {
userSettings.CaptureRawRecord = captureRawRecord;
}
public String getRawRecord() {
return rawRecord;
}
/**
* Gets whether leading and trailing whitespace characters are being trimmed
* from non-textqualified column data. Default is true.
*
* @return Whether leading and trailing whitespace characters are being
* trimmed from non-textqualified column data.
*/
public boolean getTrimWhitespace() {
return userSettings.TrimWhitespace;
}
/**
* Sets whether leading and trailing whitespace characters should be trimmed
* from non-textqualified column data or not. Default is true.
*
* @param trimWhitespace
* Whether leading and trailing whitespace characters should be
* trimmed from non-textqualified column data or not.
*/
public void setTrimWhitespace(boolean trimWhitespace) {
userSettings.TrimWhitespace = trimWhitespace;
}
/**
* Gets the character being used as the column delimiter. Default is comma,
* ','.
*
* @return The character being used as the column delimiter.
*/
public char getDelimiter() {
return userSettings.Delimiter;
}
/**
* Sets the character to use as the column delimiter. Default is comma, ','.
*
* @param delimiter
* The character to use as the column delimiter.
*/
public void setDelimiter(char delimiter) {
userSettings.Delimiter = delimiter;
}
public char getRecordDelimiter() {
return userSettings.RecordDelimiter;
}
/**
* Sets the character to use as the record delimiter.
*
* @param recordDelimiter
* The character to use as the record delimiter. Default is
* combination of standard end of line characters for Windows,
* Unix, or Mac.
*/
public void setRecordDelimiter(char recordDelimiter) {
useCustomRecordDelimiter = true;
userSettings.RecordDelimiter = recordDelimiter;
}
/**
* Gets the character to use as a text qualifier in the data.
*
* @return The character to use as a text qualifier in the data.
*/
public char getTextQualifier() {
return userSettings.TextQualifier;
}
/**
* Sets the character to use as a text qualifier in the data.
*
* @param textQualifier
* The character to use as a text qualifier in the data.
*/
public void setTextQualifier(char textQualifier) {
userSettings.TextQualifier = textQualifier;
}
/**
* Whether text qualifiers will be used while parsing or not.
*
* @return Whether text qualifiers will be used while parsing or not.
*/
public boolean getUseTextQualifier() {
return userSettings.UseTextQualifier;
}
/**
* Sets whether text qualifiers will be used while parsing or not.
*
* @param useTextQualifier
* Whether to use a text qualifier while parsing or not.
*/
public void setUseTextQualifier(boolean useTextQualifier) {
userSettings.UseTextQualifier = useTextQualifier;
}
/**
* Gets the character being used as a comment signal.
*
* @return The character being used as a comment signal.
*/
public char getComment() {
return userSettings.Comment;
}
/**
* Sets the character to use as a comment signal.
*
* @param comment
* The character to use as a comment signal.
*/
public void setComment(char comment) {
userSettings.Comment = comment;
}
/**
* Gets whether comments are being looked for while parsing or not.
*
* @return Whether comments are being looked for while parsing or not.
*/
public boolean getUseComments() {
return userSettings.UseComments;
}
/**
* Sets whether comments are being looked for while parsing or not.
*
* @param useComments
* Whether comments are being looked for while parsing or not.
*/
public void setUseComments(boolean useComments) {
userSettings.UseComments = useComments;
}
/**
* Gets the current way to escape an occurance of the text qualifier inside
* qualified data.
*
* @return The current way to escape an occurance of the text qualifier
* inside qualified data.
*/
public int getEscapeMode() {
return userSettings.EscapeMode;
}
/**
* Sets the current way to escape an occurance of the text qualifier inside
* qualified data.
*
* @param escapeMode
* The way to escape an occurance of the text qualifier inside
* qualified data.
* @exception IllegalArgumentException
* When an illegal value is specified for escapeMode.
*/
public void setEscapeMode(int escapeMode) throws IllegalArgumentException {
if (escapeMode != ESCAPE_MODE_DOUBLED
&& escapeMode != ESCAPE_MODE_BACKSLASH) {
throw new IllegalArgumentException(
"Parameter escapeMode must be a valid value.");
}
userSettings.EscapeMode = escapeMode;
}
public boolean getSkipEmptyRecords() {
return userSettings.SkipEmptyRecords;
}
public void setSkipEmptyRecords(boolean skipEmptyRecords) {
userSettings.SkipEmptyRecords = skipEmptyRecords;
}
/**
* Safety caution to prevent the parser from using large amounts of memory
* in the case where parsing settings like file encodings don't end up
* matching the actual format of a file. This switch can be turned off if
* the file format is known and tested. With the switch off, the max column
* lengths and max column count per record supported by the parser will
* greatly increase. Default is true.
*
* @return The current setting of the safety switch.
*/
public boolean getSafetySwitch() {
return userSettings.SafetySwitch;
}
/**
* Safety caution to prevent the parser from using large amounts of memory
* in the case where parsing settings like file encodings don't end up
* matching the actual format of a file. This switch can be turned off if
* the file format is known and tested. With the switch off, the max column
* lengths and max column count per record supported by the parser will
* greatly increase. Default is true.
*
* @param safetySwitch
*/
public void setSafetySwitch(boolean safetySwitch) {
userSettings.SafetySwitch = safetySwitch;
}
/**
* Gets the count of columns found in this record.
*
* @return The count of columns found in this record.
*/
public int getColumnCount() {
return columnsCount;
}
/**
* Gets the index of the current record.
*
* @return The index of the current record.
*/
public long getCurrentRecord() {
return currentRecord - 1;
}
/**
* Gets the count of headers read in by a previous call to
* {@link com.csvreader.CsvReader#readHeaders readHeaders()}.
*
* @return The count of headers read in by a previous call to
* {@link com.csvreader.CsvReader#readHeaders readHeaders()}.
*/
public int getHeaderCount() {
return headersHolder.Length;
}
/**
* Returns the header values as a string array.
*
* @return The header values as a String array.
* @exception IOException
* Thrown if this object has already been closed.
*/
public String[] getHeaders() throws IOException {
checkClosed();
if (headersHolder.Headers == null) {
return null;
} else {
// use clone here to prevent the outside code from
// setting values on the array directly, which would
// throw off the index lookup based on header name
String[] clone = new String[headersHolder.Length];
System.arraycopy(headersHolder.Headers, 0, clone, 0,
headersHolder.Length);
return clone;
}
}
public void setHeaders(String[] headers) {
headersHolder.Headers = headers;
headersHolder.IndexByName.clear();
if (headers != null) {
headersHolder.Length = headers.length;
} else {
headersHolder.Length = 0;
}
// use headersHolder.Length here in case headers is null
for (int i = 0; i < headersHolder.Length; i++) {
headersHolder.IndexByName.put(headers[i], new Integer(i));
}
}
public String[] getValues() throws IOException {
checkClosed();
// need to return a clone, and can't use clone because values.Length
// might be greater than columnsCount
String[] clone = new String[columnsCount];
System.arraycopy(values, 0, clone, 0, columnsCount);
return clone;
}
/**
* Returns the current column value for a given column index.
*
* @param columnIndex
* The index of the column.
* @return The current column value.
* @exception IOException
* Thrown if this object has already been closed.
*/
public String get(int columnIndex) throws IOException {
checkClosed();
if (columnIndex > -1 && columnIndex < columnsCount) {
return values[columnIndex];
} else {
return "";
}
}
/**
* Returns the current column value for a given column header name.
*
* @param headerName
* The header name of the column.
* @return The current column value.
* @exception IOException
* Thrown if this object has already been closed.
*/
public String get(String headerName) throws IOException {
checkClosed();
return get(getIndex(headerName));
}
/**
* Creates a {@link com.csvreader.CsvReader CsvReader} object using a string
* of data as the source. Uses ISO-8859-1 as the
* {@link Charset Charset}.
*
* @param data
* The String of data to use as the source.
* @return A {@link com.csvreader.CsvReader CsvReader} object using the
* String of data as the source.
*/
public static CsvReader parse(String data) {
if (data == null) {
throw new IllegalArgumentException(
"Parameter data can not be null.");
}
return new CsvReader(new StringReader(data));
}
/**
* Reads another record.
*
* @return Whether another record was successfully read or not.
* @exception IOException
* Thrown if an error occurs while reading data from the
* source stream.
*/
public boolean readRecord() throws IOException {
checkClosed();
columnsCount = 0;
rawBuffer.Position = 0;
dataBuffer.LineStart = dataBuffer.Position;
hasReadNextLine = false;
// check to see if we've already found the end of data
if (hasMoreData) {
// loop over the data stream until the end of data is found
// or the end of the record is found
do {
if (dataBuffer.Position == dataBuffer.Count) {
checkDataLength();
} else {
startedWithQualifier = false;
// grab the current letter as a char
char currentLetter = dataBuffer.Buffer[dataBuffer.Position];
if (userSettings.UseTextQualifier
&& currentLetter == userSettings.TextQualifier) {
// this will be a text qualified column, so
// we need to set startedWithQualifier to make it
// enter the seperate branch to handle text
// qualified columns
lastLetter = currentLetter;
// read qualified
startedColumn = true;
dataBuffer.ColumnStart = dataBuffer.Position + 1;
startedWithQualifier = true;
boolean lastLetterWasQualifier = false;
char escapeChar = userSettings.TextQualifier;
if (userSettings.EscapeMode == ESCAPE_MODE_BACKSLASH) {
escapeChar = Letters.BACKSLASH;
}
boolean eatingTrailingJunk = false;
boolean lastLetterWasEscape = false;
boolean readingComplexEscape = false;
int escape = ComplexEscape.UNICODE;
int escapeLength = 0;
char escapeValue = (char) 0;
dataBuffer.Position++;
do {
if (dataBuffer.Position == dataBuffer.Count) {
checkDataLength();
} else {
// grab the current letter as a char
currentLetter = dataBuffer.Buffer[dataBuffer.Position];
if (eatingTrailingJunk) {
dataBuffer.ColumnStart = dataBuffer.Position + 1;
if (currentLetter == userSettings.Delimiter) {
endColumn();
} else if ((!useCustomRecordDelimiter && (currentLetter == Letters.CR || currentLetter == Letters.LF))
|| (useCustomRecordDelimiter && currentLetter == userSettings.RecordDelimiter)) {
endColumn();
endRecord();
}
} else if (readingComplexEscape) {
escapeLength++;
switch (escape) {
case ComplexEscape.UNICODE:
escapeValue *= (char) 16;
escapeValue += hexToDec(currentLetter);
if (escapeLength == 4) {
readingComplexEscape = false;
}
break;
case ComplexEscape.OCTAL:
escapeValue *= (char) 8;
escapeValue += (char) (currentLetter - '0');
if (escapeLength == 3) {
readingComplexEscape = false;
}
break;
case ComplexEscape.DECIMAL:
escapeValue *= (char) 10;
escapeValue += (char) (currentLetter - '0');
if (escapeLength == 3) {
readingComplexEscape = false;
}
break;
case ComplexEscape.HEX:
escapeValue *= (char) 16;
escapeValue += hexToDec(currentLetter);
if (escapeLength == 2) {
readingComplexEscape = false;
}
break;
}
if (!readingComplexEscape) {
appendLetter(escapeValue);
} else {
dataBuffer.ColumnStart = dataBuffer.Position + 1;
}
} else if (currentLetter == userSettings.TextQualifier) {
if (lastLetterWasEscape) {
lastLetterWasEscape = false;
lastLetterWasQualifier = false;
} else {
updateCurrentValue();
if (userSettings.EscapeMode == ESCAPE_MODE_DOUBLED) {
lastLetterWasEscape = true;
}
lastLetterWasQualifier = true;
}
} else if (userSettings.EscapeMode == ESCAPE_MODE_BACKSLASH
&& lastLetterWasEscape) {
switch (currentLetter) {
case 'n':
appendLetter(Letters.LF);
break;
case 'r':
appendLetter(Letters.CR);
break;
case 't':
appendLetter(Letters.TAB);
break;
case 'b':
appendLetter(Letters.BACKSPACE);
break;
case 'f':
appendLetter(Letters.FORM_FEED);
break;
case 'e':
appendLetter(Letters.ESCAPE);
break;
case 'v':
appendLetter(Letters.VERTICAL_TAB);
break;
case 'a':
appendLetter(Letters.ALERT);
break;
case '0':
case '1':
case '2':
case '3':
case '4':
case '5':
case '6':
case '7':
escape = ComplexEscape.OCTAL;
readingComplexEscape = true;
escapeLength = 1;
escapeValue = (char) (currentLetter - '0');
dataBuffer.ColumnStart = dataBuffer.Position + 1;
break;
case 'u':
case 'x':
case 'o':
case 'd':
case 'U':
case 'X':
case 'O':
case 'D':
switch (currentLetter) {
case 'u':
case 'U':
escape = ComplexEscape.UNICODE;
break;
case 'x':
case 'X':
escape = ComplexEscape.HEX;
break;
case 'o':
case 'O':
escape = ComplexEscape.OCTAL;
break;
case 'd':
case 'D':
escape = ComplexEscape.DECIMAL;
break;
}
readingComplexEscape = true;
escapeLength = 0;
escapeValue = (char) 0;
dataBuffer.ColumnStart = dataBuffer.Position + 1;
break;
default:
break;
}
lastLetterWasEscape = false;
// can only happen for ESCAPE_MODE_BACKSLASH
} else if (currentLetter == escapeChar) {
updateCurrentValue();
lastLetterWasEscape = true;
} else {
if (lastLetterWasQualifier) {
if (currentLetter == userSettings.Delimiter) {
endColumn();
} else if ((!useCustomRecordDelimiter && (currentLetter == Letters.CR || currentLetter == Letters.LF))
|| (useCustomRecordDelimiter && currentLetter == userSettings.RecordDelimiter)) {
endColumn();
endRecord();
} else {
dataBuffer.ColumnStart = dataBuffer.Position + 1;
eatingTrailingJunk = true;
}
// make sure to clear the flag for next
// run of the loop
lastLetterWasQualifier = false;
}
}
// keep track of the last letter because we need
// it for several key decisions
lastLetter = currentLetter;
if (startedColumn) {
dataBuffer.Position++;
if (userSettings.SafetySwitch
&& dataBuffer.Position
- dataBuffer.ColumnStart
+ columnBuffer.Position > 100000) {
close();
throw new IOException(
"Maximum column length of 100,000 exceeded in column "
+ NumberFormat
.getIntegerInstance()
.format(
columnsCount)
+ " in record "
+ NumberFormat
.getIntegerInstance()
.format(
currentRecord)
+ ". Set the SafetySwitch property to false"
+ " if you're expecting column lengths greater than 100,000 characters to"
+ " avoid this error.");
}
}
} // end else
} while (hasMoreData && startedColumn);
} else if (currentLetter == userSettings.Delimiter) {
// we encountered a column with no data, so
// just send the end column
lastLetter = currentLetter;
endColumn();
} else if (useCustomRecordDelimiter
&& currentLetter == userSettings.RecordDelimiter) {
// this will skip blank lines
if (startedColumn || columnsCount > 0
|| !userSettings.SkipEmptyRecords) {
endColumn();
endRecord();
} else {
dataBuffer.LineStart = dataBuffer.Position + 1;
}
lastLetter = currentLetter;
} else if (!useCustomRecordDelimiter
&& (currentLetter == Letters.CR || currentLetter == Letters.LF)) {
// this will skip blank lines
if (startedColumn
|| columnsCount > 0
|| (!userSettings.SkipEmptyRecords && (currentLetter == Letters.CR || lastLetter != Letters.CR))) {
endColumn();
endRecord();
} else {
dataBuffer.LineStart = dataBuffer.Position + 1;
}
lastLetter = currentLetter;
} else if (userSettings.UseComments && columnsCount == 0
&& currentLetter == userSettings.Comment) {
// encountered a comment character at the beginning of
// the line so just ignore the rest of the line
lastLetter = currentLetter;
skipLine();
} else if (userSettings.TrimWhitespace
&& (currentLetter == Letters.SPACE || currentLetter == Letters.TAB)) {
// do nothing, this will trim leading whitespace
// for both text qualified columns and non
startedColumn = true;
dataBuffer.ColumnStart = dataBuffer.Position + 1;
} else {
// since the letter wasn't a special letter, this
// will be the first letter of our current column
startedColumn = true;
dataBuffer.ColumnStart = dataBuffer.Position;
boolean lastLetterWasBackslash = false;
boolean readingComplexEscape = false;
int escape = ComplexEscape.UNICODE;
int escapeLength = 0;
char escapeValue = (char) 0;
boolean firstLoop = true;
do {
if (!firstLoop
&& dataBuffer.Position == dataBuffer.Count) {
checkDataLength();
} else {
if (!firstLoop) {
// grab the current letter as a char
currentLetter = dataBuffer.Buffer[dataBuffer.Position];
}
if (!userSettings.UseTextQualifier
&& userSettings.EscapeMode == ESCAPE_MODE_BACKSLASH
&& currentLetter == Letters.BACKSLASH) {
if (lastLetterWasBackslash) {
lastLetterWasBackslash = false;
} else {
updateCurrentValue();
lastLetterWasBackslash = true;
}
} else if (readingComplexEscape) {
escapeLength++;
switch (escape) {
case ComplexEscape.UNICODE:
escapeValue *= (char) 16;
escapeValue += hexToDec(currentLetter);
if (escapeLength == 4) {
readingComplexEscape = false;
}
break;
case ComplexEscape.OCTAL:
escapeValue *= (char) 8;
escapeValue += (char) (currentLetter - '0');
if (escapeLength == 3) {
readingComplexEscape = false;
}
break;
case ComplexEscape.DECIMAL:
escapeValue *= (char) 10;
escapeValue += (char) (currentLetter - '0');
if (escapeLength == 3) {
readingComplexEscape = false;
}
break;
case ComplexEscape.HEX:
escapeValue *= (char) 16;
escapeValue += hexToDec(currentLetter);
if (escapeLength == 2) {
readingComplexEscape = false;
}
break;
}
if (!readingComplexEscape) {
appendLetter(escapeValue);
} else {
dataBuffer.ColumnStart = dataBuffer.Position + 1;
}
} else if (userSettings.EscapeMode == ESCAPE_MODE_BACKSLASH
&& lastLetterWasBackslash) {
switch (currentLetter) {
case 'n':
appendLetter(Letters.LF);
break;
case 'r':
appendLetter(Letters.CR);
break;
case 't':
appendLetter(Letters.TAB);
break;
case 'b':
appendLetter(Letters.BACKSPACE);
break;
case 'f':
appendLetter(Letters.FORM_FEED);
break;
case 'e':
appendLetter(Letters.ESCAPE);
break;
case 'v':
appendLetter(Letters.VERTICAL_TAB);
break;
case 'a':
appendLetter(Letters.ALERT);
break;
case '0':
case '1':
case '2':
case '3':
case '4':
case '5':
case '6':
case '7':
escape = ComplexEscape.OCTAL;
readingComplexEscape = true;
escapeLength = 1;
escapeValue = (char) (currentLetter - '0');
dataBuffer.ColumnStart = dataBuffer.Position + 1;
break;
case 'u':
case 'x':
case 'o':
case 'd':
case 'U':
case 'X':
case 'O':
case 'D':
switch (currentLetter) {
case 'u':
case 'U':
escape = ComplexEscape.UNICODE;
break;
case 'x':
case 'X':
escape = ComplexEscape.HEX;
break;
case 'o':
case 'O':
escape = ComplexEscape.OCTAL;
break;
case 'd':
case 'D':
escape = ComplexEscape.DECIMAL;
break;
}
readingComplexEscape = true;
escapeLength = 0;
escapeValue = (char) 0;
dataBuffer.ColumnStart = dataBuffer.Position + 1;
break;
default:
break;
}
lastLetterWasBackslash = false;
} else {
if (currentLetter == userSettings.Delimiter) {
endColumn();
} else if ((!useCustomRecordDelimiter && (currentLetter == Letters.CR || currentLetter == Letters.LF))
|| (useCustomRecordDelimiter && currentLetter == userSettings.RecordDelimiter)) {
endColumn();
endRecord();
}
}
// keep track of the last letter because we need
// it for several key decisions
lastLetter = currentLetter;
firstLoop = false;
if (startedColumn) {
dataBuffer.Position++;
if (userSettings.SafetySwitch
&& dataBuffer.Position
- dataBuffer.ColumnStart
+ columnBuffer.Position > 100000) {
close();
throw new IOException(
"Maximum column length of 100,000 exceeded in column "
+ NumberFormat
.getIntegerInstance()
.format(
columnsCount)
+ " in record "
+ NumberFormat
.getIntegerInstance()
.format(
currentRecord)
+ ". Set the SafetySwitch property to false"
+ " if you're expecting column lengths greater than 100,000 characters to"
+ " avoid this error.");
}
}
} // end else
} while (hasMoreData && startedColumn);
}
if (hasMoreData) {
dataBuffer.Position++;
}
} // end else
} while (hasMoreData && !hasReadNextLine);
// check to see if we hit the end of the file
// without processing the current record
if (startedColumn || lastLetter == userSettings.Delimiter) {
endColumn();
endRecord();
}
}
if (userSettings.CaptureRawRecord) {
if (hasMoreData) {
if (rawBuffer.Position == 0) {
rawRecord = new String(dataBuffer.Buffer,
dataBuffer.LineStart, dataBuffer.Position
- dataBuffer.LineStart - 1);
} else {
rawRecord = new String(rawBuffer.Buffer, 0,
rawBuffer.Position)
+ new String(dataBuffer.Buffer,
dataBuffer.LineStart, dataBuffer.Position
- dataBuffer.LineStart - 1);
}
} else {
// for hasMoreData to ever be false, all data would have had to
// have been
// copied to the raw buffer
rawRecord = new String(rawBuffer.Buffer, 0, rawBuffer.Position);
}
} else {
rawRecord = "";
}
return hasReadNextLine;
}
/**
* @exception IOException
* Thrown if an error occurs while reading data from the
* source stream.
*/
private void checkDataLength() throws IOException {
if (!initialized) {
if (fileName != null) {
inputStream = new BufferedReader(new InputStreamReader(
new FileInputStream(fileName), charset),
StaticSettings.MAX_FILE_BUFFER_SIZE);
}
charset = null;
initialized = true;
}
updateCurrentValue();
if (userSettings.CaptureRawRecord && dataBuffer.Count > 0) {
if (rawBuffer.Buffer.length - rawBuffer.Position < dataBuffer.Count
- dataBuffer.LineStart) {
int newLength = rawBuffer.Buffer.length
+ Math.max(dataBuffer.Count - dataBuffer.LineStart,
rawBuffer.Buffer.length);
char[] holder = new char[newLength];
System.arraycopy(rawBuffer.Buffer, 0, holder, 0,
rawBuffer.Position);
rawBuffer.Buffer = holder;
}
System.arraycopy(dataBuffer.Buffer, dataBuffer.LineStart,
rawBuffer.Buffer, rawBuffer.Position, dataBuffer.Count
- dataBuffer.LineStart);
rawBuffer.Position += dataBuffer.Count - dataBuffer.LineStart;
}
try {
dataBuffer.Count = inputStream.read(dataBuffer.Buffer, 0,
dataBuffer.Buffer.length);
} catch (IOException ex) {
close();
throw ex;
}
// if no more data could be found, set flag stating that
// the end of the data was found
if (dataBuffer.Count == -1) {
hasMoreData = false;
}
dataBuffer.Position = 0;
dataBuffer.LineStart = 0;
dataBuffer.ColumnStart = 0;
}
/**
* Read the first record of data as column headers.
*
* @return Whether the header record was successfully read or not.
* @exception IOException
* Thrown if an error occurs while reading data from the
* source stream.
*/
public boolean readHeaders() throws IOException {
boolean result = readRecord();
// copy the header data from the column array
// to the header string array
headersHolder.Length = columnsCount;
headersHolder.Headers = new String[columnsCount];
for (int i = 0; i < headersHolder.Length; i++) {
String columnValue = get(i);
headersHolder.Headers[i] = columnValue;
// if there are duplicate header names, we will save the last one
headersHolder.IndexByName.put(columnValue, new Integer(i));
}
if (result) {
currentRecord--;
}
columnsCount = 0;
return result;
}
/**
* Returns the column header value for a given column index.
*
* @param columnIndex
* The index of the header column being requested.
* @return The value of the column header at the given column index.
* @exception IOException
* Thrown if this object has already been closed.
*/
public String getHeader(int columnIndex) throws IOException {
checkClosed();
// check to see if we have read the header record yet
// check to see if the column index is within the bounds
// of our header array
if (columnIndex > -1 && columnIndex < headersHolder.Length) {
// return the processed header data for this column
return headersHolder.Headers[columnIndex];
} else {
return "";
}
}
public boolean isQualified(int columnIndex) throws IOException {
checkClosed();
if (columnIndex < columnsCount && columnIndex > -1) {
return isQualified[columnIndex];
} else {
return false;
}
}
/**
* @exception IOException
* Thrown if a very rare extreme exception occurs during
* parsing, normally resulting from improper data format.
*/
private void endColumn() throws IOException {
String currentValue = "";
// must be called before setting startedColumn = false
if (startedColumn) {
if (columnBuffer.Position == 0) {
if (dataBuffer.ColumnStart < dataBuffer.Position) {
int lastLetter = dataBuffer.Position - 1;
if (userSettings.TrimWhitespace && !startedWithQualifier) {
while (lastLetter >= dataBuffer.ColumnStart
&& (dataBuffer.Buffer[lastLetter] == Letters.SPACE || dataBuffer.Buffer[lastLetter] == Letters.TAB)) {
lastLetter--;
}
}
currentValue = new String(dataBuffer.Buffer,
dataBuffer.ColumnStart, lastLetter
- dataBuffer.ColumnStart + 1);
}
} else {
updateCurrentValue();
int lastLetter = columnBuffer.Position - 1;
if (userSettings.TrimWhitespace && !startedWithQualifier) {
while (lastLetter >= 0
&& (columnBuffer.Buffer[lastLetter] == Letters.SPACE || columnBuffer.Buffer[lastLetter] == Letters.SPACE)) {
lastLetter--;
}
}
currentValue = new String(columnBuffer.Buffer, 0,
lastLetter + 1);
}
}
columnBuffer.Position = 0;
startedColumn = false;
if (columnsCount >= 100000 && userSettings.SafetySwitch) {
close();
throw new IOException(
"Maximum column count of 100,000 exceeded in record "
+ NumberFormat.getIntegerInstance().format(
currentRecord)
+ ". Set the SafetySwitch property to false"
+ " if you're expecting more than 100,000 columns per record to"
+ " avoid this error.");
}
// check to see if our current holder array for
// column chunks is still big enough to handle another
// column chunk
if (columnsCount == values.length) {
// holder array needs to grow to be able to hold another column
int newLength = values.length * 2;
String[] holder = new String[newLength];
System.arraycopy(values, 0, holder, 0, values.length);
values = holder;
boolean[] qualifiedHolder = new boolean[newLength];
System.arraycopy(isQualified, 0, qualifiedHolder, 0,
isQualified.length);
isQualified = qualifiedHolder;
}
values[columnsCount] = currentValue;
isQualified[columnsCount] = startedWithQualifier;
currentValue = "";
columnsCount++;
}
private void appendLetter(char letter) {
if (columnBuffer.Position == columnBuffer.Buffer.length) {
int newLength = columnBuffer.Buffer.length * 2;
char[] holder = new char[newLength];
System.arraycopy(columnBuffer.Buffer, 0, holder, 0,
columnBuffer.Position);
columnBuffer.Buffer = holder;
}
columnBuffer.Buffer[columnBuffer.Position++] = letter;
dataBuffer.ColumnStart = dataBuffer.Position + 1;
}
private void updateCurrentValue() {
if (startedColumn && dataBuffer.ColumnStart < dataBuffer.Position) {
if (columnBuffer.Buffer.length - columnBuffer.Position < dataBuffer.Position
- dataBuffer.ColumnStart) {
int newLength = columnBuffer.Buffer.length
+ Math.max(
dataBuffer.Position - dataBuffer.ColumnStart,
columnBuffer.Buffer.length);
char[] holder = new char[newLength];
System.arraycopy(columnBuffer.Buffer, 0, holder, 0,
columnBuffer.Position);
columnBuffer.Buffer = holder;
}
System.arraycopy(dataBuffer.Buffer, dataBuffer.ColumnStart,
columnBuffer.Buffer, columnBuffer.Position,
dataBuffer.Position - dataBuffer.ColumnStart);
columnBuffer.Position += dataBuffer.Position
- dataBuffer.ColumnStart;
}
dataBuffer.ColumnStart = dataBuffer.Position + 1;
}
/**
* @exception IOException
* Thrown if an error occurs while reading data from the
* source stream.
*/
private void endRecord() throws IOException {
// this flag is used as a loop exit condition
// during parsing
hasReadNextLine = true;
currentRecord++;
}
/**
* Gets the corresponding column index for a given column header name.
*
* @param headerName
* The header name of the column.
* @return The column index for the given column header name. Returns
* -1 if not found.
* @exception IOException
* Thrown if this object has already been closed.
*/
public int getIndex(String headerName) throws IOException {
checkClosed();
Object indexValue = headersHolder.IndexByName.get(headerName);
if (indexValue != null) {
return ((Integer) indexValue).intValue();
} else {
return -1;
}
}
/**
* Skips the next record of data by parsing each column. Does not
* increment
* {@link com.csvreader.CsvReader#getCurrentRecord getCurrentRecord()}.
*
* @return Whether another record was successfully skipped or not.
* @exception IOException
* Thrown if an error occurs while reading data from the
* source stream.
*/
public boolean skipRecord() throws IOException {
checkClosed();
boolean recordRead = false;
if (hasMoreData) {
recordRead = readRecord();
if (recordRead) {
currentRecord--;
}
}
return recordRead;
}
/**
* Skips the next line of data using the standard end of line characters and
* does not do any column delimited parsing.
*
* @return Whether a line was successfully skipped or not.
* @exception IOException
* Thrown if an error occurs while reading data from the
* source stream.
*/
public boolean skipLine() throws IOException {
checkClosed();
// clear public column values for current line
columnsCount = 0;
boolean skippedLine = false;
if (hasMoreData) {
boolean foundEol = false;
do {
if (dataBuffer.Position == dataBuffer.Count) {
checkDataLength();
} else {
skippedLine = true;
// grab the current letter as a char
char currentLetter = dataBuffer.Buffer[dataBuffer.Position];
if (currentLetter == Letters.CR
|| currentLetter == Letters.LF) {
foundEol = true;
}
// keep track of the last letter because we need
// it for several key decisions
lastLetter = currentLetter;
if (!foundEol) {
dataBuffer.Position++;
}
} // end else
} while (hasMoreData && !foundEol);
columnBuffer.Position = 0;
dataBuffer.LineStart = dataBuffer.Position + 1;
}
rawBuffer.Position = 0;
rawRecord = "";
return skippedLine;
}
/**
* Closes and releases all related resources.
*/
public void close() {
if (!closed) {
close(true);
closed = true;
}
}
/**
*
*/
private void close(boolean closing) {
if (!closed) {
if (closing) {
charset = null;
headersHolder.Headers = null;
headersHolder.IndexByName = null;
dataBuffer.Buffer = null;
columnBuffer.Buffer = null;
rawBuffer.Buffer = null;
}
try {
if (initialized) {
inputStream.close();
}
} catch (Exception e) {
// just eat the exception
}
inputStream = null;
closed = true;
}
}
/**
* @exception IOException
* Thrown if this object has already been closed.
*/
private void checkClosed() throws IOException {
if (closed) {
throw new IOException(
"This instance of the CsvReader class has already been closed.");
}
}
/**
*
*/
protected void finalize() {
close(false);
}
private class ComplexEscape {
private static final int UNICODE = 1;
private static final int OCTAL = 2;
private static final int DECIMAL = 3;
private static final int HEX = 4;
}
private static char hexToDec(char hex) {
char result;
if (hex >= 'a') {
result = (char) (hex - 'a' + 10);
} else if (hex >= 'A') {
result = (char) (hex - 'A' + 10);
} else {
result = (char) (hex - '0');
}
return result;
}
private class DataBuffer {
public char[] Buffer;
public int Position;
// / <summary>
// / How much usable data has been read into the stream,
// / which will not always be as long as Buffer.Length.
// / </summary>
public int Count;
// / <summary>
// / The position of the cursor in the buffer when the
// / current column was started or the last time data
// / was moved out to the column buffer.
// / </summary>
public int ColumnStart;
public int LineStart;
public DataBuffer() {
Buffer = new char[StaticSettings.MAX_BUFFER_SIZE];
Position = 0;
Count = 0;
ColumnStart = 0;
LineStart = 0;
}
}
private class ColumnBuffer {
public char[] Buffer;
public int Position;
public ColumnBuffer() {
Buffer = new char[StaticSettings.INITIAL_COLUMN_BUFFER_SIZE];
Position = 0;
}
}
private class RawRecordBuffer {
public char[] Buffer;
public int Position;
public RawRecordBuffer() {
Buffer = new char[StaticSettings.INITIAL_COLUMN_BUFFER_SIZE
* StaticSettings.INITIAL_COLUMN_COUNT];
Position = 0;
}
}
private class Letters {
public static final char LF = '\n';
public static final char CR = '\r';
public static final char QUOTE = '"';
public static final char COMMA = ',';
public static final char SPACE = ' ';
public static final char TAB = '\t';
public static final char POUND = '#';
public static final char BACKSLASH = '\\';
public static final char NULL = '\0';
public static final char BACKSPACE = '\b';
public static final char FORM_FEED = '\f';
public static final char ESCAPE = '\u001B'; // ASCII/ANSI escape
public static final char VERTICAL_TAB = '\u000B';
public static final char ALERT = '\u0007';
}
private class UserSettings {
// having these as publicly accessible members will prevent
// the overhead of the method call that exists on properties
public boolean CaseSensitive;
public char TextQualifier;
public boolean TrimWhitespace;
public boolean UseTextQualifier;
public char Delimiter;
public char RecordDelimiter;
public char Comment;
public boolean UseComments;
public int EscapeMode;
public boolean SafetySwitch;
public boolean SkipEmptyRecords;
public boolean CaptureRawRecord;
public UserSettings() {
CaseSensitive = true;
TextQualifier = Letters.QUOTE;
TrimWhitespace = true;
UseTextQualifier = true;
Delimiter = Letters.COMMA;
RecordDelimiter = Letters.NULL;
Comment = Letters.POUND;
UseComments = false;
EscapeMode = CsvReader.ESCAPE_MODE_DOUBLED;
SafetySwitch = true;
SkipEmptyRecords = true;
CaptureRawRecord = true;
}
}
private class HeadersHolder {
public String[] Headers;
public int Length;
public HashMap IndexByName;
public HeadersHolder() {
Headers = null;
Length = 0;
IndexByName = new HashMap();
}
}
private class StaticSettings {
// these are static instead of final so they can be changed in unit test
// isn't visible outside this class and is only accessed once during
// CsvReader construction
public static final int MAX_BUFFER_SIZE = 1024;
public static final int MAX_FILE_BUFFER_SIZE = 4 * 1024;
public static final int INITIAL_COLUMN_COUNT = 10;
public static final int INITIAL_COLUMN_BUFFER_SIZE = 50;
}
}CsvWriter.java 源码：
/*
* Java CSV is a stream based library for reading and writing
* CSV and other delimited data.
*
* Copyright (C) Bruce Dunwiddie bruce@csvreader.com
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with this library; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA
*/
import java.io.FileOutputStream;
import java.io.IOException;
import java.io.OutputStream;
import java.io.OutputStreamWriter;
import java.io.BufferedWriter;
import java.io.Writer;
import java.nio.charset.Charset;
/**
* A stream based writer for writing delimited text data to a file or a stream.
* 每次写,都是重新覆盖写.有丢失文件内容风险,谨慎操作.
*/
public class CsvWriter {
private Writer outputStream = null;
private String fileName = null;
private boolean firstColumn = true;
private boolean useCustomRecordDelimiter = false;
private Charset charset = null;
// this holds all the values for switches that the user is allowed to set
private UserSettings userSettings = new UserSettings();
private boolean initialized = false;
private boolean closed = false;
private String systemRecordDelimiter = System.getProperty("line.separator");
/**
* Double up the text qualifier to represent an occurrence of the text
* qualifier.
*/
public static final int ESCAPE_MODE_DOUBLED = 1;
/**
* Use a backslash character before the text qualifier to represent an
* occurrence of the text qualifier.
*/
public static final int ESCAPE_MODE_BACKSLASH = 2;
/**
* Creates a {@link com.csvreader.CsvWriter CsvWriter} object using a file
* as the data destination.
*
* @param fileName
* The path to the file to output the data.
* @param delimiter
* The character to use as the column delimiter.
* @param charset
* The {@link Charset Charset} to use while
* writing the data.
*/
public CsvWriter(String fileName, char delimiter, Charset charset) {
if (fileName == null) {
throw new IllegalArgumentException("Parameter fileName can not be null.");
}
if (charset == null) {
throw new IllegalArgumentException("Parameter charset can not be null.");
}
this.fileName = fileName;
userSettings.Delimiter = delimiter;
this.charset = charset;
}
/**
* Creates a {@link com.csvreader.CsvWriter CsvWriter} object using a file
* as the data destination. Uses a comma as the column delimiter and
* ISO-8859-1 as the {@link Charset Charset}.
*
* @param fileName
* The path to the file to output the data.
*/
public CsvWriter(String fileName) {
this(fileName, Letters.COMMA, Charset.forName("ISO-8859-1"));
}
/**
* Creates a {@link com.csvreader.CsvWriter CsvWriter} object using a Writer
* to write data to.
*
* @param outputStream
* The stream to write the column delimited data to.
* @param delimiter
* The character to use as the column delimiter.
*/
public CsvWriter(Writer outputStream, char delimiter) {
if (outputStream == null) {
throw new IllegalArgumentException("Parameter outputStream can not be null.");
}
this.outputStream = outputStream;
userSettings.Delimiter = delimiter;
initialized = true;
}
/**
* Creates a {@link com.csvreader.CsvWriter CsvWriter} object using an
* OutputStream to write data to.
*
* @param outputStream
* The stream to write the column delimited data to.
* @param delimiter
* The character to use as the column delimiter.
* @param charset
* The {@link Charset Charset} to use while
* writing the data.
*/
public CsvWriter(OutputStream outputStream, char delimiter, Charset charset) {
this(new OutputStreamWriter(outputStream, charset), delimiter);
}
/**
* Gets the character being used as the column delimiter.
*
* @return The character being used as the column delimiter.
*/
public char getDelimiter() {
return userSettings.Delimiter;
}
/**
* Sets the character to use as the column delimiter.
*
* @param delimiter
* The character to use as the column delimiter.
*/
public void setDelimiter(char delimiter) {
userSettings.Delimiter = delimiter;
}
public char getRecordDelimiter() {
return userSettings.RecordDelimiter;
}
/**
* Sets the character to use as the record delimiter.
*
* @param recordDelimiter
* The character to use as the record delimiter. Default is
* combination of standard end of line characters for Windows,
* Unix, or Mac.
*/
public void setRecordDelimiter(char recordDelimiter) {
useCustomRecordDelimiter = true;
userSettings.RecordDelimiter = recordDelimiter;
}
/**
* Gets the character to use as a text qualifier in the data.
*
* @return The character to use as a text qualifier in the data.
*/
public char getTextQualifier() {
return userSettings.TextQualifier;
}
/**
* Sets the character to use as a text qualifier in the data.
*
* @param textQualifier
* The character to use as a text qualifier in the data.
*/
public void setTextQualifier(char textQualifier) {
userSettings.TextQualifier = textQualifier;
}
/**
* Whether text qualifiers will be used while writing data or not.
*
* @return Whether text qualifiers will be used while writing data or not.
*/
public boolean getUseTextQualifier() {
return userSettings.UseTextQualifier;
}
/**
* Sets whether text qualifiers will be used while writing data or not.
*
* @param useTextQualifier
* Whether to use a text qualifier while writing data or not.
*/
public void setUseTextQualifier(boolean useTextQualifier) {
userSettings.UseTextQualifier = useTextQualifier;
}
public int getEscapeMode() {
return userSettings.EscapeMode;
}
public void setEscapeMode(int escapeMode) {
userSettings.EscapeMode = escapeMode;
}
public void setComment(char comment) {
userSettings.Comment = comment;
}
public char getComment() {
return userSettings.Comment;
}
/**
* Whether fields will be surrounded by the text qualifier even if the
* qualifier is not necessarily needed to escape this field.
*
* @return Whether fields will be forced to be qualified or not.
*/
public boolean getForceQualifier() {
return userSettings.ForceQualifier;
}
/**
* Use this to force all fields to be surrounded by the text qualifier even
* if the qualifier is not necessarily needed to escape this field. Default
* is false.
*
* @param forceQualifier
* Whether to force the fields to be qualified or not.
*/
public void setForceQualifier(boolean forceQualifier) {
userSettings.ForceQualifier = forceQualifier;
}
/**
* Writes another column of data to this record.
*
* @param content
* The data for the new column.
* @param preserveSpaces
* Whether to preserve leading and trailing whitespace in this
* column of data.
* @exception IOException
* Thrown if an error occurs while writing data to the
* destination stream.
*/
public void write(String content, boolean preserveSpaces)
throws IOException {
checkClosed();
checkInit();
if (content == null) {
content = "";
}
if (!firstColumn) {
outputStream.write(userSettings.Delimiter);
}
boolean textQualify = userSettings.ForceQualifier;
if (!preserveSpaces && content.length() > 0) {
content = content.trim();
}
if (!textQualify
&& userSettings.UseTextQualifier
&& (content.indexOf(userSettings.TextQualifier) > -1
|| content.indexOf(userSettings.Delimiter) > -1
|| (!useCustomRecordDelimiter && (content
.indexOf(Letters.LF) > -1 || content
.indexOf(Letters.CR) > -1))
|| (useCustomRecordDelimiter && content
.indexOf(userSettings.RecordDelimiter) > -1)
|| (firstColumn && content.length() > 0 && content
.charAt(0) == userSettings.Comment) ||
// check for empty first column, which if on its own line must
// be qualified or the line will be skipped
(firstColumn && content.length() == 0))) {
textQualify = true;
}
if (userSettings.UseTextQualifier && !textQualify
&& content.length() > 0 && preserveSpaces) {
char firstLetter = content.charAt(0);
if (firstLetter == Letters.SPACE || firstLetter == Letters.TAB) {
textQualify = true;
}
if (!textQualify && content.length() > 1) {
char lastLetter = content.charAt(content.length() - 1);
if (lastLetter == Letters.SPACE || lastLetter == Letters.TAB) {
textQualify = true;
}
}
}
if (textQualify) {
outputStream.write(userSettings.TextQualifier);
if (userSettings.EscapeMode == ESCAPE_MODE_BACKSLASH) {
content = replace(content, "" + Letters.BACKSLASH, ""
+ Letters.BACKSLASH + Letters.BACKSLASH);
content = replace(content, "" + userSettings.TextQualifier, ""
+ Letters.BACKSLASH + userSettings.TextQualifier);
} else {
content = replace(content, "" + userSettings.TextQualifier, ""
+ userSettings.TextQualifier
+ userSettings.TextQualifier);
}
} else if (userSettings.EscapeMode == ESCAPE_MODE_BACKSLASH) {
content = replace(content, "" + Letters.BACKSLASH, ""
+ Letters.BACKSLASH + Letters.BACKSLASH);
content = replace(content, "" + userSettings.Delimiter, ""
+ Letters.BACKSLASH + userSettings.Delimiter);
if (useCustomRecordDelimiter) {
content = replace(content, "" + userSettings.RecordDelimiter,
"" + Letters.BACKSLASH + userSettings.RecordDelimiter);
} else {
content = replace(content, "" + Letters.CR, ""
+ Letters.BACKSLASH + Letters.CR);
content = replace(content, "" + Letters.LF, ""
+ Letters.BACKSLASH + Letters.LF);
}
if (firstColumn && content.length() > 0
&& content.charAt(0) == userSettings.Comment) {
if (content.length() > 1) {
content = "" + Letters.BACKSLASH + userSettings.Comment
+ content.substring(1);
} else {
content = "" + Letters.BACKSLASH + userSettings.Comment;
}
}
}
outputStream.write(content);
if (textQualify) {
outputStream.write(userSettings.TextQualifier);
}
firstColumn = false;
}
/**
* Writes another column of data to this record. Does not preserve
* leading and trailing whitespace in this column of data.
*
* @param content
* The data for the new column.
* @exception IOException
* Thrown if an error occurs while writing data to the
* destination stream.
*/
public void write(String content) throws IOException {
write(content, false);
}
public void writeComment(String commentText) throws IOException {
checkClosed();
checkInit();
outputStream.write(userSettings.Comment);
outputStream.write(commentText);
if (useCustomRecordDelimiter) {
outputStream.write(userSettings.RecordDelimiter);
} else {
outputStream.write(systemRecordDelimiter);
}
firstColumn = true;
}
/**
* Writes a new record using the passed in array of values.
*
* @param values
* Values to be written.
*
* @param preserveSpaces
* Whether to preserver leading and trailing spaces in columns
* while writing out to the record or not.
*
* @throws IOException
* Thrown if an error occurs while writing data to the
* destination stream.
*/
public void writeRecord(String[] values, boolean preserveSpaces)
throws IOException {
if (values != null && values.length > 0) {
for (int i = 0; i < values.length; i++) {
write(values[i], preserveSpaces);
}
endRecord();
}
}
/**
* Writes a new record using the passed in array of values.
*
* @param values
* Values to be written.
*
* @throws IOException
* Thrown if an error occurs while writing data to the
* destination stream.
*/
public void writeRecord(String[] values) throws IOException {
writeRecord(values, false);
}
/**
* Ends the current record by sending the record delimiter.
*
* @exception IOException
* Thrown if an error occurs while writing data to the
* destination stream.
*/
public void endRecord() throws IOException {
checkClosed();
checkInit();
if (useCustomRecordDelimiter) {
outputStream.write(userSettings.RecordDelimiter);
} else {
outputStream.write(systemRecordDelimiter);
}
firstColumn = true;
}
/**
*
*/
private void checkInit() throws IOException {
if (!initialized) {
if (fileName != null) {
outputStream = new BufferedWriter(new OutputStreamWriter(
new FileOutputStream(fileName), charset));
}
initialized = true;
}
}
/**
* Clears all buffers for the current writer and causes any buffered data to
* be written to the underlying device.
* @exception IOException
* Thrown if an error occurs while writing data to the
* destination stream.
*/
public void flush() throws IOException {
outputStream.flush();
}
/**
* Closes and releases all related resources.
*/
public void close() {
if (!closed) {
close(true);
closed = true;
}
}
/**
*
*/
private void close(boolean closing) {
if (!closed) {
if (closing) {
charset = null;
}
try {
if (initialized) {
outputStream.close();
}
} catch (Exception e) {
// just eat the exception
}
outputStream = null;
closed = true;
}
}
/**
*
*/
private void checkClosed() throws IOException {
if (closed) {
throw new IOException(
"This instance of the CsvWriter class has already been closed.");
}
}
/**
*
*/
protected void finalize() {
close(false);
}
private class Letters {
public static final char LF = '\n';
public static final char CR = '\r';
public static final char QUOTE = '"';
public static final char COMMA = ',';
public static final char SPACE = ' ';
public static final char TAB = '\t';
public static final char POUND = '#';
public static final char BACKSLASH = '\\';
public static final char NULL = '\0';
}
private class UserSettings {
// having these as publicly accessible members will prevent
// the overhead of the method call that exists on properties
public char TextQualifier;
public boolean UseTextQualifier;
public char Delimiter;
public char RecordDelimiter;
public char Comment;
public int EscapeMode;
public boolean ForceQualifier;
public UserSettings() {
TextQualifier = Letters.QUOTE;
UseTextQualifier = true;
Delimiter = Letters.COMMA;
RecordDelimiter = Letters.NULL;
Comment = Letters.POUND;
EscapeMode = ESCAPE_MODE_DOUBLED;
ForceQualifier = true;
}
}
public static String replace(String original, String pattern, String replace) {
final int len = pattern.length();
int found = original.indexOf(pattern);
if (found > -1) {
StringBuffer sb = new StringBuffer();
int start = 0;
while (found != -1) {
sb.append(original.substring(start, found));
sb.append(replace);
start = found + len;
found = original.indexOf(pattern, start);
}
sb.append(original.substring(start));
return sb.toString();
} else {
return original;
}
}
}
290
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
