java 清除 bom

参考工具  http://akini.mbnet.fi/java/unicodereader/

Utf8BomRemover 清除bom的方法

1. package cn.com.do1.component.common.util;
3. import java.io.*;
4. import java.nio.charset.Charset;
6. public class Utf8BomRemover {
11. /**
12. * 读取流中前面的字符，看是否有bom，如果有bom，将bom头先读掉丢弃
13. *
14. * @param in
15. * @return
16. * @throws java.io.IOException
17. */
18. public static InputStream getInputStream(InputStream in) throws IOException {
20. PushbackInputStream testin = new PushbackInputStream(in);
21. int ch = testin.read();
22. if (ch != 0xEF) {
23. testin.unread(ch);
24. } else if ((ch = testin.read()) != 0xBB) {
25. testin.unread(ch);
26. testin.unread(0xef);
27. } else if ((ch = testin.read()) != 0xBF) {
28. throw new IOException("错误的UTF-8格式文件");
29. } else {
30. // 不需要做，这里是bom头被读完了
31. // // System.out.println("still exist bom");
32. }
33. return testin;
35. }
37. /**
38. * 根据一个文件名，读取完文件，干掉bom头。
39. *
40. * @param fileName
41. * @throws java.io.IOException
42. */
43. public static void trimBom(String fileName) throws IOException {
45. FileInputStream fin = new FileInputStream(fileName);
46. // 开始写临时文件
47. InputStream in = getInputStream(fin);
48. ByteArrayOutputStream bos = new ByteArrayOutputStream();
49. byte b[] = new byte[4096];
51. int len = 0;
52. while (in.available() > 0) {
53. len = in.read(b, 0, 4096);
54. // out.write(b, 0, len);
55. bos.write(b, 0, len);
56. }
58. in.close();
59. fin.close();
60. bos.close();
62. // 临时文件写完，开始将临时文件写回本文件。
63. System.out.println("[" + fileName + "]");
64. FileOutputStream out = new FileOutputStream(fileName);
65. out.write(bos.toByteArray());
66. out.close();
67. System.out.println("处理文件" + fileName);
68. }
72. public static void main(String[] args) throws IOException {
73. //刪除指定文件夾下（含子文件夾）所有java文件的BOM，若構造器中參數為null則刪除所有文件頭部BOM
74. new Utf8BomRemover( "java" ).start( "\"F:\\\\flwork\\\\gmmsDGYH\\\\src\\\\com");
76. }
77. /**
78. * 根据一个文件名，读取完文件，干掉bom头2 这里使用了第三方类UnicodeReader
79. *
80. * @throws java.io.IOException
81. */
82. public void saveFile(String file) throws IOException {
83. InputStream in= new FileInputStream( file);
84. BufferedReader bre = null;
85. OutputStreamWriter pw = null;//定义一个流
86. CharArrayWriter writer = new CharArrayWriter();
87. //这一句会读取BOM头
88. //bre = new BufferedReader(new InputStreamReader(in, "UTF-8"));
89. //这一句会干掉BOM头
90. bre = new BufferedReader(new UnicodeReader(in, Charset.defaultCharset().name()));//
91. String line = bre.readLine();
92. while(line != null)
93. {
94. writer.write(line);
96. /*
97. 加上这段代码可以查看更详细的16进制
98. byte[] allbytes = line.getBytes("UTF-8");
99. for (int i=0; i < allbytes.length; i++)
100. {
101. int tmp = allbytes[i];
102. String hexString = Integer.toHexString(tmp);
103. // 1个byte变成16进制的，只需要2位就可以表示了，取后面两位，去掉前面的符号填充
104. hexString = hexString.substring(hexString.length() -2);
105. System.out.print(hexString.toUpperCase());
106. System.out.print(" ");
107. }*/
110. line = bre.readLine();
111. }
112. writer.flush();
113. bre.close();
114. FileWriter f2 = new FileWriter(file);
115. writer.writeTo(f2);
116. f2.close();
117. writer.close();
118. }
119. private String extension = null ;
120. public Utf8BomRemover(String extension) {
121. super ();
122. this .extension = extension;
123. }
124. /** 啟動對某個文件夾的篩選 */
125. @SuppressWarnings ( "unchecked" )
126. public void start(String rootDir) throws IOException {
127. traverseFolder2(rootDir);
128. }
131. public void traverseFolder2(String path) throws IOException {
133. File file = new File(path);
134. if (file.exists()) {
135. File[] files = file.listFiles();
136. if (files.length == 0) {
137. System.out.println("文件夹是空的!");
138. return;
139. } else {
140. for (File file2 : files) {
141. if (file2.isDirectory()) {
142. System.out.println("文件夹:" + file2.getAbsolutePath());
143. traverseFolder2(file2.getAbsolutePath());
144. } else {
145. remove(file2.getAbsolutePath());
146. }
147. }
148. }
149. } else {
150. System.out.println("文件不存在!");
151. }
152. }
153. /** 移除UTF-8的BOM */
154. private void remove(String path) throws IOException {
155. saveFile(path);
156. trimBom(path);
157. }
163. }

第二种方法的 UnicodeReader 类

1. package cn.com.do1.component.common.util;
2. import java.io.IOException;
3. import java.io.InputStream;
4. import java.io.InputStreamReader;
5. import java.io.PushbackInputStream;
6. import java.io.Reader;
7. /**
8. version: 1.1 / 2007-01-25
9. - changed BOM recognition ordering (longer boms first)
10. 网络地址：http://koti.mbnet.fi/akini/java/unicodereader/UnicodeReader.java.txt
11. Original pseudocode : Thomas Weidenfeller
12. Implementation tweaked: Aki Nieminen
13. http://www.unicode.org/unicode/faq/utf_bom.html
14. BOMs:
15. 00 00 FE FF = UTF-32, big-endian
16. FF FE 00 00 = UTF-32, little-endian
17. EF BB BF = UTF-8,
18. FE FF = UTF-16, big-endian
19. FF FE = UTF-16, little-endian
20. Win2k Notepad:
21. Unicode format = UTF-16LE
22. ***/
23. /**
24. * Generic unicode textreader, which will use BOM mark
25. * to identify the encoding to be used. If BOM is not found
26. * then use a given default or system encoding.
27. */
28. public class UnicodeReader extends Reader {
29. PushbackInputStream internalIn;
30. InputStreamReader internalIn2 = null;
31. String defaultEnc;
32. private static final int BOM_SIZE = 4;
33. /**
34. *
35. * @param in inputstream to be read
36. * @param defaultEnc default encoding if stream does not have
37. * BOM marker. Give NULL to use system-level default.
38. */
39. UnicodeReader(InputStream in, String defaultEnc) {
40. internalIn = new PushbackInputStream(in, BOM_SIZE);
41. this.defaultEnc = defaultEnc;
42. }
43. public String getDefaultEncoding() {
44. return defaultEnc;
45. }
46. /**
47. * Get stream encoding or NULL if stream is uninitialized.
48. * Call init() or read() method to initialize it.
49. */
50. public String getEncoding() {
51. if (internalIn2 == null) return null;
52. return internalIn2.getEncoding();
53. }
54. /**
55. * Read-ahead four bytes and check for BOM marks. Extra bytes are
56. * unread back to the stream, only BOM bytes are skipped.
57. */
58. protected void init() throws IOException {
59. if (internalIn2 != null) return;
60. String encoding;
61. byte bom[] = new byte[BOM_SIZE];
62. int n, unread;
63. n = internalIn.read(bom, 0, bom.length);
64. if ( (bom[0] == (byte)0x00) && (bom[1] == (byte)0x00) &&
65. (bom[2] == (byte)0xFE) && (bom[3] == (byte)0xFF) ) {
66. encoding = "UTF-32BE";
67. unread = n - 4;
68. } else if ( (bom[0] == (byte)0xFF) && (bom[1] == (byte)0xFE) &&
69. (bom[2] == (byte)0x00) && (bom[3] == (byte)0x00) ) {
70. encoding = "UTF-32LE";
71. unread = n - 4;
72. } else if ( (bom[0] == (byte)0xEF) && (bom[1] == (byte)0xBB) &&
73. (bom[2] == (byte)0xBF) ) {
74. encoding = "UTF-8";
75. unread = n - 3;
76. } else if ( (bom[0] == (byte)0xFE) && (bom[1] == (byte)0xFF) ) {
77. encoding = "UTF-16BE";
78. unread = n - 2;
79. } else if ( (bom[0] == (byte)0xFF) && (bom[1] == (byte)0xFE) ) {
80. encoding = "UTF-16LE";
81. unread = n - 2;
82. } else {
83. // Unicode BOM mark not found, unread all bytes
84. encoding = defaultEnc;
85. unread = n;
86. }
87. //System.out.println("read=" + n + ", unread=" + unread);
88. if (unread > 0) internalIn.unread(bom, (n - unread), unread);
89. // Use given encoding
90. if (encoding == null) {
91. internalIn2 = new InputStreamReader(internalIn);
92. } else {
93. internalIn2 = new InputStreamReader(internalIn, encoding);
94. }
95. }
96. public void close() throws IOException {
97. init();
98. internalIn2.close();
99. }
100. public int read(char[] cbuf, int off, int len) throws IOException {
101. init();
102. return internalIn2.read(cbuf, off, len);
103. }
104. }