Base64是网络上最常见的用于传输8Bit字节代码的编码方式之一,大家可以查看RFC2045~RFC2049,上面有MIME的详细规范。Base64编码可用于在HTTP环境下传递较长的标识信息。例如,在Java Persistence系统Hibernate中,就采用了Base64来将一个较长的唯一标识符(一般为128-bit的UUID)编码为一个字符串,用作HTTP表单和HTTP GET URL中的参数。在其他应用程序中,也常常需要把二进制数据编码为适合放在URL(包括隐藏表单域)中的形式。此时,采用Base64编码不仅比较简短,同时也具有不可读性,即所编码的数据不会被人用肉眼所直接看到。
import java.io.UnsupportedEncodingException; import java.util.Arrays; public class Base64 { private static final char[] CA = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/" .toCharArray(); private static final int[] IA = new int[256]; static { Arrays.fill(IA, -1); for (int i = 0, iS = CA.length; i < iS; i++) IA[CA[i]] = i; IA['='] = 0; } public final static char[] encodeToChar(byte[] sArr, boolean lineSep) { // Check special case int sLen = sArr != null ? sArr.length : 0; if (sLen == 0) return new char[0]; int eLen = (sLen / 3) * 3; // Length of even 24-bits. int cCnt = ((sLen - 1) / 3 + 1) << 2; // Returned character count int dLen = cCnt + (lineSep ? (cCnt - 1) / 76 << 1 : 0); // Length of // returned // array char[] dArr = new char[dLen]; // Encode even 24-bits for (int s = 0, d = 0, cc = 0; s < eLen;) { // Copy next three bytes into lower 24 bits of int, paying attension // to sign. int i = (sArr[s++] & 0xff) << 16 | (sArr[s++] & 0xff) << 8 | (sArr[s++] & 0xff); // Encode the int into four chars dArr[d++] = CA[(i >>> 18) & 0x3f]; dArr[d++] = CA[(i >>> 12) & 0x3f]; dArr[d++] = CA[(i >>> 6) & 0x3f]; dArr[d++] = CA[i & 0x3f]; // Add optional line separator if (lineSep && ++cc == 19 && d < dLen - 2) { dArr[d++] = '\r'; dArr[d++] = '\n'; cc = 0; } } // Pad and encode last bits if source isn't even 24 bits. int left = sLen - eLen; // 0 - 2. if (left > 0) { // Prepare the int int i = ((sArr[eLen] & 0xff) << 10) | (left == 2 ? ((sArr[sLen - 1] & 0xff) << 2) : 0); // Set last four chars dArr[dLen - 4] = CA[i >> 12]; dArr[dLen - 3] = CA[(i >>> 6) & 0x3f]; dArr[dLen - 2] = left == 2 ? CA[i & 0x3f] : '='; dArr[dLen - 1] = '='; } return dArr; } public final static byte[] decode(char[] sArr) { // Check special case int sLen = sArr != null ? sArr.length : 0; if (sLen == 0) return new byte[0]; // Count illegal characters (including '\r', '\n') to know what size the // returned array will be, // so we don't have to reallocate & copy it later. int sepCnt = 0; // Number of separator characters. (Actually illegal // characters, but that's a bonus) for (int i = 0; i < sLen; i++) // If input is "pure" (I.e. no line separators or illegal chars) // base64 this loop can be commented out. if (IA[sArr[i]] < 0) sepCnt++; // Check so that legal chars (including '=') are evenly divideable by 4 // as specified in RFC 2045. if ((sLen - sepCnt) % 4 != 0) return null; int pad = 0; for (int i = sLen; i > 1 && IA[sArr[--i]] <= 0;) if (sArr[i] == '=') pad++; int len = ((sLen - sepCnt) * 6 >> 3) - pad; byte[] dArr = new byte[len]; // Preallocate byte[] of exact length for (int s = 0, d = 0; d < len;) { // Assemble three bytes into an int from four "valid" characters. int i = 0; for (int j = 0; j < 4; j++) { // j only increased if a valid char // was found. int c = IA[sArr[s++]]; if (c >= 0) i |= c << (18 - j * 6); else j--; } // Add the bytes dArr[d++] = (byte) (i >> 16); if (d < len) { dArr[d++] = (byte) (i >> 8); if (d < len) dArr[d++] = (byte) i; } } return dArr; } public final static byte[] decodeFast(char[] sArr) { // Check special case int sLen = sArr.length; if (sLen == 0) return new byte[0]; int sIx = 0, eIx = sLen - 1; // Start and end index after trimming. // Trim illegal chars from start while (sIx < eIx && IA[sArr[sIx]] < 0) sIx++; // Trim illegal chars from end while (eIx > 0 && IA[sArr[eIx]] < 0) eIx--; // get the padding count (=) (0, 1 or 2) int pad = sArr[eIx] == '=' ? (sArr[eIx - 1] == '=' ? 2 : 1) : 0; // Count // '=' // at // end. int cCnt = eIx - sIx + 1; // Content count including possible // separators int sepCnt = sLen > 76 ? (sArr[76] == '\r' ? cCnt / 78 : 0) << 1 : 0; int len = ((cCnt - sepCnt) * 6 >> 3) - pad; // The number of decoded // bytes byte[] dArr = new byte[len]; // Preallocate byte[] of exact length // Decode all but the last 0 - 2 bytes. int d = 0; for (int cc = 0, eLen = (len / 3) * 3; d < eLen;) { // Assemble three bytes into an int from four "valid" characters. int i = IA[sArr[sIx++]] << 18 | IA[sArr[sIx++]] << 12 | IA[sArr[sIx++]] << 6 | IA[sArr[sIx++]]; // Add the bytes dArr[d++] = (byte) (i >> 16); dArr[d++] = (byte) (i >> 8); dArr[d++] = (byte) i; // If line separator, jump over it. if (sepCnt > 0 && ++cc == 19) { sIx += 2; cc = 0; } } if (d < len) { // Decode last 1-3 bytes (incl '=') into 1-3 bytes int i = 0; for (int j = 0; sIx <= eIx - pad; j++) i |= IA[sArr[sIx++]] << (18 - j * 6); for (int r = 16; d < len; r -= 8) dArr[d++] = (byte) (i >> r); } return dArr; } public final static byte[] encodeToByte(byte[] sArr, boolean lineSep) { // Check special case int sLen = sArr != null ? sArr.length : 0; if (sLen == 0) return new byte[0]; int eLen = (sLen / 3) * 3; // Length of even 24-bits. int cCnt = ((sLen - 1) / 3 + 1) << 2; // Returned character count int dLen = cCnt + (lineSep ? (cCnt - 1) / 76 << 1 : 0); // Length of // returned // array byte[] dArr = new byte[dLen]; // Encode even 24-bits for (int s = 0, d = 0, cc = 0; s < eLen;) { // Copy next three bytes into lower 24 bits of int, paying attension // to sign. int i = (sArr[s++] & 0xff) << 16 | (sArr[s++] & 0xff) << 8 | (sArr[s++] & 0xff); // Encode the int into four chars dArr[d++] = (byte) CA[(i >>> 18) & 0x3f]; dArr[d++] = (byte) CA[(i >>> 12) & 0x3f]; dArr[d++] = (byte) CA[(i >>> 6) & 0x3f]; dArr[d++] = (byte) CA[i & 0x3f]; // Add optional line separator if (lineSep && ++cc == 19 && d < dLen - 2) { dArr[d++] = '\r'; dArr[d++] = '\n'; cc = 0; } } // Pad and encode last bits if source isn't an even 24 bits. int left = sLen - eLen; // 0 - 2. if (left > 0) { // Prepare the int int i = ((sArr[eLen] & 0xff) << 10) | (left == 2 ? ((sArr[sLen - 1] & 0xff) << 2) : 0); // Set last four chars dArr[dLen - 4] = (byte) CA[i >> 12]; dArr[dLen - 3] = (byte) CA[(i >>> 6) & 0x3f]; dArr[dLen - 2] = left == 2 ? (byte) CA[i & 0x3f] : (byte) '='; dArr[dLen - 1] = '='; } return dArr; } public final static byte[] decode(byte[] sArr) { // Check special case int sLen = sArr.length; // Count illegal characters (including '\r', '\n') to know what size the // returned array will be, // so we don't have to reallocate & copy it later. int sepCnt = 0; // Number of separator characters. (Actually illegal // characters, but that's a bonus) for (int i = 0; i < sLen; i++) // If input is "pure" (I.e. no line separators or illegal chars) // base64 this loop can be commented out. if (IA[sArr[i] & 0xff] < 0) sepCnt++; // Check so that legal chars (including '=') are evenly divideable by 4 // as specified in RFC 2045. if ((sLen - sepCnt) % 4 != 0) return null; int pad = 0; for (int i = sLen; i > 1 && IA[sArr[--i] & 0xff] <= 0;) if (sArr[i] == '=') pad++; int len = ((sLen - sepCnt) * 6 >> 3) - pad; byte[] dArr = new byte[len]; // Preallocate byte[] of exact length for (int s = 0, d = 0; d < len;) { // Assemble three bytes into an int from four "valid" characters. int i = 0; for (int j = 0; j < 4; j++) { // j only increased if a valid char // was found. int c = IA[sArr[s++] & 0xff]; if (c >= 0) i |= c << (18 - j * 6); else j--; } // Add the bytes dArr[d++] = (byte) (i >> 16); if (d < len) { dArr[d++] = (byte) (i >> 8); if (d < len) dArr[d++] = (byte) i; } } return dArr; } public final static byte[] decodeFast(byte[] sArr) { // Check special case int sLen = sArr.length; if (sLen == 0) return new byte[0]; int sIx = 0, eIx = sLen - 1; // Start and end index after trimming. // Trim illegal chars from start while (sIx < eIx && IA[sArr[sIx] & 0xff] < 0) sIx++; // Trim illegal chars from end while (eIx > 0 && IA[sArr[eIx] & 0xff] < 0) eIx--; // get the padding count (=) (0, 1 or 2) int pad = sArr[eIx] == '=' ? (sArr[eIx - 1] == '=' ? 2 : 1) : 0; // Count // '=' // at // end. int cCnt = eIx - sIx + 1; // Content count including possible // separators int sepCnt = sLen > 76 ? (sArr[76] == '\r' ? cCnt / 78 : 0) << 1 : 0; int len = ((cCnt - sepCnt) * 6 >> 3) - pad; // The number of decoded // bytes byte[] dArr = new byte[len]; // Preallocate byte[] of exact length // Decode all but the last 0 - 2 bytes. int d = 0; for (int cc = 0, eLen = (len / 3) * 3; d < eLen;) { // Assemble three bytes into an int from four "valid" characters. int i = IA[sArr[sIx++]] << 18 | IA[sArr[sIx++]] << 12 | IA[sArr[sIx++]] << 6 | IA[sArr[sIx++]]; // Add the bytes dArr[d++] = (byte) (i >> 16); dArr[d++] = (byte) (i >> 8); dArr[d++] = (byte) i; // If line separator, jump over it. if (sepCnt > 0 && ++cc == 19) { sIx += 2; cc = 0; } } if (d < len) { // Decode last 1-3 bytes (incl '=') into 1-3 bytes int i = 0; for (int j = 0; sIx <= eIx - pad; j++) i |= IA[sArr[sIx++]] << (18 - j * 6); for (int r = 16; d < len; r -= 8) dArr[d++] = (byte) (i >> r); } return dArr; } public final static String encodeToString(byte[] sArr, boolean lineSep) { // Reuse char[] since we can't create a String incrementally anyway and // StringBuffer/Builder would be slower. return new String(encodeToChar(sArr, lineSep)); } public final static String encode(String s) { // Reuse char[] since we can't create a String incrementally anyway and // StringBuffer/Builder would be slower. try { return new String(encodeToChar(s.getBytes("UTF-8"), false)); } catch (UnsupportedEncodingException e) { System.err.println("Base64 encoding error: " + e.getMessage()); e.printStackTrace(); } return null; } public final static byte[] decode(String str, boolean used) { // Check special case int sLen = str != null ? str.length() : 0; if (sLen == 0) return new byte[0]; // Count illegal characters (including '\r', '\n') to know what size the // returned array will be, // so we don't have to reallocate & copy it later. int sepCnt = 0; // Number of separator characters. (Actually illegal // characters, but that's a bonus) for (int i = 0; i < sLen; i++) // If input is "pure" (I.e. no line separators or illegal chars) // base64 this loop can be commented out. if (IA[str.charAt(i)] < 0) sepCnt++; // Check so that legal chars (including '=') are evenly divideable by 4 // as specified in RFC 2045. if ((sLen - sepCnt) % 4 != 0) return null; // Count '=' at end int pad = 0; for (int i = sLen; i > 1 && IA[str.charAt(--i)] <= 0;) if (str.charAt(i) == '=') pad++; int len = ((sLen - sepCnt) * 6 >> 3) - pad; byte[] dArr = new byte[len]; // Preallocate byte[] of exact length for (int s = 0, d = 0; d < len;) { // Assemble three bytes into an int from four "valid" characters. int i = 0; for (int j = 0; j < 4; j++) { // j only increased if a valid char // was found. int c = IA[str.charAt(s++)]; if (c >= 0) i |= c << (18 - j * 6); else j--; } // Add the bytes dArr[d++] = (byte) (i >> 16); if (d < len) { dArr[d++] = (byte) (i >> 8); if (d < len) dArr[d++] = (byte) i; } } return dArr; } public final static byte[] decodeFast(String s) { // Check special case int sLen = s.length(); if (sLen == 0) return new byte[0]; int sIx = 0, eIx = sLen - 1; // Start and end index after trimming. // Trim illegal chars from start while (sIx < eIx && IA[s.charAt(sIx) & 0xff] < 0) sIx++; // Trim illegal chars from end while (eIx > 0 && IA[s.charAt(eIx) & 0xff] < 0) eIx--; // get the padding count (=) (0, 1 or 2) int pad = s.charAt(eIx) == '=' ? (s.charAt(eIx - 1) == '=' ? 2 : 1) : 0; // Count // '=' // at // end. int cCnt = eIx - sIx + 1; // Content count including possible // separators int sepCnt = sLen > 76 ? (s.charAt(76) == '\r' ? cCnt / 78 : 0) << 1 : 0; int len = ((cCnt - sepCnt) * 6 >> 3) - pad; // The number of decoded // bytes byte[] dArr = new byte[len]; // Preallocate byte[] of exact length // Decode all but the last 0 - 2 bytes. int d = 0; for (int cc = 0, eLen = (len / 3) * 3; d < eLen;) { // Assemble three bytes into an int from four "valid" characters. int i = IA[s.charAt(sIx++)] << 18 | IA[s.charAt(sIx++)] << 12 | IA[s.charAt(sIx++)] << 6 | IA[s.charAt(sIx++)]; // Add the bytes dArr[d++] = (byte) (i >> 16); dArr[d++] = (byte) (i >> 8); dArr[d++] = (byte) i; // If line separator, jump over it. if (sepCnt > 0 && ++cc == 19) { sIx += 2; cc = 0; } } if (d < len) { // Decode last 1-3 bytes (incl '=') into 1-3 bytes int i = 0; for (int j = 0; sIx <= eIx - pad; j++) i |= IA[s.charAt(sIx++)] << (18 - j * 6); for (int r = 16; d < len; r -= 8) dArr[d++] = (byte) (i >> r); } return dArr; } public static String decode(String s) throws UnsupportedEncodingException { return new String(Base64.decodeFast(s), "utf-8"); } public static void main(String[] args) throws UnsupportedEncodingException { String str="测试"; String encode=encode(str); System.out.println(encode); String decode=decode(encode); System.out.println(decode); } }
运行结果如下:
5rWL6K+V
测试