stringutil.java

google gdata API 很好用的API

  public static String stripPrefix(String str, String prefix) {
    return str.startsWith(prefix) ? str.substring(prefix.length()) : null;
  }

  /**
   * Case insensitive version of stripPrefix.
   * Analogous to the c++ functions strcaseprefix and var_strcaseprefix.
   */
  public static String stripPrefixIgnoreCase(String str, String prefix) {
    if (str.length() >= prefix.length()
        && str.substring(0, prefix.length()).equalsIgnoreCase(prefix)) {
      return str.substring(prefix.length());
    }

    return null;
  }

  /**
   * Strips all non-digit characters from a string.
   *
   * The resulting string will only contain characters for which isDigit()
   * returns true.
   *
   * @param str the string to strip
   * @return a string consisting of digits only, or an empty string
   */
  public static String stripNonDigits(String str) {
    StringBuffer result = new StringBuffer(str.length());
    for (char candidate : str.toCharArray()) {
      if (Character.isDigit(candidate)) {
        result.append(candidate);
      }
    }
    return result.toString();
  }

  /**
   * Counts the number of (not necessarily distinct) characters in the
   * string that also happen to be in 'chars'
   */
  public static int numSharedChars(final String str, final String chars) {

    if (str == null || chars == null) {
      return 0;
    }

    int total = 0, pos = -1;
    while ((pos = indexOfChars(str, chars, pos + 1)) != -1) {
      total++;
    }

    return total;
  }

  /**
   * Like String.indexOf() except that it will look for any of the
   * characters in 'chars' (similar to C's strpbrk)
   */
  public static int indexOfChars(String str, String chars, int fromIndex) {
    final int len = str.length();

    for (int pos = fromIndex; pos < len; pos++) {
      if (chars.indexOf(str.charAt(pos)) >= 0) {
        return pos;
      }
    }

    return -1;
  }

  /**
   * Like String.indexOf() except that it will look for any of the
   * characters in 'chars' (similar to C's strpbrk)
   */
  public static int indexOfChars(String str, String chars) {
    return indexOfChars(str, chars, 0);
  }

  /**
   * Finds the last index in str of a character not in the characters
   * in 'chars' (similar to ANSI string.find_last_not_of).
   *
   * Returns -1 if no such character can be found.
   */
  public static int lastIndexNotOf(String str, String chars, int fromIndex) {
    fromIndex = Math.min(fromIndex, str.length() - 1);

    for (int pos = fromIndex; pos >= 0; pos--) {
      if (chars.indexOf(str.charAt(pos)) < 0) {
        return pos;
      }
    }

    return -1;
  }

  /**
   * Like String.replace() except that it accepts any number of old chars.
   * Replaces any occurrances of 'oldchars' in 'str' with 'newchar'.
   * Example: replaceChars("Hello, world!", "H,!", ' ') returns " ello  world "
   */
  public static String replaceChars(String str, String oldchars, char newchar) {
    int pos = indexOfChars(str, oldchars);
    if (pos == -1) {
      return str;
    }

    StringBuilder buf = new StringBuilder(str);
    do {
      buf.setCharAt(pos, newchar);
      pos = indexOfChars(str, oldchars, pos + 1);
    } while (pos != -1);

    return buf.toString();
  }

  /**
   * Remove any occurrances of 'oldchars' in 'str'.
   * Example: removeChars("Hello, world!", ",!") returns "Hello world"
   */
  public static String removeChars(String str, String oldchars) {
    int pos = indexOfChars(str, oldchars);
    if (pos == -1) {
      return str;
    }

    StringBuilder buf = new StringBuilder();
    int start = 0;
    do {
      buf.append(str.substring(start, pos));
      start = pos + 1;
      pos = indexOfChars(str, oldchars, start);
    } while (pos != -1);

    if (start < str.length()) {
      buf.append(str.substring(start));
    }
    return buf.toString();
  }

  /**
   * Removes all characters from 'str' that are not in 'retainChars'.
   * Example: retainAllChars("Hello, world!", "lo") returns "llool"
   */
  public static String retainAllChars(String str, String retainChars) {
    int pos = indexOfChars(str, retainChars);
    if (pos == -1) {
      return "";
    }
    StringBuilder buf = new StringBuilder();
    do {
      buf.append(str.charAt(pos));
      pos = indexOfChars(str, retainChars, pos + 1);
    } while (pos != -1);
    return buf.toString();
  }

  /**
   * Replaces microsoft "smart quotes" (curly " and ') with their
   * ascii counterparts.
   */
  public static String replaceSmartQuotes(String str) {
    // See http://www.microsoft.com/typography/unicode/1252.htm
    str = replaceChars(str, "\u0091\u0092\u2018\u2019", '\'');
    str = replaceChars(str, "\u0093\u0094\u201c\u201d", '"');
    return str;
  }

  /**
   * Convert a string of hex digits to a byte array, with the first
   * byte in the array being the MSB. The string passed in should be
   * just the raw digits (upper or lower case), with no leading
   * or trailing characters (like '0x' or 'h').
   * An odd number of characters is supported.
   * If the string is empty, an empty array will be returned.
   *
   * This is significantly faster than using
   *   new BigInteger(str, 16).toByteArray();
   * especially with larger strings. Here are the results of some
   * microbenchmarks done on a P4 2.8GHz 2GB RAM running
   * linux 2.4.22-gg11 and JDK 1.5 with an optimized build:
   *
   * String length        hexToBytes (usec)   BigInteger
   * -----------------------------------------------------
   * 16                       0.570                 1.43
   * 256                      8.21                 44.4
   * 1024                    32.8                 526
   * 16384                  546                121000
   */
  public static byte[] hexToBytes(String str) {
    byte[] bytes = new byte[(str.length() + 1) / 2];
    if (str.length() == 0) {
      return bytes;
    }
    bytes[0] = 0;
    int nibbleIdx = (str.length() % 2);
    for (int i = 0; i < str.length(); i++) {
      char c = str.charAt(i);
      if (!isHex(c)) {
        throw new IllegalArgumentException("string contains non-hex chars");
      }
      if ((nibbleIdx % 2) == 0) {
        bytes[nibbleIdx >> 1] = (byte) (hexValue(c) << 4);
      } else {
        bytes[nibbleIdx >> 1] += (byte) hexValue(c);
      }
      nibbleIdx++;
    }
    return bytes;
  }

  /**
   * Converts any instances of "\r" or "\r\n" style EOLs into "\n" (Line Feed).
   */
  public static String convertEOLToLF(String input) {
    StringBuilder res = new StringBuilder(input.length());
    char[] s = input.toCharArray();
    int from = 0;
    final int end = s.length;
    for (int i = 0; i < end; i++) {
      if (s[i] == '\r') {
        res.append(s, from, i - from);
        res.append('\n');
        if (i + 1 < end && s[i + 1] == '\n') {
          i++;
        }

        from = i + 1;
      }
    }

    if (from == 0) {   // no \r!
      return input;
    }

    res.append(s, from, end - from);
    return res.toString();
  }

  /** @deprecated Please inline this method. */
  @Deprecated public static String convertEOLToCRLF(String input) {
    return input.replaceAll("(\r\n|\r|\n)", "\r\n");
  }

  /**
   * Returns a string consisting of "s", plus enough copies of "pad_ch" on the
   * left hand side to make the length of "s" equal to or greater than len (if
   * "s" is already longer than "len", then "s" is returned).
   */
  public static String padLeft(String s, int len, char pad_ch) {
    if (s.length() >= len) {
      return s;
    } else {
      StringBuilder sb = new StringBuilder();
      int n = len - s.length();
      for (int i = 0; i < n; i++) {
        sb.append(pad_ch);
      }
      sb.append(s);
      return sb.toString();
    }
  }

  /**
   * Returns a string consisting of "s", plus enough copies of "pad_ch" on the
   * right hand side to make the length of "s" equal to or greater than len (if
   * "s" is already longer than "len", then "s" is returned).
   */
  public static String padRight(String s, int len, char pad_ch) {
    if (s.length() >= len) {
      return s;
    } else {
      StringBuilder sb = new StringBuilder();
      int n = len - s.length();
      sb.append(s);
      for (int i = 0; i < n; i++) {
        sb.append(pad_ch);
      }
      return sb.toString();
    }
  }

  /**
   * Returns a string consisting of "s", with each of the first "len" characters
   * replaced by "mask_ch" character.
   */
  public static String maskLeft(String s, int len, char mask_ch) {
    if (len <= 0) {
      return s;
    }
    len = Math.min(len, s.length());
    StringBuilder sb = new StringBuilder();
    for (int i = 0; i < len; i++) {
      sb.append(mask_ch);
    }
    sb.append(s.substring(len));
    return sb.toString();
  }

  /**
   * Returns a string consisting of "s", with each of the last "len" characters
   * replaces by "mask_ch" character.
   */
  public static String maskRight(String s, int len, char mask_ch) {
    if (len <= 0) {
      return s;
    }
    len = Math.min(len, s.length());
    StringBuilder sb = new StringBuilder();
    sb.append(s.substring(0, s.length() - len));
    for (int i = 0; i < len; i++) {
      sb.append(mask_ch);
    }
    return sb.toString();
  }

  private static boolean isOctal(char c) {
    return (c >= '0') && (c <= '7');
  }

  private static boolean isHex(char c) {
    return ((c >= '0') && (c <= '9')) ||
           ((c >= 'a') && (c <= 'f')) ||
           ((c >= 'A') && (c <= 'F'));
  }

  private static int hexValue(char c) {
    if ((c >= '0') && (c <= '9')) {
      return (c - '0');
    } else if ((c >= 'a') && (c <= 'f')) {
      return (c - 'a') + 10;
    } else {
      return (c - 'A') + 10;
    }
  }

  /**
   * Unescape any C escape sequences (\n, \r, \\, \ooo, etc) and return the
   * resulting string.
   */
  public static String unescapeCString(String s) {
    if (s.indexOf('\\') < 0) {
      // Fast path: nothing to unescape
      return s;
    }

    StringBuilder sb = new StringBuilder();
    int len = s.length();
    for (int i = 0; i < len;) {
      char c = s.charAt(i++);
      if (c == '\\' && (i < len)) {
        c = s.charAt(i++);
        switch (c) {
          case 'a':  c = '\007';  break;
          case 'b':  c = '\b';    break;
          case 'f':  c = '\f';    break;
          case 'n':  c = '\n';    break;
          case 'r':  c = '\r';    break;
          case 't':  c = '\t';    break;
          case 'v':  c = '\013';  break;
          case '\\': c = '\\';    break;
          case '?':  c = '?';     break;
          case '\'': c = '\'';    break;
          case '"':  c = '\"';    break;

          default: {
            if ((c == 'x') && (i < len) && isHex(s.charAt(i))) {
              // "\xXX"
              int v = hexValue(s.charAt(i++));
              if ((i < len) && isHex(s.charAt(i))) {
                v = v*16 + hexValue(s.charAt(i++));
              }
              c = (char)v;
            } else if (isOctal(c)) {
              // "\OOO"
              int v = (c - '0');
              if ((i < len) && isOctal(s.charAt(i))) {