Package net.i2p.data

Class DataHelper


  • public class DataHelper
    extends Object
    Defines some simple IO routines for dealing with marshalling data structures
    Author:
    jrandom
    • Constructor Summary

      Constructors 
      Constructor Description
      DataHelper()  
    • Method Summary

      All Methods Static Methods Concrete Methods Deprecated Methods 
      Modifier and Type Method Description
      static int compareTo​(byte[] lhs, byte[] rhs)
      Big endian compare, treats bytes as unsigned.
      static byte[] compress​(byte[] orig)
      Compress the data and return a new GZIP compressed byte array.
      static byte[] compress​(byte[] orig, int offset, int size)
      Compress the data and return a new GZIP compressed byte array.
      static byte[] compress​(byte[] orig, int offset, int size, int level)
      Compress the data and return a new GZIP compressed byte array.
      static void copy​(InputStream in, OutputStream out)
      Copy in to out.
      static byte[] decompress​(byte[] orig)
      Decompress the GZIP compressed data (returning null on error).
      static byte[] decompress​(byte[] orig, int offset, int length)
      Decompress the GZIP compressed data (returning null on error).
      static boolean eq​(byte[] lhs, byte[] rhs)
      Run a comparison on the byte arrays, byte by byte.
      static boolean eq​(byte[] lhs, int offsetLeft, byte[] rhs, int offsetRight, int length)
      Unlike eq(byte[], byte[]), this returns false if either lhs or rhs is null.
      static boolean eq​(byte lhs, byte rhs)
      Deprecated.
      inefficient
      static boolean eq​(int lhs, int rhs)
      Deprecated.
      inefficient
      static boolean eq​(long lhs, long rhs)
      Deprecated.
      inefficient
      static boolean eq​(Object lhs, Object rhs)
      Helper util to compare two objects, including null handling.
      static boolean eq​(Collection<?> lhs, Collection<?> rhs)
      Run a deep comparison across the two collections.
      static boolean eqCT​(byte[] lhs, int offsetLeft, byte[] rhs, int offsetRight, int length)
      Unlike eq(), this throws NPE if either lhs or rhs is null.
      static String escapeHTML​(String unescaped)
      Escape a string for inclusion in HTML
      static String formatDate​(long now)
      The default formatting for date, current locale, local time zone.
      static String formatDuration​(long ms)
      NOTE: formatDuration2() recommended in most cases for readability
      static String formatDuration2​(double ms)
      Like formatDuration2(long) but with microsec and nanosec also.
      static String formatDuration2​(long ms)
      Like formatDuration but with a non-breaking space after the number, 0 is unitless, and the unit is translated.
      static String formatSize​(long bytes)
      This is binary, i.e.
      static String formatSize2​(long bytes)
      This is binary, i.e.
      static String formatSize2​(long bytes, boolean nonBreaking)
      This is binary, i.e.
      static String formatSize2Decimal​(long bytes)
      This is decimal, i.e.
      static String formatSize2Decimal​(long bytes, boolean nonBreaking)
      This is decimal, i.e.
      static String formatTime​(long now)
      The default formatting for date/time, current locale, local time zone.
      static Date fromDate​(byte[] src, int offset)  
      static byte[] fromHexString​(String val)
      Deprecated.
      unused
      static long fromLong​(byte[] src, int offset, int numBytes)
      Big endian.
      static long fromLong8​(byte[] src, int offset)
      Big endian.
      static long fromLongLE​(byte[] src, int offset, int numBytes)
      Little endian, i.e.
      static int fromProperties​(byte[] source, int offset, Properties target)
      Reads the props from the byte array and puts them in the Properties target See readProperties() for the format.
      static byte[] getASCII​(String orig)
      Roughly the same as orig.getBytes("ISO-8859-1") but much faster and will not throw an exception.
      static String getUTF8​(byte[] orig)
      Same as new String(orig, "UTF-8") but throws an unchecked RuntimeException instead of an UnsupportedEncodingException if no UTF-8, for ease of use.
      static String getUTF8​(byte[] orig, int offset, int len)
      Same as new String(orig, "UTF-8") but throws an unchecked RuntimeException instead of an UnsupportedEncodingException if no UTF-8, for ease of use.
      static byte[] getUTF8​(String orig)
      Same as orig.getBytes("UTF-8") but throws an unchecked RuntimeException instead of an UnsupportedEncodingException if no UTF-8, for ease of use.
      static byte[] getUTF8​(StringBuffer orig)
      Deprecated.
      unused
      static int hashCode​(byte[] b)
      Calculate the hashcode of the byte array, using 0 for null
      static int hashCode​(Object obj)
      Calculate the hashcode of the object, using 0 for null
      static int hashCode​(Collection<?> col)
      Calculate the hashcode of the collection, using 0 for null
      static int hashCode​(Date obj)
      Calculate the hashcode of the date, using 0 for null
      static void loadProps​(Properties props, File file)
      A more efficient Properties.load Some of the other differences: - UTF-8 encoding, not ISO-8859-1 - No escaping! This does not process or drop backslashes - '#' or ';' starts a comment line, but '!' does not - Leading whitespace is not trimmed - '=' is the only key-termination character (not ':' or whitespace) As of 0.9.10, an empty value is allowed.
      static void loadProps​(Properties props, File file, boolean forceLowerCase)  
      static void loadProps​(Properties props, InputStream inStr)  
      static void loadProps​(Properties props, InputStream inStr, boolean forceLowerCase)  
      static int read​(InputStream in, byte[] target)
      This is different than InputStream.read(target), in that it does repeated reads until the full data is received.
      static int read​(InputStream in, byte[] target, int offset, int length)
      WARNING - This is different than InputStream.read(target, offset, length) for a nonzero offset, in that it returns the new offset (== old offset + length).
      static Date readDate​(InputStream in)
      Read in a date from the stream as specified by the I2P data structure spec.
      static String readLine​(InputStream in)
      Read a newline delimited line from the stream, returning the line (without the newline), or null if EOF reached on an empty line Warning - strips \n but not \r Warning - 8KB line length limit as of 0.7.13, @throws IOException if exceeded Warning - not UTF-8
      static boolean readLine​(InputStream in, StringBuilder buf)
      Read in a line, placing it into the buffer (excluding the newline).
      static boolean readLine​(InputStream in, StringBuilder buf, MessageDigest hash)
      update the hash along the way Warning - strips \n but not \r Warning - 8KB line length limit as of 0.7.13, @throws IOException if exceeded Warning - not UTF-8
      static String readLine​(InputStream in, MessageDigest hash)
      update the hash along the way Warning - strips \n but not \r Warning - 8KB line length limit as of 0.7.13, @throws IOException if exceeded Warning - not UTF-8
      static long readLong​(InputStream rawStream, int numBytes)
      Read the stream for an integer as defined by the I2P data structure specification.
      static Properties readProperties​(InputStream rawStream)
      Read a mapping from the stream, as defined by the I2P data structure spec, and store it into a Properties object.
      static Properties readProperties​(InputStream rawStream, Properties props)
      Ditto, load into an existing properties As of 0.9.18, throws DataFormatException on duplicate key
      static String readString​(InputStream in)
      Read in a string from the stream as specified by the I2P data structure spec.
      static void replace​(StringBuilder buf, String from, String to)
      Replace all instances of "from" with "to" in the StringBuilder buf.
      static void skip​(InputStream in, long n)
      This is different than InputStream.skip(), in that it does repeated reads until the full amount is skipped.
      static <T> void sort​(List<T> list, Comparator<? super T> c)
      Same as Collections.sort(), but guaranteed not to throw an IllegalArgumentException if the sort is unstable.
      static <T> void sort​(T[] a, Comparator<? super T> c)
      Same as Arrays.sort(), but guaranteed not to throw an IllegalArgumentException if the sort is unstable.
      static String[] split​(String s, String regex)
      Same as s.split(regex) but caches the compiled pattern for speed.
      static String[] split​(String s, String regex, int limit)
      Same as s.split(regex, limit) but caches the compiled pattern for speed.
      static void storeProps​(Properties props, File file)
      Writes the props to the file, unsorted (unless props is an OrderedProperties) Note that this does not escape the \r or \n that are unescaped in loadProps() above.
      static String stripHTML​(String orig)
      Strip out any HTML (simply removing any less than / greater than symbols)
      static void toDate​(byte[] target, int offset, long when)  
      static byte[] toDate​(Date date)
      Deprecated.
      unused
      static String toDecimalString​(byte[] buf, int len)
      Deprecated.
      unused
      static String toHexString​(byte[] data)
      Lower-case hex without leading zeros.
      static void toLong​(byte[] target, int offset, int numBytes, long value)
      Big endian.
      static byte[] toLong​(int numBytes, long value)
      Big endian.
      static void toLong8​(byte[] target, int offset, long value)
      Big endian.
      static void toLongLE​(byte[] target, int offset, int numBytes, long value)
      Little endian, i.e.
      static int toProperties​(byte[] target, int offset, Properties props)  
      static byte[] toProperties​(Properties opts)
      Writes the props to returned byte array, not sorted (unless the opts param is an OrderedProperties) Strings will be UTF-8 encoded in the byte array.
      static String toString​(byte[] buf)
      Lower-case hex with leading zeros.
      static String toString​(byte[] buf, int len)
      Lower-case hex with leading zeros.
      static String toString​(Collection<?> col)
      Pretty print the collection
      static String toString​(Map<?,​?> options)
      Pretty print the mapping, unsorted (unless the options param is an OrderedProperties)
      static String toString​(Properties options)
      Pretty print the mapping, unsorted (unless the options param is an OrderedProperties)
      static void write​(OutputStream out, byte[] data, MessageDigest hash)
      update the hash along the way
      static void writeDate​(OutputStream out, Date date)
      Write out a date to the stream as specified by the I2P data structure spec.
      static void writeLong​(OutputStream rawStream, int numBytes, long value)
      Write an integer as defined by the I2P data structure specification to the stream.
      static void writeProperties​(OutputStream rawStream, Properties props)
      Write a mapping to the stream, as defined by the I2P data structure spec, and store it into a Properties object.
      static void writeProperties​(OutputStream rawStream, Properties props, boolean utf8)
      Writes the props to the stream, sorted by property name.
      static void writeProperties​(OutputStream rawStream, Properties props, boolean utf8, boolean sort)
      Writes the props to the stream, sorted by property name if sort == true or if props is an OrderedProperties.
      static void writeString​(OutputStream out, String string)
      Write out a string to the stream as specified by the I2P data structure spec.
      static void writeStringUTF8​(OutputStream out, String string)
      Write out a string to the stream as specified by the I2P data structure spec.
      static byte[] xor​(byte[] lhs, byte[] rhs)  
      static void xor​(byte[] lhs, int startLeft, byte[] rhs, int startRight, byte[] out, int startOut, int len)
      xor the lhs with the rhs, storing the result in out.
    • Field Detail

      • MAX_COMPRESSION

        public static final int MAX_COMPRESSION
        Appx. 30% slower, 2.5% smaller than MEDIUM_COMPRESSION
        See Also:
        Constant Field Values
      • HIGH_COMPRESSION

        public static final int HIGH_COMPRESSION
        Appx. 15% slower, 1.5% smaller than MEDIUM_COMPRESSION
        Since:
        0.9.47
        See Also:
        Constant Field Values
      • MEDIUM_COMPRESSION

        public static final int MEDIUM_COMPRESSION
        New default as of 0.9.47
        Since:
        0.9.47
        See Also:
        Constant Field Values
    • Constructor Detail

      • DataHelper

        public DataHelper()
    • Method Detail

      • readProperties

        public static Properties readProperties​(InputStream rawStream)
                                         throws DataFormatException,
                                                IOException
        Read a mapping from the stream, as defined by the I2P data structure spec, and store it into a Properties object. A mapping is a set of key / value pairs. It starts with a 2 byte Integer (ala readLong(rawStream, 2)) defining how many bytes make up the mapping. After that comes that many bytes making up a set of UTF-8 encoded characters. The characters are organized as key=value;. The key is a String (ala readString(rawStream)) unique as a key within the current mapping that does not include the UTF-8 characters '=' or ';'. After the key comes the literal UTF-8 character '='. After that comes a String (ala readString(rawStream)) for the value. Finally after that comes the literal UTF-8 character ';'. This key=value; is repeated until there are no more bytes (not characters!) left as defined by the first two byte integer. As of 0.9.18, throws DataFormatException on duplicate key
        Parameters:
        rawStream - stream to read the mapping from
        Returns:
        an OrderedProperties
        Throws:
        DataFormatException - if the format is invalid
        IOException - if there is a problem reading the data
      • readProperties

        public static Properties readProperties​(InputStream rawStream,
                                                Properties props)
                                         throws DataFormatException,
                                                IOException
        Ditto, load into an existing properties As of 0.9.18, throws DataFormatException on duplicate key
        Parameters:
        props - The Properties to load into. As of 0.9.38, if null, a new OrderedProperties will be created.
        rawStream - stream to read the mapping from
        Returns:
        the parameter props, or (as of 0.9.38) a new OrderedProperties if props is null, and an immutable EmptyProperties if empty.
        Throws:
        DataFormatException - if the format is invalid
        IOException - if there is a problem reading the data
        Since:
        0.8.13
      • writeProperties

        public static void writeProperties​(OutputStream rawStream,
                                           Properties props)
                                    throws DataFormatException,
                                           IOException
        Write a mapping to the stream, as defined by the I2P data structure spec, and store it into a Properties object. See readProperties for the format. Output is sorted by property name. Property keys and values must not contain '=' or ';', this is not checked and they are not escaped Keys and values must be 255 bytes or less, Formatted length must not exceed 65535 bytes Properties from the defaults table of props (if any) are not written out by this method.
        Parameters:
        rawStream - stream to write to
        props - properties to write out, may be null
        Throws:
        DataFormatException - if there is not enough valid data to write out, or a length limit is exceeded
        IOException - if there is an IO error writing out the data
      • writeProperties

        public static void writeProperties​(OutputStream rawStream,
                                           Properties props,
                                           boolean utf8)
                                    throws DataFormatException,
                                           IOException
        Writes the props to the stream, sorted by property name. See readProperties() for the format. Property keys and values must not contain '=' or ';', this is not checked and they are not escaped Keys and values must be 255 bytes or less, Formatted length must not exceed 65535 bytes Properties from the defaults table of props (if any) are not written out by this method. jrandom disabled UTF-8 in mid-2004, for performance reasons, i.e. slow foo.getBytes("UTF-8") Re-enable it so we can pass UTF-8 tunnel names through the I2CP SessionConfig. Use utf8 = false for RouterAddress (fast, non UTF-8) Use utf8 = true for SessionConfig (slow, UTF-8)
        Parameters:
        props - source may be null
        Throws:
        DataFormatException - if a length limit is exceeded
        IOException
      • writeProperties

        public static void writeProperties​(OutputStream rawStream,
                                           Properties props,
                                           boolean utf8,
                                           boolean sort)
                                    throws DataFormatException,
                                           IOException
        Writes the props to the stream, sorted by property name if sort == true or if props is an OrderedProperties. See readProperties() for the format. Property keys and values must not contain '=' or ';', this is not checked and they are not escaped Keys and values must be 255 bytes or less, Formatted length must not exceed 65535 bytes Properties from the defaults table of props (if any) are not written out by this method. jrandom disabled UTF-8 in mid-2004, for performance reasons, i.e. slow foo.getBytes("UTF-8") Re-enable it so we can pass UTF-8 tunnel names through the I2CP SessionConfig. Use utf8 = false for RouterAddress (fast, non UTF-8) Use utf8 = true for SessionConfig (slow, UTF-8)
        Parameters:
        props - source may be null
        sort - should we sort the properties? (set to false if already sorted, e.g. OrderedProperties)
        Throws:
        DataFormatException - if any string is over 255 bytes long, or if the total length (not including the two length bytes) is greater than 65535 bytes.
        IOException
        Since:
        0.8.7
      • fromProperties

        public static int fromProperties​(byte[] source,
                                         int offset,
                                         Properties target)
                                  throws DataFormatException
        Reads the props from the byte array and puts them in the Properties target See readProperties() for the format. Warning - confusing method name, Properties is the target. Strings must be UTF-8 encoded in the byte array. As of 0.9.18, throws DataFormatException on duplicate key
        Parameters:
        source - source
        target - returned Properties
        Returns:
        new offset
        Throws:
        DataFormatException
      • toProperties

        public static byte[] toProperties​(Properties opts)
                                   throws DataFormatException
        Writes the props to returned byte array, not sorted (unless the opts param is an OrderedProperties) Strings will be UTF-8 encoded in the byte array. See readProperties() for the format. Property keys and values must not contain '=' or ';', this is not checked and they are not escaped Keys and values must be 255 bytes or less, Formatted length must not exceed 65535 bytes Warning - confusing method name, Properties is the source. Properties from the defaults table of props (if any) are not written out by this method.
        Throws:
        DataFormatException - if key, value, or total is too long
      • toString

        public static String toString​(Properties options)
        Pretty print the mapping, unsorted (unless the options param is an OrderedProperties)
      • toString

        public static String toString​(Map<?,​?> options)
        Pretty print the mapping, unsorted (unless the options param is an OrderedProperties)
        Since:
        0.9.4, as of 0.9.38 supports non-String values
      • loadProps

        public static void loadProps​(Properties props,
                                     File file)
                              throws IOException
        A more efficient Properties.load Some of the other differences: - UTF-8 encoding, not ISO-8859-1 - No escaping! This does not process or drop backslashes - '#' or ';' starts a comment line, but '!' does not - Leading whitespace is not trimmed - '=' is the only key-termination character (not ':' or whitespace) As of 0.9.10, an empty value is allowed. As in Java Properties, duplicate keys are allowed, last one wins.
        Throws:
        IOException
      • loadProps

        public static void loadProps​(Properties props,
                                     File file,
                                     boolean forceLowerCase)
                              throws IOException
        Parameters:
        forceLowerCase - if true forces the keys to lower case (not the values)
        Throws:
        IOException
      • loadProps

        public static void loadProps​(Properties props,
                                     InputStream inStr,
                                     boolean forceLowerCase)
                              throws IOException
        Parameters:
        forceLowerCase - if true forces the keys to lower case (not the values)
        Throws:
        IOException
      • storeProps

        public static void storeProps​(Properties props,
                                      File file)
                               throws IOException
        Writes the props to the file, unsorted (unless props is an OrderedProperties) Note that this does not escape the \r or \n that are unescaped in loadProps() above. As of 0.8.1, file will be mode 600. Properties from the defaults table of props (if any) are not written out by this method. Leading or trailing whitespace in values is not checked but will be trimmed by loadProps()
        Throws:
        IllegalArgumentException - if a key contains any of "#=\n" or starts with ';', or a value contains '#' or '\n'
        IOException
      • toString

        public static String toString​(Collection<?> col)
        Pretty print the collection
      • toString

        public static String toString​(byte[] buf)
        Lower-case hex with leading zeros. Use toHexString(byte[]) to not get leading zeros
        Parameters:
        buf - may be null (returns "")
        Returns:
        String of length 2*buf.length
      • toString

        public static String toString​(byte[] buf,
                                      int len)
        Lower-case hex with leading zeros. Use toHexString(byte[]) to not get leading zeros
        Parameters:
        buf - may be null
        len - number of bytes. If greater than buf.length, additional zeros will be prepended
        Returns:
        String of length 2*len
      • toDecimalString

        public static String toDecimalString​(byte[] buf,
                                             int len)
        Deprecated.
        unused
        Positive decimal without leading zeros.
        Parameters:
        buf - may be null (returns "0")
        len - unused
        Returns:
        (new BigInteger(1, buf)).toString()
      • toHexString

        public static final String toHexString​(byte[] data)
        Lower-case hex without leading zeros. Use toString(byte[]) to get leading zeros
        Parameters:
        data - may be null (returns "00")
      • fromHexString

        public static final byte[] fromHexString​(String val)
        Deprecated.
        unused
        Parameters:
        val - non-null, may have leading minus sign
        Returns:
        minimum-length representation (with possible leading 0 byte)
      • readLong

        public static long readLong​(InputStream rawStream,
                                    int numBytes)
                             throws DataFormatException,
                                    IOException
        Read the stream for an integer as defined by the I2P data structure specification. Integers are a fixed number of bytes (numBytes), stored as unsigned integers in network byte order.
        Parameters:
        rawStream - stream to read from
        numBytes - number of bytes to read and format into a number, 1 to 8
        Returns:
        number
        Throws:
        DataFormatException - if negative (only possible if numBytes = 8) (since 0.8.12)
        EOFException - since 0.8.2, if there aren't enough bytes to read the number
        IOException - if there is an IO error reading the number
      • writeLong

        public static void writeLong​(OutputStream rawStream,
                                     int numBytes,
                                     long value)
                              throws DataFormatException,
                                     IOException
        Write an integer as defined by the I2P data structure specification to the stream. Integers are a fixed number of bytes (numBytes), stored as unsigned integers in network byte order.
        Parameters:
        value - value to write out, non-negative
        rawStream - stream to write to
        numBytes - number of bytes to write the number into, 1-8 (padding as necessary)
        Throws:
        DataFormatException - if value is negative or if numBytes not 1-8
        IOException - if there is an IO error writing to the stream
      • toLongLE

        public static void toLongLE​(byte[] target,
                                    int offset,
                                    int numBytes,
                                    long value)
        Little endian, i.e. backwards. Not for use in I2P protocols.
        Parameters:
        numBytes - 1-8
        value - non-negative
        Since:
        0.8.12
      • fromLong

        public static long fromLong​(byte[] src,
                                    int offset,
                                    int numBytes)
        Big endian.
        Parameters:
        src - if null returns 0
        numBytes - 1-8
        Returns:
        non-negative
        Throws:
        ArrayIndexOutOfBoundsException
        IllegalArgumentException - if negative (only possible if numBytes = 8)
      • fromLongLE

        public static long fromLongLE​(byte[] src,
                                      int offset,
                                      int numBytes)
        Little endian, i.e. backwards. Not for use in I2P protocols.
        Parameters:
        numBytes - 1-8
        Returns:
        non-negative
        Throws:
        ArrayIndexOutOfBoundsException
        IllegalArgumentException - if negative (only possible if numBytes = 8)
        Since:
        0.8.12
      • fromLong8

        public static long fromLong8​(byte[] src,
                                     int offset)
        Big endian. Same as fromLong(src, offset, 8) but allows negative result
        Throws:
        ArrayIndexOutOfBoundsException
        Since:
        0.9.47 moved from NTCP2Payload
      • toLong8

        public static void toLong8​(byte[] target,
                                   int offset,
                                   long value)
        Big endian. Same as toLong(target, offset, 8, value) but allows negative value
        Throws:
        ArrayIndexOutOfBoundsException
        Since:
        0.9.47 moved from NTCP2Payload
      • readDate

        public static Date readDate​(InputStream in)
                             throws DataFormatException,
                                    IOException
        Read in a date from the stream as specified by the I2P data structure spec. A date is an 8 byte unsigned integer in network byte order specifying the number of milliseconds since midnight on January 1, 1970 in the GMT timezone. If the number is 0, the date is undefined or null. (yes, this means you can't represent midnight on 1/1/1970)
        Parameters:
        in - stream to read from
        Returns:
        date read, or null
        Throws:
        DataFormatException - if the stream doesn't contain a validly formatted date
        IOException - if there is an IO error reading the date
      • readString

        public static String readString​(InputStream in)
                                 throws DataFormatException,
                                        IOException
        Read in a string from the stream as specified by the I2P data structure spec. A string is 1 or more bytes where the first byte is the number of bytes (not characters!) in the string and the remaining 0-255 bytes are the non-null terminated UTF-8 encoded character array.
        Parameters:
        in - stream to read from
        Returns:
        UTF-8 string
        Throws:
        DataFormatException - if the stream doesn't contain a validly formatted string
        EOFException - since 0.8.2, if there aren't enough bytes to read the string
        IOException - if there is an IO error reading the string
      • writeString

        public static void writeString​(OutputStream out,
                                       String string)
                                throws DataFormatException,
                                       IOException
        Write out a string to the stream as specified by the I2P data structure spec. Note that the max size for a string allowed by the spec is 255 bytes. WARNING - this method destroys the encoding, and therefore violates the data structure spec.
        Parameters:
        out - stream to write string
        string - string to write out: null strings are perfectly valid, but strings of excess length will cause a DataFormatException to be thrown
        Throws:
        DataFormatException - if the string is not valid
        IOException - if there is an IO error writing the string
      • writeStringUTF8

        public static void writeStringUTF8​(OutputStream out,
                                           String string)
                                    throws DataFormatException,
                                           IOException
        Write out a string to the stream as specified by the I2P data structure spec. Note that the max size for a string allowed by the spec is 255 bytes. This method correctly uses UTF-8
        Parameters:
        out - stream to write string
        string - UTF-8 string to write out: null strings are perfectly valid, but strings of excess length will cause a DataFormatException to be thrown
        Throws:
        DataFormatException - if the string is not valid
        IOException - if there is an IO error writing the string
        Since:
        public since 0.9.26
      • eq

        public static final boolean eq​(Object lhs,
                                       Object rhs)
        Helper util to compare two objects, including null handling.

        This treats (null == null) as true, and (null == (!null)) as false.

      • eq

        public static final boolean eq​(Collection<?> lhs,
                                       Collection<?> rhs)
        Run a deep comparison across the two collections.

        This treats (null == null) as true, (null == (!null)) as false, and then comparing each element via eq(object, object).

        If the size of the collections are not equal, the comparison returns false. The collection order should be consistent, as this simply iterates across both and compares based on the value of each at each step along the way.

      • eq

        public static final boolean eq​(byte[] lhs,
                                       byte[] rhs)
        Run a comparison on the byte arrays, byte by byte.

        This treats (null == null) as true, (null == (!null)) as false, and unequal length arrays as false. Variable time.

        Returns:
        Arrays.equals(lhs, rhs)
      • eq

        @Deprecated
        public static final boolean eq​(int lhs,
                                       int rhs)
        Deprecated.
        inefficient
        Compare two integers, really just for consistency.
      • eq

        @Deprecated
        public static final boolean eq​(long lhs,
                                       long rhs)
        Deprecated.
        inefficient
        Compare two longs, really just for consistency.
      • eq

        @Deprecated
        public static final boolean eq​(byte lhs,
                                       byte rhs)
        Deprecated.
        inefficient
        Compare two bytes, really just for consistency.
      • eq

        public static final boolean eq​(byte[] lhs,
                                       int offsetLeft,
                                       byte[] rhs,
                                       int offsetRight,
                                       int length)
        Unlike eq(byte[], byte[]), this returns false if either lhs or rhs is null. Variable time.
        Throws:
        ArrayIndexOutOfBoundsException - if either array isn't long enough
      • eqCT

        public static final boolean eqCT​(byte[] lhs,
                                         int offsetLeft,
                                         byte[] rhs,
                                         int offsetRight,
                                         int length)
        Unlike eq(), this throws NPE if either lhs or rhs is null. Constant time.
        Throws:
        NullPointerException - if lhs or rhs is null
        ArrayIndexOutOfBoundsException - if either array isn't long enough
        Since:
        0.9.13
      • compareTo

        public static final int compareTo​(byte[] lhs,
                                          byte[] rhs)
        Big endian compare, treats bytes as unsigned. Shorter arg is lesser. Args may be null, null is less than non-null. Variable time.
      • xor

        public static final byte[] xor​(byte[] lhs,
                                       byte[] rhs)
        Returns:
        null if either arg is null or the args are not equal length
      • xor

        public static final void xor​(byte[] lhs,
                                     int startLeft,
                                     byte[] rhs,
                                     int startRight,
                                     byte[] out,
                                     int startOut,
                                     int len)
        xor the lhs with the rhs, storing the result in out.
        Parameters:
        lhs - one of the source arrays
        startLeft - starting index in the lhs array to begin the xor
        rhs - the other source array
        startRight - starting index in the rhs array to begin the xor
        out - output array
        startOut - starting index in the out array to store the result
        len - how many bytes into the various arrays to xor
      • hashCode

        public static int hashCode​(Object obj)
        Calculate the hashcode of the object, using 0 for null
      • hashCode

        public static int hashCode​(Date obj)
        Calculate the hashcode of the date, using 0 for null
      • hashCode

        public static int hashCode​(byte[] b)
        Calculate the hashcode of the byte array, using 0 for null
      • hashCode

        public static int hashCode​(Collection<?> col)
        Calculate the hashcode of the collection, using 0 for null
      • skip

        public static void skip​(InputStream in,
                                long n)
                         throws IOException
        This is different than InputStream.skip(), in that it does repeated reads until the full amount is skipped. To fix findbugs issues with skip(). Guaranteed to skip exactly n bytes or throw an IOE. http://stackoverflow.com/questions/14057720/robust-skipping-of-data-in-a-java-io-inputstream-and-its-subtypes http://stackoverflow.com/questions/11511093/java-inputstream-skip-return-value-near-end-of-file
        Throws:
        IOException
        Since:
        0.9.9
      • read

        public static int read​(InputStream in,
                               byte[] target)
                        throws IOException
        This is different than InputStream.read(target), in that it does repeated reads until the full data is received. As of 0.9.27, throws EOFException if the full length is not read.
        Returns:
        target.length
        Throws:
        EOFException - if the full length is not read (since 0.9.27)
        IOException
      • read

        public static int read​(InputStream in,
                               byte[] target,
                               int offset,
                               int length)
                        throws IOException
        WARNING - This is different than InputStream.read(target, offset, length) for a nonzero offset, in that it returns the new offset (== old offset + length). It also does repeated reads until the full data is received. WARNING - Broken for nonzero offset before 0.9.27. As of 0.9.27, throws EOFException if the full length is not read.
        Returns:
        the new offset (== old offset + length)
        Throws:
        EOFException - if the full length is not read (since 0.9.27)
        IOException
      • readLine

        public static String readLine​(InputStream in)
                               throws IOException
        Read a newline delimited line from the stream, returning the line (without the newline), or null if EOF reached on an empty line Warning - strips \n but not \r Warning - 8KB line length limit as of 0.7.13, @throws IOException if exceeded Warning - not UTF-8
        Returns:
        null on EOF
        Throws:
        IOException
      • readLine

        public static String readLine​(InputStream in,
                                      MessageDigest hash)
                               throws IOException
        update the hash along the way Warning - strips \n but not \r Warning - 8KB line length limit as of 0.7.13, @throws IOException if exceeded Warning - not UTF-8
        Parameters:
        hash - null OK
        Returns:
        null on EOF
        Throws:
        IOException
        Since:
        0.8.8
      • readLine

        public static boolean readLine​(InputStream in,
                                       StringBuilder buf)
                                throws IOException
        Read in a line, placing it into the buffer (excluding the newline). Warning - strips \n but not \r Warning - 8KB line length limit as of 0.7.13, @throws IOException if exceeded Warning - not UTF-8
        Returns:
        true if the line was read, false if eof was reached on an empty line (returns true for non-empty last line without a newline)
        Throws:
        IOException
      • readLine

        public static boolean readLine​(InputStream in,
                                       StringBuilder buf,
                                       MessageDigest hash)
                                throws IOException
        update the hash along the way Warning - strips \n but not \r Warning - 8KB line length limit as of 0.7.13, @throws IOException if exceeded Warning - not UTF-8
        Parameters:
        hash - null OK
        Returns:
        true if the line was read, false if eof was reached on an empty line (returns true for non-empty last line without a newline)
        Throws:
        IOException
        Since:
        0.8.8
      • formatDuration

        public static String formatDuration​(long ms)
        NOTE: formatDuration2() recommended in most cases for readability
      • formatDuration2

        public static String formatDuration2​(long ms)
        Like formatDuration but with a non-breaking space after the number, 0 is unitless, and the unit is translated. This seems consistent with most style guides out there. Use only in HTML. Thresholds are a little lower than in formatDuration() also, as precision is less important in the GUI than in logging. Negative numbers handled correctly.
        Since:
        0.8.2
      • formatDuration2

        public static String formatDuration2​(double ms)
        Like formatDuration2(long) but with microsec and nanosec also.
        Since:
        0.9.19
      • formatSize

        public static String formatSize​(long bytes)
        This is binary, i.e. multiples of 1024. For decimal, see formatSize2Decimal(). Caller should append 'B' or 'b' as appropriate. No space between the number and the letter. NOTE: formatSize2() recommended in most cases for readability
        Returns:
        e.g. "123.05Ki"
      • formatSize2

        public static String formatSize2​(long bytes)
        This is binary, i.e. multiples of 1024. For decimal, see formatSize2Decimal(). Caller should append 'B' or 'b' as appropriate. Like formatSize but with a non-breaking space after the number This seems consistent with most style guides out there. Use only in HTML, and not inside form values (use formatSize2(bytes, false) there instead).
        Returns:
        e.g. "123.05&#8239;Ki"
        Since:
        0.7.14, uses thin non-breaking space since 0.9.31
      • formatSize2

        public static String formatSize2​(long bytes,
                                         boolean nonBreaking)
        This is binary, i.e. multiples of 1024. For decimal, see formatSize2Decimal(). Caller should append 'B' or 'b' as appropriate, Like formatSize but with a space after the number This seems consistent with most style guides out there.
        Parameters:
        nonBreaking - use an HTML thin non-breaking space (&#8239;)
        Returns:
        e.g. "123.05&#8239;Ki" or "123.05 Ki"
        Since:
        0.9.31
      • formatSize2Decimal

        public static String formatSize2Decimal​(long bytes)
        This is decimal, i.e. multiples of 1000. For binary, see formatSize2(). Caller should append 'B' or 'b' as appropriate. Like formatSize but with a space after the number This seems consistent with most style guides out there.
        Returns:
        e.g. "123.05&#8239;K"
        Since:
        0.9.34
      • formatSize2Decimal

        public static String formatSize2Decimal​(long bytes,
                                                boolean nonBreaking)
        This is decimal, i.e. multiples of 1000. For binary, see formatSize2(). Caller should append 'B' or 'b' as appropriate. Like formatSize but with a space after the number This seems consistent with most style guides out there.
        Parameters:
        nonBreaking - use an HTML thin non-breaking space (&#8239;)
        Returns:
        e.g. "123.05&#8239;K" or "123.05 K"
        Since:
        0.9.34
      • formatDate

        public static String formatDate​(long now)
        The default formatting for date, current locale, local time zone. Warning - NOT UTC! Examples: en: Aug 30, 2019 de: 30.08.2019
        Since:
        0.9.43
      • formatTime

        public static String formatTime​(long now)
        The default formatting for date/time, current locale, local time zone. Warning - NOT UTC! Examples: en: Aug 30, 2019 12:38 PM de: 30.08.2019 12:38
        Since:
        0.9.43
      • stripHTML

        public static String stripHTML​(String orig)
        Strip out any HTML (simply removing any less than / greater than symbols)
        Parameters:
        orig - may be null, returns empty string if null
      • escapeHTML

        public static String escapeHTML​(String unescaped)
        Escape a string for inclusion in HTML
        Parameters:
        unescaped - the unescaped string, may be null
        Returns:
        the escaped string, or null if null is passed in
      • compress

        public static byte[] compress​(byte[] orig)
        Compress the data and return a new GZIP compressed byte array. The compressed data conforms to RFC 1952, with a 10-byte gzip header and a 8-byte gzip checksum footer. Prior to 0.9.29, this would return a zero-length output for a zero-length input. As of 0.9.29, output is valid for a zero-length input also. As of 0.9.47, this uses a level of MEDIUM_COMPRESSION, which is a good space/speed tradeoff. Prior to that, it used MAX_COMPRESSION.
        Returns:
        null if orig is null
        Throws:
        IllegalArgumentException - if input size is over 40KB
        IllegalStateException - on compression failure, as of 0.9.29
      • compress

        public static byte[] compress​(byte[] orig,
                                      int offset,
                                      int size)
        Compress the data and return a new GZIP compressed byte array. The compressed data conforms to RFC 1952, with a 10-byte gzip header and a 8-byte gzip checksum footer. Prior to 0.9.29, this would return a zero-length output for a zero-length input. As of 0.9.29, output is valid for a zero-length input also. As of 0.9.47, this uses a level of MEDIUM_COMPRESSION, which is a good space/speed tradeoff. Prior to that, it used MAX_COMPRESSION.
        Returns:
        null if orig is null
        Throws:
        IllegalArgumentException - if size is over 40KB
        IllegalStateException - on compression failure, as of 0.9.29
      • compress

        public static byte[] compress​(byte[] orig,
                                      int offset,
                                      int size,
                                      int level)
        Compress the data and return a new GZIP compressed byte array. The compressed data conforms to RFC 1952, with a 10-byte gzip header and a 8-byte gzip checksum footer. Prior to 0.9.29, this would return a zero-length output for a zero-length input. As of 0.9.29, output is valid for a zero-length input also.
        Parameters:
        level - the compression level, 0 to 9
        Returns:
        null if orig is null
        Throws:
        IllegalArgumentException - if size is over 40KB
        IllegalStateException - on compression failure, as of 0.9.29
      • decompress

        public static byte[] decompress​(byte[] orig)
                                 throws IOException
        Decompress the GZIP compressed data (returning null on error).
        Returns:
        null if orig is null
        Throws:
        IOException - if uncompressed is over 40 KB, or on a decompression error
      • decompress

        public static byte[] decompress​(byte[] orig,
                                        int offset,
                                        int length)
                                 throws IOException
        Decompress the GZIP compressed data (returning null on error).
        Returns:
        null if orig is null
        Throws:
        IOException - if uncompressed is over 40 KB, or on a decompression error
      • getUTF8

        public static byte[] getUTF8​(String orig)
        Same as orig.getBytes("UTF-8") but throws an unchecked RuntimeException instead of an UnsupportedEncodingException if no UTF-8, for ease of use.
        Returns:
        null if orig is null
        Throws:
        RuntimeException
      • getUTF8

        public static byte[] getUTF8​(StringBuffer orig)
        Deprecated.
        unused
        Same as orig.getBytes("UTF-8") but throws an unchecked RuntimeException instead of an UnsupportedEncodingException if no UTF-8, for ease of use.
        Returns:
        null if orig is null
        Throws:
        RuntimeException
      • getUTF8

        public static String getUTF8​(byte[] orig)
        Same as new String(orig, "UTF-8") but throws an unchecked RuntimeException instead of an UnsupportedEncodingException if no UTF-8, for ease of use. Used by Syndie.
        Returns:
        null if orig is null
        Throws:
        RuntimeException
      • getUTF8

        public static String getUTF8​(byte[] orig,
                                     int offset,
                                     int len)
        Same as new String(orig, "UTF-8") but throws an unchecked RuntimeException instead of an UnsupportedEncodingException if no UTF-8, for ease of use.
        Returns:
        null if orig is null
        Throws:
        RuntimeException
      • getASCII

        public static byte[] getASCII​(String orig)
        Roughly the same as orig.getBytes("ISO-8859-1") but much faster and will not throw an exception. Warning - misnamed, converts to ISO-8859-1.
        Parameters:
        orig - non-null, truncates to 8-bit chars
        Since:
        0.9.5
      • split

        public static String[] split​(String s,
                                     String regex)
        Same as s.split(regex) but caches the compiled pattern for speed. This saves about 10 microseconds (Bulldozer) on subsequent invocations. Note: For an input "" this returns [""], not a zero-length array. This is the same behavior as String.split().
        Parameters:
        s - non-null
        regex - non-null, don't forget to enclose multiple choices with []
        Throws:
        PatternSyntaxException - unchecked
        Since:
        0.9.24
      • split

        public static String[] split​(String s,
                                     String regex,
                                     int limit)
        Same as s.split(regex, limit) but caches the compiled pattern for speed. This saves about 10 microseconds (Bulldozer) on subsequent invocations. Note: For an input "" this returns [""], not a zero-length array. This is the same behavior as String.split().
        Parameters:
        s - non-null
        regex - non-null, don't forget to enclose multiple choices with []
        limit - result threshold
        Throws:
        PatternSyntaxException - unchecked
        Since:
        0.9.24
      • sort

        public static <T> void sort​(List<T> list,
                                    Comparator<? super T> c)
        Same as Collections.sort(), but guaranteed not to throw an IllegalArgumentException if the sort is unstable. As of Java 7, TimSort will throw an IAE if the underlying sort order changes during the sort. This catches the IAE, retries once, and then returns. If an IAE is thrown twice, this method will return, with the list possibly unsorted.
        Parameters:
        list - the list to be sorted.
        c - the comparator to determine the order of the list. A null value indicates that the elements' natural ordering should be used.
        Since:
        0.9.34
      • sort

        public static <T> void sort​(T[] a,
                                    Comparator<? super T> c)
        Same as Arrays.sort(), but guaranteed not to throw an IllegalArgumentException if the sort is unstable. As of Java 7, TimSort will throw an IAE if the underlying sort order changes during the sort. This catches the IAE, retries once, and then returns. If an IAE is thrown twice, this method will return, with the array possibly unsorted.
        Parameters:
        a - the array to be sorted.
        c - the comparator to determine the order of the array. A null value indicates that the elements' natural ordering should be used.
        Since:
        0.9.34
      • replace

        public static void replace​(StringBuilder buf,
                                   String from,
                                   String to)
        Replace all instances of "from" with "to" in the StringBuilder buf. Same as String.replace(), but in-memory with no object churn, as long as "to" is equal size or smaller than "from", or buf has capacity. Use for large Strings or for multiple replacements in a row.
        Parameters:
        buf - contains the string to be searched
        from - the string to be replaced
        to - the replacement string
        Since:
        0.9.34