oshi.util.ParseUtil Maven / Gradle / Ivy
/*
* Copyright 2016-2022 The OSHI Project Contributors
* SPDX-License-Identifier: MIT
*/
package oshi.util;
import java.math.BigInteger;
import java.net.InetAddress;
import java.net.UnknownHostException;
import java.nio.ByteBuffer;
import java.nio.ByteOrder;
import java.nio.charset.StandardCharsets;
import java.time.LocalTime;
import java.time.OffsetDateTime;
import java.time.format.DateTimeFormatter;
import java.time.format.DateTimeParseException;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.EnumMap;
import java.util.EnumSet;
import java.util.HashMap;
import java.util.LinkedHashMap;
import java.util.List;
import java.util.Locale;
import java.util.Map;
import java.util.TimeZone;
import java.util.regex.Matcher;
import java.util.regex.Pattern;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import oshi.annotation.concurrent.ThreadSafe;
import oshi.util.tuples.Pair;
import oshi.util.tuples.Triplet;
/**
* String parsing utility.
*/
@ThreadSafe
public final class ParseUtil {
private static final Logger LOG = LoggerFactory.getLogger(ParseUtil.class);
private static final String DEFAULT_LOG_MSG = "{} didn't parse. Returning default. {}";
/*
* Used for matching
*/
private static final Pattern HERTZ_PATTERN = Pattern.compile("(\\d+(.\\d+)?) ?([kKMGT]?Hz).*");
private static final Pattern BYTES_PATTERN = Pattern.compile("(\\d+) ?([kKMGT]?B?).*");
private static final Pattern UNITS_PATTERN = Pattern.compile("(\\d+(.\\d+)?)[\\s]?([kKMGT])?");
/*
* Used to check validity of a hexadecimal string
*/
private static final Pattern VALID_HEX = Pattern.compile("[0-9a-fA-F]+");
/*
* Pattern for [dd-[hh:[mm:[ss[.sss]]]]]
*/
private static final Pattern DHMS = Pattern.compile("(?:(\\d+)-)?(?:(\\d+):)??(?:(\\d+):)?(\\d+)(?:\\.(\\d+))?");
/*
* Pattern for a UUID
*/
private static final Pattern UUID_PATTERN = Pattern
.compile(".*([0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12}).*");
/*
* Pattern for Windows DeviceID vendor and product ID and serial
*/
private static final Pattern VENDOR_PRODUCT_ID_SERIAL = Pattern
.compile(".*(?:VID|VEN)_(\\p{XDigit}{4})&(?:PID|DEV)_(\\p{XDigit}{4})(.*)\\\\(.*)");
/*
* Pattern for Linux lspci machine readable
*/
private static final Pattern LSPCI_MACHINE_READABLE = Pattern.compile("(.+)\\s\\[(.*?)\\]");
/*
* Pattern for Linux lspci memory
*/
private static final Pattern LSPCI_MEMORY_SIZE = Pattern.compile(".+\\s\\[size=(\\d+)([kKMGT])\\]");
/*
* Hertz related variables.
*/
private static final String HZ = "Hz";
private static final String KHZ = "kHz";
private static final String MHZ = "MHz";
private static final String GHZ = "GHz";
private static final String THZ = "THz";
private static final String PHZ = "PHz";
private static final Map multipliers;
// PDH timestamps are 1601 epoch, local time
// Constants to convert to UTC millis
private static final long EPOCH_DIFF = 11_644_473_600_000L;
private static final int TZ_OFFSET = TimeZone.getDefault().getOffset(System.currentTimeMillis());
/** Constant whitespacesColonWhitespace */
public static final Pattern whitespacesColonWhitespace = Pattern.compile("\\s+:\\s");
/** Constant whitespaces */
public static final Pattern whitespaces = Pattern.compile("\\s+");
/** Constant notDigits */
public static final Pattern notDigits = Pattern.compile("[^0-9]+");
/** Constant startWithNotDigits */
public static final Pattern startWithNotDigits = Pattern.compile("^[^0-9]*");
/** Constant forwardSlash */
public static final Pattern slash = Pattern.compile("\\/");
static {
multipliers = new HashMap<>();
multipliers.put(HZ, 1L);
multipliers.put(KHZ, 1_000L);
multipliers.put(MHZ, 1_000_000L);
multipliers.put(GHZ, 1_000_000_000L);
multipliers.put(THZ, 1_000_000_000_000L);
multipliers.put(PHZ, 1_000_000_000_000_000L);
}
// Fast decimal exponentiation: pow(10,y) --> POWERS_OF_10[y]
private static final long[] POWERS_OF_TEN = { 1L, 10L, 100L, 1_000L, 10_000L, 100_000L, 1_000_000L, 10_000_000L,
100_000_000L, 1_000_000_000L, 10_000_000_000L, 100_000_000_000L, 1_000_000_000_000L, 10_000_000_000_000L,
100_000_000_000_000L, 1_000_000_000_000_000L, 10_000_000_000_000_000L, 100_000_000_000_000_000L,
1_000_000_000_000_000_000L };
// Format returned by WMI for DateTime
private static final DateTimeFormatter CIM_FORMAT = DateTimeFormatter.ofPattern("yyyyMMddHHmmss.SSSSSSZZZZZ",
Locale.US);
private ParseUtil() {
}
/**
* Parse hertz from a string, eg. "2.00MHz" is 2000000L.
*
* @param hertz Hertz size.
* @return {@link java.lang.Long} Hertz value or -1 if not parseable.
*/
public static long parseHertz(String hertz) {
Matcher matcher = HERTZ_PATTERN.matcher(hertz.trim());
if (matcher.find() && matcher.groupCount() == 3) {
// Regexp enforces #(.#) format so no test for NFE required
double value = Double.valueOf(matcher.group(1)) * multipliers.getOrDefault(matcher.group(3), -1L);
if (value >= 0d) {
return (long) value;
}
}
return -1L;
}
/**
* Parse the last element of a space-delimited string to a value
*
* @param s The string to parse
* @param i Default integer if not parsable
* @return value or the given default if not parsable
*/
public static int parseLastInt(String s, int i) {
try {
String ls = parseLastString(s);
if (ls.toLowerCase().startsWith("0x")) {
return Integer.decode(ls);
} else {
return Integer.parseInt(ls);
}
} catch (NumberFormatException e) {
LOG.trace(DEFAULT_LOG_MSG, s, e);
return i;
}
}
/**
* Parse the last element of a space-delimited string to a value
*
* @param s The string to parse
* @param li Default long integer if not parsable
* @return value or the given default if not parsable
*/
public static long parseLastLong(String s, long li) {
try {
String ls = parseLastString(s);
if (ls.toLowerCase().startsWith("0x")) {
return Long.decode(ls);
} else {
return Long.parseLong(ls);
}
} catch (NumberFormatException e) {
LOG.trace(DEFAULT_LOG_MSG, s, e);
return li;
}
}
/**
* Parse the last element of a space-delimited string to a value
*
* @param s The string to parse
* @param d Default double if not parsable
* @return value or the given default if not parsable
*/
public static double parseLastDouble(String s, double d) {
try {
return Double.parseDouble(parseLastString(s));
} catch (NumberFormatException e) {
LOG.trace(DEFAULT_LOG_MSG, s, e);
return d;
}
}
/**
* Parse the last element of a space-delimited string to a string
*
* @param s The string to parse
* @return last space-delimited element
*/
public static String parseLastString(String s) {
String[] ss = whitespaces.split(s);
// guaranteed at least one element
return ss[ss.length - 1];
}
/**
* Parse a byte array into a string of hexadecimal digits including all array bytes as digits
*
* @param bytes The byte array to represent
* @return A string of hex characters corresponding to the bytes. The string is upper case.
*/
public static String byteArrayToHexString(byte[] bytes) {
StringBuilder sb = new StringBuilder(bytes.length * 2);
for (byte b : bytes) {
sb.append(Character.forDigit((b & 0xf0) >>> 4, 16));
sb.append(Character.forDigit(b & 0x0f, 16));
}
return sb.toString().toUpperCase();
}
/**
* Parse a string of hexadecimal digits into a byte array
*
* @param digits The string to be parsed
* @return a byte array with each pair of characters converted to a byte, or empty array if the string is not valid
* hex
*/
public static byte[] hexStringToByteArray(String digits) {
int len = digits.length();
// Check if string is valid hex
if (!VALID_HEX.matcher(digits).matches() || (len & 0x1) != 0) {
LOG.warn("Invalid hexadecimal string: {}", digits);
return new byte[0];
}
byte[] data = new byte[len / 2];
for (int i = 0; i < len; i += 2) {
data[i / 2] = (byte) (Character.digit(digits.charAt(i), 16) << 4
| Character.digit(digits.charAt(i + 1), 16));
}
return data;
}
/**
* Parse a human readable ASCII string into a byte array, truncating or padding with zeros (if necessary) so the
* array has the specified length.
*
* @param text The string to be parsed
* @param length Length of the returned array.
* @return A byte array of specified length, with each of the first length characters converted to a byte. If length
* is longer than the provided string length, will be filled with zeroes.
*/
public static byte[] asciiStringToByteArray(String text, int length) {
return Arrays.copyOf(text.getBytes(StandardCharsets.US_ASCII), length);
}
/**
* Convert a long value to a byte array using Big Endian, truncating or padding with zeros (if necessary) so the
* array has the specified length.
*
* @param value The value to be converted
* @param valueSize Number of bytes representing the value
* @param length Number of bytes to return
* @return A byte array of specified length representing the long in the first valueSize bytes
*/
public static byte[] longToByteArray(long value, int valueSize, int length) {
long val = value;
// Convert the long to 8-byte BE representation
byte[] b = new byte[8];
for (int i = 7; i >= 0 && val != 0L; i--) {
b[i] = (byte) val;
val >>>= 8;
}
// Then copy the rightmost valueSize bytes
// e.g., for an integer we want rightmost 4 bytes
return Arrays.copyOfRange(b, 8 - valueSize, 8 + length - valueSize);
}
/**
* Convert a string to an integer representation.
*
* @param str A human readable ASCII string
* @param size Number of characters to convert to the long. May not exceed 8.
* @return An integer representing the string where each character is treated as a byte
*/
public static long strToLong(String str, int size) {
return byteArrayToLong(str.getBytes(StandardCharsets.US_ASCII), size);
}
/**
* Convert a byte array to its (long) integer representation assuming big endian ordering.
*
* @param bytes An array of bytes no smaller than the size to be converted
* @param size Number of bytes to convert to the long. May not exceed 8.
* @return A long integer representing the byte array
*/
public static long byteArrayToLong(byte[] bytes, int size) {
return byteArrayToLong(bytes, size, true);
}
/**
* Convert a byte array to its (long) integer representation in the specified endianness.
*
* @param bytes An array of bytes no smaller than the size to be converted
* @param size Number of bytes to convert to the long. May not exceed 8.
* @param bigEndian True to parse big-endian, false to parse little-endian
* @return An long integer representing the byte array
*/
public static long byteArrayToLong(byte[] bytes, int size, boolean bigEndian) {
if (size > 8) {
throw new IllegalArgumentException("Can't convert more than 8 bytes.");
}
if (size > bytes.length) {
throw new IllegalArgumentException("Size can't be larger than array length.");
}
long total = 0L;
for (int i = 0; i < size; i++) {
if (bigEndian) {
total = total << 8 | bytes[i] & 0xff;
} else {
total = total << 8 | bytes[size - i - 1] & 0xff;
}
}
return total;
}
/**
* Convert a byte array to its floating point representation.
*
* @param bytes An array of bytes no smaller than the size to be converted
* @param size Number of bytes to convert to the float. May not exceed 8.
* @param fpBits Number of bits representing the decimal
* @return A float; the integer portion representing the byte array as an integer shifted by the bits specified in
* fpBits; with the remaining bits used as a decimal
*/
public static float byteArrayToFloat(byte[] bytes, int size, int fpBits) {
return byteArrayToLong(bytes, size) / (float) (1 << fpBits);
}
/**
* Convert an unsigned integer to a long value. The method assumes that all bits in the specified integer value are
* 'data' bits, including the most-significant bit which Java normally considers a sign bit. The method must be used
* only when it is certain that the integer value represents an unsigned integer, for example when the integer is
* returned by JNA library in a structure which holds unsigned integers.
*
* @param unsignedValue The unsigned integer value to convert.
* @return The unsigned integer value widened to a long.
*/
public static long unsignedIntToLong(int unsignedValue) {
// use standard Java widening conversion to long which does
// sign-extension,
// then drop any copies of the sign bit, to prevent the value being
// considered a negative one by Java if it is set
long longValue = unsignedValue;
return longValue & 0xffff_ffffL;
}
/**
* Convert an unsigned long to a signed long value by stripping the sign bit. This method "rolls over" long values
* greater than the max value but ensures the result is never negative.
*
* @param unsignedValue The unsigned long value to convert.
* @return The signed long value.
*/
public static long unsignedLongToSignedLong(long unsignedValue) {
return unsignedValue & 0x7fff_ffff_ffff_ffffL;
}
/**
* Parses a string of hex digits to a string where each pair of hex digits represents an ASCII character
*
* @param hexString A sequence of hex digits
* @return The corresponding string if valid hex; otherwise the original hexString
*/
public static String hexStringToString(String hexString) {
// Odd length strings won't parse, return
if (hexString.length() % 2 > 0) {
return hexString;
}
int charAsInt;
StringBuilder sb = new StringBuilder();
try {
for (int pos = 0; pos < hexString.length(); pos += 2) {
charAsInt = Integer.parseInt(hexString.substring(pos, pos + 2), 16);
if (charAsInt < 32 || charAsInt > 127) {
return hexString;
}
sb.append((char) charAsInt);
}
} catch (NumberFormatException e) {
LOG.trace(DEFAULT_LOG_MSG, hexString, e);
// Hex failed to parse, just return the existing string
return hexString;
}
return sb.toString();
}
/**
* Attempts to parse a string to an int. If it fails, returns the default
*
* @param s The string to parse
* @param defaultInt The value to return if parsing fails
* @return The parsed int, or the default if parsing failed
*/
public static int parseIntOrDefault(String s, int defaultInt) {
try {
return Integer.parseInt(s);
} catch (NumberFormatException e) {
LOG.trace(DEFAULT_LOG_MSG, s, e);
return defaultInt;
}
}
/**
* Attempts to parse a string to a long. If it fails, returns the default
*
* @param s The string to parse
* @param defaultLong The value to return if parsing fails
* @return The parsed long, or the default if parsing failed
*/
public static long parseLongOrDefault(String s, long defaultLong) {
try {
return Long.parseLong(s);
} catch (NumberFormatException e) {
LOG.trace(DEFAULT_LOG_MSG, s, e);
return defaultLong;
}
}
/**
* Attempts to parse a string to an "unsigned" long. If it fails, returns the default
*
* @param s The string to parse
* @param defaultLong The value to return if parsing fails
* @return The parsed long containing the same 64 bits that an unsigned long would contain (which may produce a
* negative value)
*/
public static long parseUnsignedLongOrDefault(String s, long defaultLong) {
try {
return new BigInteger(s).longValue();
} catch (NumberFormatException e) {
LOG.trace(DEFAULT_LOG_MSG, s, e);
return defaultLong;
}
}
/**
* Attempts to parse a string to a double. If it fails, returns the default
*
* @param s The string to parse
* @param defaultDouble The value to return if parsing fails
* @return The parsed double, or the default if parsing failed
*/
public static double parseDoubleOrDefault(String s, double defaultDouble) {
try {
return Double.parseDouble(s);
} catch (NumberFormatException e) {
LOG.trace(DEFAULT_LOG_MSG, s, e);
return defaultDouble;
}
}
/**
* Attempts to parse a string of the form [DD-[hh:]]mm:ss[.ddd] to a number of milliseconds. If it fails, returns
* the default.
*
* @param s The string to parse
* @param defaultLong The value to return if parsing fails
* @return The parsed number of seconds, or the default if parsing fails
*/
public static long parseDHMSOrDefault(String s, long defaultLong) {
Matcher m = DHMS.matcher(s);
if (m.matches()) {
long milliseconds = 0L;
if (m.group(1) != null) {
milliseconds += parseLongOrDefault(m.group(1), 0L) * 86_400_000L;
}
if (m.group(2) != null) {
milliseconds += parseLongOrDefault(m.group(2), 0L) * 3_600_000L;
}
if (m.group(3) != null) {
milliseconds += parseLongOrDefault(m.group(3), 0L) * 60_000L;
}
milliseconds += parseLongOrDefault(m.group(4), 0L) * 1000L;
if (m.group(5) != null) {
milliseconds += (long) (1000 * parseDoubleOrDefault("0." + m.group(5), 0d));
}
return milliseconds;
}
return defaultLong;
}
/**
* Attempts to parse a UUID. If it fails, returns the default.
*
* @param s The string to parse
* @param defaultStr The value to return if parsing fails
* @return The parsed UUID, or the default if parsing fails
*/
public static String parseUuidOrDefault(String s, String defaultStr) {
Matcher m = UUID_PATTERN.matcher(s.toLowerCase());
if (m.matches()) {
return m.group(1);
}
return defaultStr;
}
/**
* Parses a string key = 'value' (string)
*
* @param line The entire string
* @return the value contained between single tick marks
*/
public static String getSingleQuoteStringValue(String line) {
return getStringBetween(line, '\'');
}
/**
* Parse a string key = "value" (string)
*
* @param line the entire string
* @return the value contained between double tick marks
*/
public static String getDoubleQuoteStringValue(String line) {
return getStringBetween(line, '"');
}
/**
* Gets a value between two characters having multiple same characters between them. Examples :
*
* - "name = 'James Gosling's Java'" returns "James Gosling's Java"
* - "pci.name = 'Realtek AC'97 Audio Device'" returns "Realtek AC'97 Audio Device"
*
*
* @param line The "key-value" pair line.
* @param c The Trailing And Leading characters of the string line
* @return : The value having the characters between them.
*/
public static String getStringBetween(String line, char c) {
int firstOcc = line.indexOf(c);
if (firstOcc < 0) {
return "";
}
return line.substring(firstOcc + 1, line.lastIndexOf(c)).trim();
}
/**
* Parses a string such as "10.12.2" or "key = 1 (0x1) (int)" to find the integer value of the first set of one or
* more consecutive digits
*
* @param line The entire string
* @return the value of first integer if any; 0 otherwise
*/
public static int getFirstIntValue(String line) {
return getNthIntValue(line, 1);
}
/**
* Parses a string such as "10.12.2" or "key = 1 (0x1) (int)" to find the integer value of the nth set of one or
* more consecutive digits
*
* @param line The entire string
* @param n Which set of integers to return
* @return the value of nth integer if any; 0 otherwise
*/
public static int getNthIntValue(String line, int n) {
// Split the string by non-digits,
String[] split = notDigits.split(startWithNotDigits.matcher(line).replaceFirst(""));
if (split.length >= n) {
return parseIntOrDefault(split[n - 1], 0);
}
return 0;
}
/**
* Removes all matching sub strings from the string. More efficient than regexp.
*
* @param original source String to remove from
* @param toRemove the sub string to be removed
* @return The string with all matching substrings removed
*/
public static String removeMatchingString(final String original, final String toRemove) {
if (original == null || original.isEmpty() || toRemove == null || toRemove.isEmpty()) {
return original;
}
int matchIndex = original.indexOf(toRemove, 0);
if (matchIndex == -1) {
return original;
}
StringBuilder buffer = new StringBuilder(original.length() - toRemove.length());
int currIndex = 0;
do {
buffer.append(original.substring(currIndex, matchIndex));
currIndex = matchIndex + toRemove.length();
matchIndex = original.indexOf(toRemove, currIndex);
} while (matchIndex != -1);
buffer.append(original.substring(currIndex));
return buffer.toString();
}
/**
* Parses a delimited string to an array of longs. Optimized for processing predictable-length arrays such as
* outputs of reliably formatted Linux proc or sys filesystem, minimizing new object creation. Users should perform
* other sanity checks of data.
*
* As a special case, non-numeric fields (such as UUIDs in OpenVZ) at the end of the list are ignored. Values
* greater than the max long value return the max long value.
*
* The indices parameters are referenced assuming the length as specified, and leading characters are ignored. For
* example, if the string is "foo 12 34 5" and the length is 3, then index 0 is 12, index 1 is 34, and index 2 is 5.
*
* @param s The string to parse
* @param indices An array indicating which indexes should be populated in the final array; other values will be
* skipped. This idex is zero-referenced assuming the rightmost delimited fields of the string
* contain the array.
* @param length The total number of elements in the string array. It is permissible for the string to have more
* elements than this; leading elements will be ignored. This should be calculated once per text
* format by {@link #countStringToLongArray}.
* @param delimiter The character to delimit by.
* @return If successful, an array of parsed longs. If parsing errors occurred, will be an array of zeros.
*/
public static long[] parseStringToLongArray(String s, int[] indices, int length, char delimiter) {
long[] parsed = new long[indices.length];
// Iterate from right-to-left of String
// Fill right to left of result array using index array
int charIndex = s.length();
int parsedIndex = indices.length - 1;
int stringIndex = length - 1;
int power = 0;
int c;
boolean delimCurrent = false;
boolean numeric = true;
boolean numberFound = false; // ignore nonnumeric at end
boolean dashSeen = false; // to flag uuids as nonnumeric
while (--charIndex > 0 && parsedIndex >= 0) {
c = s.charAt(charIndex);
if (c == delimiter) {
// first parseable number?
if (!numberFound && numeric) {
numberFound = true;
}
if (!delimCurrent) {
if (numberFound && indices[parsedIndex] == stringIndex--) {
parsedIndex--;
}
delimCurrent = true;
power = 0;
dashSeen = false;
numeric = true;
}
} else if (indices[parsedIndex] != stringIndex || c == '+' || !numeric) {
// Doesn't impact parsing, ignore
delimCurrent = false;
} else if (c >= '0' && c <= '9' && !dashSeen) {
if (power > 18 || power == 17 && c == '9' && parsed[parsedIndex] > 223_372_036_854_775_807L) {
parsed[parsedIndex] = Long.MAX_VALUE;
} else {
parsed[parsedIndex] += (c - '0') * ParseUtil.POWERS_OF_TEN[power++];
}
delimCurrent = false;
} else if (c == '-') {
parsed[parsedIndex] *= -1L;
delimCurrent = false;
dashSeen = true;
} else {
// Flag as nonnumeric and continue unless we've seen a numeric
// error on everything else
if (numberFound) {
if (!noLog(s)) {
LOG.error("Illegal character parsing string '{}' to long array: {}", s, s.charAt(charIndex));
}
return new long[indices.length];
}
parsed[parsedIndex] = 0;
numeric = false;
}
}
if (parsedIndex > 0) {
if (!noLog(s)) {
LOG.error("Not enough fields in string '{}' parsing to long array: {}", s,
indices.length - parsedIndex);
}
return new long[indices.length];
}
return parsed;
}
/**
* Test whether to log this message
*
* @param s The string to log
* @return True if the string begins with {@code NOLOG}
*/
private static boolean noLog(String s) {
return s.startsWith("NOLOG: ");
}
/**
* Parses a delimited string to count elements of an array of longs. Intended to be called once to calculate the
* {@code length} field for {@link #parseStringToLongArray}.
*
* As a special case, non-numeric fields (such as UUIDs in OpenVZ) at the end of the list are ignored.
*
* @param s The string to parse
* @param delimiter The character to delimit by
* @return The number of parsable long values which follow the last unparsable value.
*/
public static int countStringToLongArray(String s, char delimiter) {
// Iterate from right-to-left of String
// Fill right to left of result array using index array
int charIndex = s.length();
int numbers = 0;
int c;
boolean delimCurrent = false;
boolean numeric = true;
boolean dashSeen = false; // to flag uuids as nonnumeric
while (--charIndex > 0) {
c = s.charAt(charIndex);
if (c == delimiter) {
if (!delimCurrent) {
if (numeric) {
numbers++;
}
delimCurrent = true;
dashSeen = false;
numeric = true;
}
} else if (c == '+' || !numeric) {
// Doesn't impact parsing, ignore
delimCurrent = false;
} else if (c >= '0' && c <= '9' && !dashSeen) {
delimCurrent = false;
} else if (c == '-') {
delimCurrent = false;
dashSeen = true;
} else {
// we found non-digit or delimiter. If not last field, exit
if (numbers > 0) {
return numbers;
}
// Else flag as nonnumeric and continue
numeric = false;
}
}
// We got to beginning of string with only numbers, count start as a delimiter
// and exit
return numbers + 1;
}
/**
* Get a String in a line of text between two marker strings
*
* @param text Text to search for match
* @param before Start matching after this text
* @param after End matching before this text
* @return Text between the strings before and after, or empty string if either marker does not exist
*/
public static String getTextBetweenStrings(String text, String before, String after) {
String result = "";
if (text.indexOf(before) >= 0 && text.indexOf(after) >= 0) {
result = text.substring(text.indexOf(before) + before.length(), text.length());
result = result.substring(0, result.indexOf(after));
}
return result;
}
/**
* Convert a long representing filetime (100-ns since 1601 epoch) to ms since 1970 epoch
*
* @param filetime A 64-bit value equivalent to FILETIME
* @param local True if converting from a local filetime (PDH counter); false if already UTC (WMI PerfRawData
* classes)
* @return Equivalent milliseconds since the epoch
*/
public static long filetimeToUtcMs(long filetime, boolean local) {
return filetime / 10_000L - EPOCH_DIFF - (local ? TZ_OFFSET : 0L);
}
/**
* Parse a date in MM-DD-YYYY or MM/DD/YYYY to YYYY-MM-DD
*
* @param dateString The date in MM DD YYYY format
* @return The date in ISO YYYY-MM-DD format if parseable, or the original string
*/
public static String parseMmDdYyyyToYyyyMmDD(String dateString) {
try {
// Date is MM-DD-YYYY, convert to YYYY-MM-DD
return String.format("%s-%s-%s", dateString.substring(6, 10), dateString.substring(0, 2),
dateString.substring(3, 5));
} catch (StringIndexOutOfBoundsException e) {
return dateString;
}
}
/**
* Converts a string in CIM Date Format, as returned by WMI for DateTime types, into a
* {@link java.time.OffsetDateTime}.
*
* @param cimDateTime A non-null DateTime String in CIM date format, e.g., 20160513072950.782000-420
* @return The parsed {@link java.time.OffsetDateTime} if the string is parsable, otherwise
* {@link oshi.util.Constants#UNIX_EPOCH}.
*/
public static OffsetDateTime parseCimDateTimeToOffset(String cimDateTime) {
// Keep first 22 characters: digits, decimal, and + or - sign
// But alter last 3 characters from a minute offset to hh:mm
try {
// From WMI as 20160513072950.782000-420,
int tzInMinutes = Integer.parseInt(cimDateTime.substring(22));
// modified to 20160513072950.782000-07:00 which can be parsed
LocalTime offsetAsLocalTime = LocalTime.MIDNIGHT.plusMinutes(tzInMinutes);
return OffsetDateTime.parse(
cimDateTime.substring(0, 22) + offsetAsLocalTime.format(DateTimeFormatter.ISO_LOCAL_TIME),
ParseUtil.CIM_FORMAT);
} catch (IndexOutOfBoundsException // if cimDate not 22+ chars
| NumberFormatException // if TZ minutes doesn't parse
| DateTimeParseException e) {
LOG.trace("Unable to parse {} to CIM DateTime.", cimDateTime);
return Constants.UNIX_EPOCH;
}
}
/**
* Checks if a file path equals or starts with an prefix in the given list
*
* @param prefixList A list of path prefixes
* @param path a string path to check
* @return true if the path exactly equals, or starts with one of the strings in prefixList
*/
public static boolean filePathStartsWith(List prefixList, String path) {
for (String match : prefixList) {
if (path.equals(match) || path.startsWith(match + "/")) {
return true;
}
}
return false;
}
/**
* Parses a string like "53G" or "54.904 M" to its long value.
*
* @param count A count with a multiplyer like "4096 M"
* @return the count parsed to a long
*/
public static long parseMultipliedToLongs(String count) {
Matcher matcher = UNITS_PATTERN.matcher(count.trim());
String[] mem;
if (matcher.find() && matcher.groupCount() == 3) {
mem = new String[2];
mem[0] = matcher.group(1);
mem[1] = matcher.group(3);
} else {
mem = new String[] { count };
}
double number = ParseUtil.parseDoubleOrDefault(mem[0], 0L);
if (mem.length == 2 && mem[1] != null && mem[1].length() >= 1) {
switch (mem[1].charAt(0)) {
case 'T':
number *= 1_000_000_000_000L;
break;
case 'G':
number *= 1_000_000_000L;
break;
case 'M':
number *= 1_000_000L;
break;
case 'K':
case 'k':
number *= 1_000L;
break;
default:
}
}
return (long) number;
}
/**
* Parses a string such as "4096 MB" to its long. Used to parse macOS and *nix memory chip sizes. Although the units
* given are decimal they must parse to binary units.
*
* @param size A string of memory sizes like "4096 MB"
* @return the size parsed to a long
*/
public static long parseDecimalMemorySizeToBinary(String size) {
String[] mem = ParseUtil.whitespaces.split(size);
if (mem.length < 2) {
// If no spaces, use regexp
Matcher matcher = BYTES_PATTERN.matcher(size.trim());
if (matcher.find() && matcher.groupCount() == 2) {
mem = new String[2];
mem[0] = matcher.group(1);
mem[1] = matcher.group(2);
}
}
long capacity = ParseUtil.parseLongOrDefault(mem[0], 0L);
if (mem.length == 2 && mem[1].length() > 1) {
switch (mem[1].charAt(0)) {
case 'T':
capacity <<= 40;
break;
case 'G':
capacity <<= 30;
break;
case 'M':
capacity <<= 20;
break;
case 'K':
case 'k':
capacity <<= 10;
break;
default:
break;
}
}
return capacity;
}
/**
* Parse a Windows DeviceID to get the vendor ID, product ID, and Serial Number
*
* @param deviceId The DeviceID
* @return A {@link Triplet} where the first element is the vendor ID, the second element is the product ID, and the
* third element is either a serial number or empty string if parsing was successful, or {@code null}
* otherwise
*/
public static Triplet parseDeviceIdToVendorProductSerial(String deviceId) {
Matcher m = VENDOR_PRODUCT_ID_SERIAL.matcher(deviceId);
if (m.matches()) {
String vendorId = "0x" + m.group(1).toLowerCase();
String productId = "0x" + m.group(2).toLowerCase();
String serial = m.group(4);
return new Triplet<>(vendorId, productId, !m.group(3).isEmpty() || serial.contains("&") ? "" : serial);
}
return null;
}
/**
* Parse a Linux lshw resources string to calculate the memory size
*
* @param resources A string containing one or more elements of the form {@code memory:b00000000-bffffffff}
* @return The number of bytes consumed by the memory in the {@code resources} string
*/
public static long parseLshwResourceString(String resources) {
long bytes = 0L;
// First split by whitespace
String[] resourceArray = whitespaces.split(resources);
for (String r : resourceArray) {
// Remove prefix
if (r.startsWith("memory:")) {
// Split to low and high
String[] mem = r.substring(7).split("-");
if (mem.length == 2) {
try {
// Parse the hex strings
bytes += Long.parseLong(mem[1], 16) - Long.parseLong(mem[0], 16) + 1;
} catch (NumberFormatException e) {
LOG.trace(DEFAULT_LOG_MSG, r, e);
}
}
}
}
return bytes;
}
/**
* Parse a Linux lspci machine readble line to its name and id
*
* @param line A string in the form Foo [bar]
* @return A pair separating the String before the square brackets and within them if found, null otherwise
*/
public static Pair parseLspciMachineReadable(String line) {
Matcher matcher = LSPCI_MACHINE_READABLE.matcher(line);
if (matcher.matches()) {
return new Pair<>(matcher.group(1), matcher.group(2));
}
return null;
}
/**
* Parse a Linux lspci line containing memory size
*
* @param line A string in the form Foo [size=256M]
* @return A the memory size in bytes
*/
public static long parseLspciMemorySize(String line) {
Matcher matcher = LSPCI_MEMORY_SIZE.matcher(line);
if (matcher.matches()) {
return parseDecimalMemorySizeToBinary(matcher.group(1) + " " + matcher.group(2) + "B");
}
return 0;
}
/**
* Parse a space-delimited list of integers which include hyphenated ranges to a list of just the integers. For
* example, 0 1 4-7 parses to a list containing 0, 1, 4, 5, 6, and 7.
*
* @param str A string containing space-delimited integers or ranges of integers with a hyphen
* @return A list of integers representing the provided range(s).
*/
public static List parseHyphenatedIntList(String str) {
List result = new ArrayList<>();
for (String s : whitespaces.split(str)) {
if (s.contains("-")) {
int first = getFirstIntValue(s);
int last = getNthIntValue(s, 2);
for (int i = first; i <= last; i++) {
result.add(i);
}
} else {
int only = ParseUtil.parseIntOrDefault(s, -1);
if (only >= 0) {
result.add(only);
}
}
}
return result;
}
/**
* Parse an integer in big endian IP format to its component bytes representing an IPv4 address
*
* @param ip The address as an integer
* @return The address as an array of four bytes
*/
public static byte[] parseIntToIP(int ip) {
return ByteBuffer.allocate(4).order(ByteOrder.LITTLE_ENDIAN).putInt(ip).array();
}
/**
* Parse an integer array in big endian IP format to its component bytes representing an IPv6 address
*
* @param ip6 The address as an integer array
* @return The address as an array of sizteen bytes
*/
public static byte[] parseIntArrayToIP(int[] ip6) {
ByteBuffer bb = ByteBuffer.allocate(16).order(ByteOrder.LITTLE_ENDIAN);
for (int i : ip6) {
bb.putInt(i);
}
return bb.array();
}
/**
* TCP network addresses and ports are in big endian format by definition. The order of the two bytes in the 16-bit
* unsigned short port value must be reversed
*
* @param port The port number in big endian order
* @return The port number
* @see ntohs
*/
public static int bigEndian16ToLittleEndian(int port) {
// 20480 = 0x5000 should be 0x0050 = 80
// 47873 = 0xBB01 should be 0x01BB = 443
return port >> 8 & 0xff | port << 8 & 0xff00;
}
/**
* Parse an integer array to an IPv4 or IPv6 as appropriate.
*
* Intended for use on Utmp structures's {@code ut_addr_v6} element.
*
* @param utAddrV6 An array of 4 integers representing an IPv6 address. IPv4 address uses just utAddrV6[0]
* @return A string representation of the IP address.
*/
public static String parseUtAddrV6toIP(int[] utAddrV6) {
if (utAddrV6.length != 4) {
throw new IllegalArgumentException("ut_addr_v6 must have exactly 4 elements");
}
// IPv4 has only first element
if (utAddrV6[1] == 0 && utAddrV6[2] == 0 && utAddrV6[3] == 0) {
// Special case for all 0's
if (utAddrV6[0] == 0) {
return "::";
}
// Parse using InetAddress
byte[] ipv4 = ByteBuffer.allocate(4).putInt(utAddrV6[0]).array();
try {
return InetAddress.getByAddress(ipv4).getHostAddress();
} catch (UnknownHostException e) {
// Shouldn't happen with length 4 or 16
return Constants.UNKNOWN;
}
}
// Parse all 16 bytes
byte[] ipv6 = ByteBuffer.allocate(16).putInt(utAddrV6[0]).putInt(utAddrV6[1]).putInt(utAddrV6[2])
.putInt(utAddrV6[3]).array();
try {
return InetAddress.getByAddress(ipv6).getHostAddress()
.replaceAll("((?:(?:^|:)0+\\b){2,8}):?(?!\\S*\\b\\1:0+\\b)(\\S*)", "::$2");
} catch (UnknownHostException e) {
// Shouldn't happen with length 4 or 16
return Constants.UNKNOWN;
}
}
/**
* Parses a string of hex digits to an int value.
*
* @param hexString A sequence of hex digits
* @param defaultValue default value to return if parsefails
* @return The corresponding int value
*/
public static int hexStringToInt(String hexString, int defaultValue) {
if (hexString != null) {
try {
if (hexString.startsWith("0x")) {
return new BigInteger(hexString.substring(2), 16).intValue();
} else {
return new BigInteger(hexString, 16).intValue();
}
} catch (NumberFormatException e) {
LOG.trace(DEFAULT_LOG_MSG, hexString, e);
}
}
// Hex failed to parse, just return the default long
return defaultValue;
}
/**
* Parses a string of hex digits to a long value.
*
* @param hexString A sequence of hex digits
* @param defaultValue default value to return if parsefails
* @return The corresponding long value
*/
public static long hexStringToLong(String hexString, long defaultValue) {
if (hexString != null) {
try {
if (hexString.startsWith("0x")) {
return new BigInteger(hexString.substring(2), 16).longValue();
} else {
return new BigInteger(hexString, 16).longValue();
}
} catch (NumberFormatException e) {
LOG.trace(DEFAULT_LOG_MSG, hexString, e);
}
}
// Hex failed to parse, just return the default long
return defaultValue;
}
/**
* Parses a String "....foo" to "foo"
*
* @param dotPrefixedStr A string with possibly leading dots
* @return The string without the dots
*/
public static String removeLeadingDots(String dotPrefixedStr) {
int pos = 0;
while (pos < dotPrefixedStr.length() && dotPrefixedStr.charAt(pos) == '.') {
pos++;
}
return pos < dotPrefixedStr.length() ? dotPrefixedStr.substring(pos) : "";
}
/**
* Parse a null-delimited byte array to a list of strings.
*
* @param bytes A byte array containing Strings delimited by null characters. Two consecutive null characters mark
* the end of the list.
* @return A list of Strings between the nulls.
*/
public static List parseByteArrayToStrings(byte[] bytes) {
List strList = new ArrayList<>();
int start = 0;
int end = 0;
// Iterate characters
do {
// If we've reached a delimiter or the end of the array, add to list
if (end == bytes.length || bytes[end] == 0) {
// Zero length string means two nulls, we're done
if (start == end) {
break;
}
// Otherwise add string and reset start
// Intentionally using platform default charset
strList.add(new String(bytes, start, end - start, StandardCharsets.UTF_8));
start = end + 1;
}
} while (end++ < bytes.length);
return strList;
}
/**
* Parse a null-delimited byte array to a map of string keys and values.
*
* @param bytes A byte array containing String key-value pairs with keys and values delimited by {@code =} and pairs
* delimited by null characters. Two consecutive null characters mark the end of the map.
* @return A map of String key-value pairs between the nulls.
*/
public static Map parseByteArrayToStringMap(byte[] bytes) {
// API does not specify any particular order of entries, but it is reasonable to
// maintain whatever order the OS provided to the end user
Map strMap = new LinkedHashMap<>();
int start = 0;
int end = 0;
String key = null;
// Iterate characters
do {
// If we've reached a delimiter or the end of the array, add to list
if (end == bytes.length || bytes[end] == 0) {
// Zero length string with no key, we're done
if (start == end && key == null) {
break;
}
// Otherwise add string (possibly empty) and reset start
// Intentionally using platform default charset
strMap.put(key, new String(bytes, start, end - start, StandardCharsets.UTF_8));
key = null;
start = end + 1;
} else if (bytes[end] == '=' && key == null) {
key = new String(bytes, start, end - start, StandardCharsets.UTF_8);
start = end + 1;
}
} while (end++ < bytes.length);
return strMap;
}
/**
* Parse a null-delimited char array to a map of string keys and values.
*
* @param chars A char array containing String key-value pairs with keys and values delimited by {@code =} and pairs
* delimited by null characters. Two consecutive null characters mark the end of the map.
* @return A map of String key-value pairs between the nulls.
*/
public static Map parseCharArrayToStringMap(char[] chars) {
// API does not specify any particular order of entries, but it is reasonable to
// maintain whatever order the OS provided to the end user
Map strMap = new LinkedHashMap<>();
int start = 0;
int end = 0;
String key = null;
// Iterate characters
do {
// If we've reached a delimiter or the end of the array, add to list
if (end == chars.length || chars[end] == 0) {
// Zero length string with no key, we're done
if (start == end && key == null) {
break;
}
// Otherwise add string (possibly empty) and reset start
// Intentionally using platform default charset
strMap.put(key, new String(chars, start, end - start));
key = null;
start = end + 1;
} else if (chars[end] == '=' && key == null) {
key = new String(chars, start, end - start);
start = end + 1;
}
} while (end++ < chars.length);
return strMap;
}
/**
* Parses a delimited String into an enum map. Multiple consecutive delimiters are treated as one.
*
* @param a type extending Enum
* @param clazz The enum class
* @param values A delimited String to be parsed into the map
* @param delim the delimiter to use
* @return An EnumMap populated in order using the delimited String values. If there are fewer String values than
* enum values, the later enum values are not mapped. The final enum value will contain the remainder of the
* String, including excess delimiters.
*/
public static > Map stringToEnumMap(Class clazz, String values, char delim) {
EnumMap map = new EnumMap<>(clazz);
int start = 0;
int len = values.length();
EnumSet keys = EnumSet.allOf(clazz);
int keySize = keys.size();
for (K key : keys) {
// If this is the last enum, put the index at the end of the string, otherwise
// put at delimiter
int idx = --keySize == 0 ? len : values.indexOf(delim, start);
if (idx >= 0) {
map.put(key, values.substring(start, idx));
start = idx;
do {
start++;
} while (start < len && values.charAt(start) == delim);
} else {
map.put(key, values.substring(start));
break;
}
}
return map;
}
}