es.weso.utils.StrUtils.scala Maven / Gradle / Ivy
The newest version!
package es.weso.utils
object StrUtils {
/** Converter takes an index i into a string and checks how many charts it can replace If it can't replace, it returns
* None (no conversion) If it replaces, it returns the characters that are replaced and the next index in the string
* Rationale: Some conversions may require some lookahead, in which case, the index will be i + characters read
*/
type Converter = (String, Int) => Option[CharConversion]
type CharConversion = (Array[Char], Int)
/** Unescape unicode numbers Given a string like: "p\u0031", return "p1" The code implements the Turtle rules:
* https://www.w3.org/TR/turtle/#sec-escapes
*
* @param str
* input string
* @return
* unscaped output string
*/
def unescapeStringLiteral(str: String): String =
cnvLoop(str, List(unescapeStringEscapeSequence, unescapeNumericSequence, unescapeReservedChar))
def unescapeIRI(str: String): String = unescapeStringLiteral(str)
def unescapeCode(str: String): String = cnvLoop(
str,
List(unescapeStringEscapeSequence, unescapeNumericSequence, unescapeReservedPatternChar, unescapeReservedChar)
)
def unescapePattern(str: String): String = cnvLoop(
str,
List(
// unescapeStringEscapeSequence,
unescapeNumericSequence,
unescapeBackSlash
// unescapeReservedPatternChar,
// unescapeReservedChar
)
)
private def cnvChar(c: Char, i: Int): Option[CharConversion] =
Some((Array(c), i))
private def unescapeBackSlash: Converter = (str, i) =>
if (str(i) == '\\' && str.length > 1) {
str(i + 1) match {
case '\\' => Some((Array('\\', '\\'), i + 1))
case _ => None
}
} else None
private def unescapeStringEscapeSequence: Converter = (str, i) =>
if (str(i) == '\\' && str.length > 1) {
str(i + 1) match {
case 't' => cnvChar('\u0009', i + 1)
case 'b' => cnvChar('\u0008', i + 1)
case 'n' => cnvChar('\u000A', i + 1)
case 'r' => cnvChar('\u000D', i + 1)
case 'f' => cnvChar('\u000C', i + 1)
case '\"' => cnvChar('\u0022', i + 1)
case '\'' => cnvChar('\'', i + 1)
case '\\' => cnvChar('\\', i + 1)
case _ => None
}
} else None
private def unescapeNumericSequence: Converter = (str, i) =>
if (str(i) == '\\' && str.length > 1) {
str(i + 1) match {
case 'u' => {
val hexValue = getHex(str, i + 2, 4)
Some((Character.toChars(hexValue), i + 5))
}
case 'U' => {
val hexValue = getHex(str, i + 2, 8)
Some((Character.toChars(hexValue), i + 9))
}
case _ => None
}
} else None
private def getHex(str: String, index: Int, num: Int): Int = {
val rs = (0 to num - 1).map(n => str(index + n)).mkString
Integer.parseInt(rs, 16)
}
private def unescapeReservedPatternChar: Converter = (str, i) =>
if (str(i) == '\\' && str.length > 1) {
str(i + 1) match {
case c if "^$[]".contains(c) => {
// println(s"unescape pattern: $c")
Some((Array('\\', c), i + 1))
}
case c => None
}
} else None
private def unescapeReservedChar: Converter = (str, i) =>
if (str(i) == '\\' && str.length > 1) {
str(i + 1) match {
case c if "~.-!$&'()*+,;=/?#@%_".contains(c) => cnvChar(c, i + 1)
case c => None
}
} else None
/** Escape a string Example: "Hi\n\t" -> "Hi\\n\\t"
* @param str
* @return
*/
def escapeStringLiteral(str: String): String = cnvLoop(str, List(cnvCtrl))
def escapePattern(str: String): String = cnvLoop(str, List())
def cnvLoop(str: String, converters: List[Converter]): String = {
var i = 0
val length = str.length
val builder = new StringBuilder(length)
while (i < str.length) {
val (nextChars, newIndex) = {
val zero: CharConversion = noConverter(str, i)
def next(f: Converter, r: CharConversion): CharConversion = {
f(str, i).getOrElse(r)
}
converters.foldRight(zero)(next)
}
i = newIndex + 1
builder.appendAll(nextChars)
}
builder.mkString
}
/* def escape(str: String): String = {
var i = 0
val length = str.length
val builder = new StringBuilder(length)
while (i < str.length) {
val (nextChars,newIndex) = cnvCtrl(str,i).getOrElse(noConverter(str,i))
i = newIndex + 1
builder.appendAll(nextChars)
}
builder.mkString
} */
private def escapeChar(c: Char, i: Int) = Some((Array('\\', c), i))
private def cnvCtrl: Converter = (str, i) =>
str(i) match {
case '\t' => escapeChar('t', i)
case '\b' => escapeChar('b', i)
case '\n' => escapeChar('n', i)
case '\r' => escapeChar('r', i)
case '\f' => escapeChar('f', i)
case '\'' => escapeChar('\'', i)
case '\"' => escapeChar('\"', i)
case _ => None
}
private def noConverter(str: String, i: Int): CharConversion =
(Array(str(i)), i)
/** escapeDot: Escapes strings to be represented as labels in Dot It follows dot conventions:
* https://graphviz.gitlab.io/_pages/doc/info/lang.html Extra characters are escaped using their Unicode
* representation
* @param str
* @return
*/
def escapeDot(str: String): String = cnvLoop(str, List(dotConverter))
private def dotConverter: Converter = (str, i) =>
str(i) match {
case c if c.isLetterOrDigit => None
case c if c.toInt > 200 && c.toInt < 377 => None
case c => {
Some((s"&#${c.toInt};".toCharArray, i))
}
}
}