azalea/kotlin-fork
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[WASM] Regex std implementation

Browse Source
This commit is contained in:
Igor Yakovlev
2021-11-26 15:23:40 +01:00
committed by TeamCityServer
parent 4f9b54da26
commit d55e16a030
78 changed files with 3403 additions and 1286 deletions
Download Patch File Download Diff File Expand all files Collapse all files
libraries/stdlib/native-wasm/src/kotlin/text/Appendable.kt
+37
View File
@@ -0,0 +1,37 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
package kotlin.text
/**
* An object to which char sequences and values can be appended.
*/
public actual interface Appendable {
/**
* Appends the specified character [value] to this Appendable and returns this instance.
*
* @param value the character to append.
*/
actual fun append(value: Char): Appendable
/**
* Appends the specified character sequence [value] to this Appendable and returns this instance.
*
* @param value the character sequence to append. If [value] is `null`, then the four characters `"null"` are appended to this Appendable.
*/
actual fun append(value: CharSequence?): Appendable
/**
* Appends a subsequence of the specified character sequence [value] to this Appendable and returns this instance.
*
* @param value the character sequence from which a subsequence is appended. If [value] is `null`,
* then characters are appended as if [value] contained the four characters `"null"`.
* @param startIndex the beginning (inclusive) of the subsequence to append.
* @param endIndex the end (exclusive) of the subsequence to append.
*
* @throws IndexOutOfBoundsException or [IllegalArgumentException] when [startIndex] or [endIndex] is out of range of the [value] character sequence indices or when `startIndex > endIndex`.
*/
actual fun append(value: CharSequence?, startIndex: Int, endIndex: Int): Appendable
}
libraries/stdlib/native-wasm/src/kotlin/text/Char.kt
+282
View File
@@ -0,0 +1,282 @@
/*
* Copyright 2010-2021 JetBrains s.r.o. and Kotlin Programming Language contributors.
* Use of this source code is governed by the Apache 2.0 license that can be found in the license/LICENSE.txt file.
*/
package kotlin.text
import kotlin.IllegalArgumentException
/**
* Returns `true` if this character (Unicode code point) is defined in Unicode.
*
* A character is considered to be defined in Unicode if its [category] is not [CharCategory.UNASSIGNED].
*/
public actual fun Char.isDefined(): Boolean {
if (this < '\u0080') {
return true
}
return getCategoryValue() != CharCategory.UNASSIGNED.value
}
/**
* Returns `true` if this character is a letter.
*
* A character is considered to be a letter if its [category] is [CharCategory.UPPERCASE_LETTER],
* [CharCategory.LOWERCASE_LETTER], [CharCategory.TITLECASE_LETTER], [CharCategory.MODIFIER_LETTER], or [CharCategory.OTHER_LETTER].
*
* @sample samples.text.Chars.isLetter
*/
public actual fun Char.isLetter(): Boolean {
if (this in 'a'..'z' || this in 'A'..'Z') {
return true
}
if (this < '\u0080') {
return false
}
return isLetterImpl()
}
/**
* Returns `true` if this character is a letter or digit.
*
* @see isLetter
* @see isDigit
*
* @sample samples.text.Chars.isLetterOrDigit
*/
public actual fun Char.isLetterOrDigit(): Boolean {
if (this in 'a'..'z' || this in 'A'..'Z' || this in '0'..'9') {
return true
}
if (this < '\u0080') {
return false
}
return isDigit() || isLetter()
}
/**
* Returns `true` if this character is a digit.
*
* A character is considered to be a digit if its [category] is [CharCategory.DECIMAL_DIGIT_NUMBER].
*
* @sample samples.text.Chars.isDigit
*/
public actual fun Char.isDigit(): Boolean {
if (this in '0'..'9') {
return true
}
if (this < '\u0080') {
return false
}
return isDigitImpl()
}
/**
* Determines whether a character is whitespace according to the Unicode standard.
* Returns `true` if the character is whitespace.
*
* @sample samples.text.Chars.isWhitespace
*/
public actual fun Char.isWhitespace(): Boolean = isWhitespaceImpl()
/**
* Returns `true` if this character is upper case.
*
* A character is considered to be an upper case character if its [category] is [CharCategory.UPPERCASE_LETTER],
* or it has contributory property `Other_Uppercase` as defined by the Unicode Standard.
*
* @sample samples.text.Chars.isUpperCase
*/
public actual fun Char.isUpperCase(): Boolean {
if (this in 'A'..'Z') {
return true
}
if (this < '\u0080') {
return false
}
return isUpperCaseImpl()
}
/**
* Returns `true` if this character is lower case.
*
* A character is considered to be a lower case character if its [category] is [CharCategory.LOWERCASE_LETTER],
* or it has contributory property `Other_Lowercase` as defined by the Unicode Standard.
*
* @sample samples.text.Chars.isLowerCase
*/
public actual fun Char.isLowerCase(): Boolean {
if (this in 'a'..'z') {
return true
}
if (this < '\u0080') {
return false
}
return isLowerCaseImpl()
}
/**
* Returns `true` if this character is a title case letter.
*
* A character is considered to be a title case letter if its [category] is [CharCategory.TITLECASE_LETTER].
*
* @sample samples.text.Chars.isTitleCase
*/
@SinceKotlin("1.5")
public actual fun Char.isTitleCase(): Boolean {
if (this < '\u0080') {
return false
}
return getCategoryValue() == CharCategory.TITLECASE_LETTER.value
}
/**
* Converts this character to upper case using Unicode mapping rules of the invariant locale.
*/
@Deprecated("Use uppercaseChar() instead.", ReplaceWith("uppercaseChar()"))
@DeprecatedSinceKotlin(warningSince = "1.5")
public actual fun Char.toUpperCase(): Char = uppercaseCharImpl()
/**
* Converts this character to upper case using Unicode mapping rules of the invariant locale.
*
* This function performs one-to-one character mapping.
* To support one-to-many character mapping use the [uppercase] function.
* If this character has no mapping equivalent, the character itself is returned.
*
* @sample samples.text.Chars.uppercase
*/
@SinceKotlin("1.5")
@WasExperimental(ExperimentalStdlibApi::class)
public actual fun Char.uppercaseChar(): Char = uppercaseCharImpl()
/**
* Converts this character to upper case using Unicode mapping rules of the invariant locale.
*
* This function supports one-to-many character mapping, thus the length of the returned string can be greater than one.
* For example, `'\uFB00'.uppercase()` returns `"\u0046\u0046"`,
* where `'\uFB00'` is the LATIN SMALL LIGATURE FF character (`ff`).
* If this character has no upper case mapping, the result of `toString()` of this char is returned.
*
* @sample samples.text.Chars.uppercase
*/
@SinceKotlin("1.5")
@WasExperimental(ExperimentalStdlibApi::class)
public actual fun Char.uppercase(): String = uppercaseImpl()
/**
* Converts this character to lower case using Unicode mapping rules of the invariant locale.
*/
@Deprecated("Use lowercaseChar() instead.", ReplaceWith("lowercaseChar()"))
@DeprecatedSinceKotlin(warningSince = "1.5")
public actual fun Char.toLowerCase(): Char = lowercaseCharImpl()
/**
* Converts this character to lower case using Unicode mapping rules of the invariant locale.
*
* This function performs one-to-one character mapping.
* To support one-to-many character mapping use the [lowercase] function.
* If this character has no mapping equivalent, the character itself is returned.
*
* @sample samples.text.Chars.lowercase
*/
@SinceKotlin("1.5")
@WasExperimental(ExperimentalStdlibApi::class)
public actual fun Char.lowercaseChar(): Char = lowercaseCharImpl()
/**
* Converts this character to lower case using Unicode mapping rules of the invariant locale.
*
* This function supports one-to-many character mapping, thus the length of the returned string can be greater than one.
* For example, `'\u0130'.lowercase()` returns `"\u0069\u0307"`,
* where `'\u0130'` is the LATIN CAPITAL LETTER I WITH DOT ABOVE character (`İ`).
* If this character has no lower case mapping, the result of `toString()` of this char is returned.
*
* @sample samples.text.Chars.lowercase
*/
@SinceKotlin("1.5")
@WasExperimental(ExperimentalStdlibApi::class)
public actual fun Char.lowercase(): String = lowercaseImpl()
/**
* Converts this character to title case using Unicode mapping rules of the invariant locale.
*
* This function performs one-to-one character mapping.
* To support one-to-many character mapping use the [titlecase] function.
* If this character has no mapping equivalent, the result of calling [uppercaseChar] is returned.
*
* @sample samples.text.Chars.titlecase
*/
@SinceKotlin("1.5")
public actual fun Char.titlecaseChar(): Char = titlecaseCharImpl()
/**
* Returns the Unicode general category of this character.
*/
public actual val Char.category: CharCategory
get() = CharCategory.valueOf(getCategoryValue())
/**
* Checks whether the given [radix] is valid radix for string to number and number to string conversion.
*/
@PublishedApi
internal actual fun checkRadix(radix: Int): Int {
if(radix !in Char.MIN_RADIX..Char.MAX_RADIX) {
throw IllegalArgumentException("radix $radix was not in valid range ${Char.MIN_RADIX..Char.MAX_RADIX}")
}
return radix
}
// Char.Compaion methods. Konan specific.
// TODO: Make public when supplementary codepoints are supported.
/** Converts a unicode code point to lower case. */
internal fun Char.Companion.toLowerCase(codePoint: Int): Int =
if (codePoint < MIN_SUPPLEMENTARY_CODE_POINT) {
@Suppress("DEPRECATION")
codePoint.toChar().lowercaseChar().toInt()
} else {
codePoint // TODO: Implement this transformation for supplementary codepoints.
}
/** Converts a unicode code point to upper case. */
internal fun Char.Companion.toUpperCase(codePoint: Int): Int =
if (codePoint < MIN_SUPPLEMENTARY_CODE_POINT) {
@Suppress("DEPRECATION")
codePoint.toChar().uppercaseChar().toInt()
} else {
codePoint // TODO: Implement this transformation for supplementary codepoints.
}
/** Converts a surrogate pair to a unicode code point. Doesn't validate that the characters are a valid surrogate pair. */
// TODO: Consider removing from public API
public fun Char.Companion.toCodePoint(high: Char, low: Char): Int =
(((high - MIN_HIGH_SURROGATE) shl 10) or (low - MIN_LOW_SURROGATE)) + 0x10000
/** Checks if the codepoint specified is a supplementary codepoint or not. */
// TODO: Consider removing from public API
public fun Char.Companion.isSupplementaryCodePoint(codepoint: Int): Boolean =
codepoint in MIN_SUPPLEMENTARY_CODE_POINT..MAX_CODE_POINT
// TODO: Consider removing from public API
public fun Char.Companion.isSurrogatePair(high: Char, low: Char): Boolean = high.isHighSurrogate() && low.isLowSurrogate()
/**
* Converts the codepoint specified to a char array. If the codepoint is not supplementary, the method will
* return an array with one element otherwise it will return an array A with a high surrogate in A[0] and
* a low surrogate in A[1].
*/
// TODO: Consider removing from public API
@Suppress("DEPRECATION")
public fun Char.Companion.toChars(codePoint: Int): CharArray =
when {
codePoint in 0 until MIN_SUPPLEMENTARY_CODE_POINT -> charArrayOf(codePoint.toChar())
codePoint in MIN_SUPPLEMENTARY_CODE_POINT..MAX_CODE_POINT -> {
val low = ((codePoint - 0x10000) and 0x3FF) + MIN_LOW_SURROGATE.toInt()
val high = (((codePoint - 0x10000) ushr 10) and 0x3FF) + MIN_HIGH_SURROGATE.toInt()
charArrayOf(high.toChar(), low.toChar())
}
else -> throw IllegalArgumentException()
}
libraries/stdlib/native-wasm/src/kotlin/text/CharCategory.kt
+174
View File
@@ -0,0 +1,174 @@
/*
* Copyright 2010-2020 JetBrains s.r.o. and Kotlin Programming Language contributors.
* Use of this source code is governed by the Apache 2.0 license that can be found in the license/LICENSE.txt file.
*/
package kotlin.text
/**
* Represents the character general category in the Unicode specification.
*/
public actual enum class CharCategory(public val value: Int, public actual val code: String) {
/**
* General category "Cn" in the Unicode specification.
*/
UNASSIGNED(0, "Cn"),
/**
* General category "Lu" in the Unicode specification.
*/
UPPERCASE_LETTER(1, "Lu"),
/**
* General category "Ll" in the Unicode specification.
*/
LOWERCASE_LETTER(2, "Ll"),
/**
* General category "Lt" in the Unicode specification.
*/
TITLECASE_LETTER(3, "Lt"),
/**
* General category "Lm" in the Unicode specification.
*/
MODIFIER_LETTER(4, "Lm"),
/**
* General category "Lo" in the Unicode specification.
*/
OTHER_LETTER(5, "Lo"),
/**
* General category "Mn" in the Unicode specification.
*/
NON_SPACING_MARK(6, "Mn"),
/**
* General category "Me" in the Unicode specification.
*/
ENCLOSING_MARK(7, "Me"),
/**
* General category "Mc" in the Unicode specification.
*/
COMBINING_SPACING_MARK(8, "Mc"),
/**
* General category "Nd" in the Unicode specification.
*/
DECIMAL_DIGIT_NUMBER(9, "Nd"),
/**
* General category "Nl" in the Unicode specification.
*/
LETTER_NUMBER(10, "Nl"),
/**
* General category "No" in the Unicode specification.
*/
OTHER_NUMBER(11, "No"),
/**
* General category "Zs" in the Unicode specification.
*/
SPACE_SEPARATOR(12, "Zs"),
/**
* General category "Zl" in the Unicode specification.
*/
LINE_SEPARATOR(13, "Zl"),
/**
* General category "Zp" in the Unicode specification.
*/
PARAGRAPH_SEPARATOR(14, "Zp"),
/**
* General category "Cc" in the Unicode specification.
*/
CONTROL(15, "Cc"),
/**
* General category "Cf" in the Unicode specification.
*/
FORMAT(16, "Cf"),
/**
* General category "Co" in the Unicode specification.
*/
PRIVATE_USE(18, "Co"),
/**
* General category "Cs" in the Unicode specification.
*/
SURROGATE(19, "Cs"),
/**
* General category "Pd" in the Unicode specification.
*/
DASH_PUNCTUATION(20, "Pd"),
/**
* General category "Ps" in the Unicode specification.
*/
START_PUNCTUATION(21, "Ps"),
/**
* General category "Pe" in the Unicode specification.
*/
END_PUNCTUATION(22, "Pe"),
/**
* General category "Pc" in the Unicode specification.
*/
CONNECTOR_PUNCTUATION(23, "Pc"),
/**
* General category "Po" in the Unicode specification.
*/
OTHER_PUNCTUATION(24, "Po"),
/**
* General category "Sm" in the Unicode specification.
*/
MATH_SYMBOL(25, "Sm"),
/**
* General category "Sc" in the Unicode specification.
*/
CURRENCY_SYMBOL(26, "Sc"),
/**
* General category "Sk" in the Unicode specification.
*/
MODIFIER_SYMBOL(27, "Sk"),
/**
* General category "So" in the Unicode specification.
*/
OTHER_SYMBOL(28, "So"),
/**
* General category "Pi" in the Unicode specification.
*/
INITIAL_QUOTE_PUNCTUATION(29, "Pi"),
/**
* General category "Pf" in the Unicode specification.
*/
FINAL_QUOTE_PUNCTUATION(30, "Pf");
/**
* Returns `true` if [char] character belongs to this category.
*/
public actual operator fun contains(char: Char): Boolean = char.getCategoryValue() == this.value
public companion object {
public fun valueOf(category: Int): CharCategory =
when (category) {
in 0..16 -> values()[category]
in 18..30 -> values()[category - 1]
else -> throw IllegalArgumentException("Category #$category is not defined.")
}
}
}
libraries/stdlib/native-wasm/src/kotlin/text/PatternSyntaxException.kt
+32
View File
@@ -0,0 +1,32 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
package kotlin.text
/**
* Encapsulates a syntax error that occurred during the compilation of a
* [Pattern]. Might include a detailed description, the original regular
* expression, and the index at which the error occurred.
*/
internal class PatternSyntaxException(
val description: String = "",
val pattern: String = "",
val index: Int = -1
) : IllegalArgumentException(formatMessage(description, pattern, index)) {
companion object {
fun formatMessage(description: String, pattern: String, index: Int): String {
if (index < 0 || pattern == "") {
return description
}
val filler = if (index >= 1) " ".repeat(index) else ""
return """
$description near index: $index
$pattern
$filler^
""".trimIndent()
}
}
}
libraries/stdlib/native-wasm/src/kotlin/text/Regex.kt
+428
View File
@@ -0,0 +1,428 @@
/*
* Copyright 2010-2021 JetBrains s.r.o. and Kotlin Programming Language contributors.
* Use of this source code is governed by the Apache 2.0 license that can be found in the license/LICENSE.txt file.
*/
package kotlin.text
import kotlin.text.regex.*
@PublishedApi
internal interface FlagEnum {
val value: Int
val mask: Int
}
private fun Iterable<FlagEnum>.toInt(): Int = this.fold(0, { value, option -> value or option.value })
private fun fromInt(value: Int): Set<RegexOption> =
RegexOption.values().filterTo(mutableSetOf<RegexOption>()) { value and it.mask == it.value }
/**
* Provides enumeration values to use to set regular expression options.
*/
public actual enum class RegexOption(override val value: Int, override val mask: Int = value) : FlagEnum {
// common
/** Enables case-insensitive matching. Case comparison is Unicode-aware. */
IGNORE_CASE(Pattern.CASE_INSENSITIVE),
/**
* Enables multiline mode.
*
* In multiline mode the expressions `^` and `$` match just after or just before,
* respectively, a line terminator or the end of the input sequence.
*/
MULTILINE(Pattern.MULTILINE),
/**
* Enables literal parsing of the pattern.
*
* Metacharacters or escape sequences in the input sequence will be given no special meaning.
*/
LITERAL(Pattern.LITERAL),
/** Enables Unix lines mode. In this mode, only the `'\n'` is recognized as a line terminator. */
UNIX_LINES(Pattern.UNIX_LINES),
/** Permits whitespace and comments in pattern. */
COMMENTS(Pattern.COMMENTS),
/** Enables the mode, when the expression `.` matches any character, including a line terminator. */
DOT_MATCHES_ALL(Pattern.DOTALL),
/** Enables equivalence by canonical decomposition. */
CANON_EQ(Pattern.CANON_EQ)
}
/**
* Represents the results from a single capturing group within a [MatchResult] of [Regex].
*
* @param value The value of captured group.
* @param range The range of indices in the input string where group was captured.
*
* The [range] property is available on JVM only.
*/
public actual data class MatchGroup(actual val value: String, val range: IntRange)
/**
* Represents a compiled regular expression.
* Provides functions to match strings in text with a pattern, replace the found occurrences and split text around matches.
*/
public actual class Regex internal constructor(internal val nativePattern: Pattern) {
internal enum class Mode {
FIND, MATCH
}
/** Creates a regular expression from the specified [pattern] string and the default options. */
actual constructor(pattern: String): this(Pattern(pattern))
/** Creates a regular expression from the specified [pattern] string and the specified single [option]. */
actual constructor(pattern: String, option: RegexOption): this(Pattern(pattern, ensureUnicodeCase(option.value)))
/** Creates a regular expression from the specified [pattern] string and the specified set of [options]. */
actual constructor(pattern: String, options: Set<RegexOption>): this(Pattern(pattern, ensureUnicodeCase(options.toInt())))
/** The pattern string of this regular expression. */
actual val pattern: String
get() = nativePattern.pattern
private val startNode = nativePattern.startNode
/** The set of options that were used to create this regular expression. */
actual val options: Set<RegexOption> = fromInt(nativePattern.flags)
actual companion object {
/**
* Returns a regular expression that matches the specified [literal] string literally.
* No characters of that string will have special meaning when searching for an occurrence of the regular expression.
*/
actual fun fromLiteral(literal: String): Regex = Regex(literal, RegexOption.LITERAL)
/**
* Returns a regular expression pattern string that matches the specified [literal] string literally.
* No characters of that string will have special meaning when searching for an occurrence of the regular expression.
*/
actual fun escape(literal: String): String = Pattern.quote(literal)
/**
* Returns a literal replacement expression for the specified [literal] string.
* No characters of that string will have special meaning when it is used as a replacement string in [Regex.replace] function.
*/
actual fun escapeReplacement(literal: String): String {
if (!literal.contains('\\') && !literal.contains('$'))
return literal
val result = StringBuilder(literal.length * 2)
literal.forEach {
if (it == '\\' || it == '$') {
result.append('\\')
}
result.append(it)
}
return result.toString()
}
// TODO: Remove
private fun ensureUnicodeCase(flags: Int) = flags
}
private fun doMatch(input: CharSequence, mode: Mode): MatchResult? {
// TODO: Harmony has a default constructor for MatchResult. Do we need it?
// TODO: Reuse the matchResult.
val matchResult = MatchResultImpl(input, this)
matchResult.mode = mode
val matches = startNode.matches(0, input, matchResult) >= 0
if (!matches) {
return null
}
matchResult.finalizeMatch()
return matchResult
}
/** Indicates whether the regular expression matches the entire [input]. */
actual infix fun matches(input: CharSequence): Boolean = doMatch(input, Mode.MATCH) != null
/** Indicates whether the regular expression can find at least one match in the specified [input]. */
actual fun containsMatchIn(input: CharSequence): Boolean = find(input) != null
@SinceKotlin("1.5")
@ExperimentalStdlibApi
public actual fun matchesAt(input: CharSequence, index: Int): Boolean =
// TODO: expand and simplify
matchAt(input, index) != null
/**
* Returns the first match of a regular expression in the [input], beginning at the specified [startIndex].
*
* @param startIndex An index to start search with, by default 0. Must be not less than zero and not greater than `input.length()`
* @return An instance of [MatchResult] if match was found or `null` otherwise.
* @throws IndexOutOfBoundsException if [startIndex] is less than zero or greater than the length of the [input] char sequence.
* @sample samples.text.Regexps.find
*/
@Suppress("ACTUAL_FUNCTION_WITH_DEFAULT_ARGUMENTS")
actual fun find(input: CharSequence, startIndex: Int = 0): MatchResult? {
if (startIndex < 0 || startIndex > input.length) {
throw IndexOutOfBoundsException("Start index is out of bounds: $startIndex, input length: ${input.length}")
}
val matchResult = MatchResultImpl(input, this)
matchResult.mode = Mode.FIND
matchResult.startIndex = startIndex
val foundIndex = startNode.find(startIndex, input, matchResult)
if (foundIndex >= 0) {
matchResult.finalizeMatch()
return matchResult
} else {
return null
}
}
/**
* Returns a sequence of all occurrences of a regular expression within the [input] string, beginning at the specified [startIndex].
*
* @throws IndexOutOfBoundsException if [startIndex] is less than zero or greater than the length of the [input] char sequence.
*
* @sample samples.text.Regexps.findAll
*/
@Suppress("ACTUAL_FUNCTION_WITH_DEFAULT_ARGUMENTS")
actual fun findAll(input: CharSequence, startIndex: Int = 0): Sequence<MatchResult> {
if (startIndex < 0 || startIndex > input.length) {
throw IndexOutOfBoundsException("Start index is out of bounds: $startIndex, input length: ${input.length}")
}
return generateSequence({ find(input, startIndex) }, MatchResult::next)
}
/**
* Attempts to match the entire [input] CharSequence against the pattern.
*
* @return An instance of [MatchResult] if the entire input matches or `null` otherwise.
*/
actual fun matchEntire(input: CharSequence): MatchResult?= doMatch(input, Mode.MATCH)
@SinceKotlin("1.5")
@ExperimentalStdlibApi
public actual fun matchAt(input: CharSequence, index: Int): MatchResult? {
if (index < 0 || index > input.length) {
throw IndexOutOfBoundsException("index is out of bounds: $index, input length: ${input.length}")
}
val matchResult = MatchResultImpl(input, this)
matchResult.mode = Mode.FIND
matchResult.startIndex = index
val matches = startNode.matches(index, input, matchResult) >= 0
if (!matches) {
return null
}
matchResult.finalizeMatch()
return matchResult
}
/**
* Replaces all occurrences of this regular expression in the specified [input] string with specified [replacement] expression.
*
* The replacement string may contain references to the captured groups during a match. Occurrences of `$index`
* in the replacement string will be substituted with the subsequences corresponding to the captured groups with the specified index.
* The first digit after '$' is always treated as part of group reference. Subsequent digits are incorporated
* into `index` only if they would form a valid group reference. Only the digits '0'..'9' are considered as potential components
* of the group reference. Note that indexes of captured groups start from 1, and the group with index 0 is the whole match.
*
* Backslash character '\' can be used to include the succeeding character as a literal in the replacement string, e.g, `\$` or `\\`.
* [Regex.escapeReplacement] can be used if [replacement] have to be treated as a literal string.
*
* Note that named capturing groups are not supported in Kotlin/Native.
*
* @param input the char sequence to find matches of this regular expression in
* @param replacement the expression to replace found matches with
* @return the result of replacing each occurrence of this regular expression in [input] with the result of evaluating the [replacement] expression
* @throws RuntimeException if [replacement] expression is malformed, or capturing group with specified `name` or `index` does not exist
*/
actual fun replace(input: CharSequence, replacement: String): String
= replace(input) { match -> substituteGroupRefs(match, replacement) }
/**
* Replaces all occurrences of this regular expression in the specified [input] string with the result of
* the given function [transform] that takes [MatchResult] and returns a string to be used as a
* replacement for that match.
*/
actual fun replace(input: CharSequence, transform: (MatchResult) -> CharSequence): String {
var match: MatchResult? = find(input) ?: return input.toString()
var lastStart = 0
val length = input.length
val sb = StringBuilder(length)
do {
val foundMatch = match!!
sb.append(input, lastStart, foundMatch.range.start)
sb.append(transform(foundMatch))
lastStart = foundMatch.range.endInclusive + 1
match = foundMatch.next()
} while (lastStart < length && match != null)
if (lastStart < length) {
sb.append(input, lastStart, length)
}
return sb.toString()
}
/**
* Replaces the first occurrence of this regular expression in the specified [input] string with specified [replacement] expression.
*
* The replacement string may contain references to the captured groups during a match. Occurrences of `$index`
* in the replacement string will be substituted with the subsequences corresponding to the captured groups with the specified index.
* The first digit after '$' is always treated as part of group reference. Subsequent digits are incorporated
* into `index` only if they would form a valid group reference. Only the digits '0'..'9' are considered as potential components
* of the group reference. Note that indexes of captured groups start from 1, and the group with index 0 is the whole match.
*
* Backslash character '\' can be used to include the succeeding character as a literal in the replacement string, e.g, `\$` or `\\`.
* [Regex.escapeReplacement] can be used if [replacement] have to be treated as a literal string.
*
* Note that named capturing groups are not supported in Kotlin/Native.
*
* @param input the char sequence to find a match of this regular expression in
* @param replacement the expression to replace the found match with
* @return the result of replacing the first occurrence of this regular expression in [input] with the result of evaluating the [replacement] expression
* @throws RuntimeException if [replacement] expression is malformed, or capturing group with specified `name` or `index` does not exist
*/
actual fun replaceFirst(input: CharSequence, replacement: String): String {
val match = find(input) ?: return input.toString()
val length = input.length
val result = StringBuilder(length)
result.append(input, 0, match.range.start)
result.append(substituteGroupRefs(match, replacement))
if (match.range.endInclusive + 1 < length) {
result.append(input, match.range.endInclusive + 1, length)
}
return result.toString()
}
/**
* Splits the [input] CharSequence to a list of strings around matches of this regular expression.
*
* @param limit Non-negative value specifying the maximum number of substrings the string can be split to.
* Zero by default means no limit is set.
*/
@Suppress("ACTUAL_FUNCTION_WITH_DEFAULT_ARGUMENTS")
actual fun split(input: CharSequence, limit: Int = 0): List<String> {
requireNonNegativeLimit(limit)
var match: MatchResult? = find(input)
if (match == null || limit == 1) return listOf(input.toString())
val result = ArrayList<String>(if (limit > 0) limit.coerceAtMost(10) else 10)
var lastStart = 0
val lastSplit = limit - 1 // negative if there's no limit
do {
result.add(input.substring(lastStart, match!!.range.start))
lastStart = match.range.endInclusive + 1
if (lastSplit >= 0 && result.size == lastSplit) break
match = match.next()
} while (match != null)
result.add(input.substring(lastStart, input.length))
return result
}
/**
* Splits the [input] CharSequence to a sequence of strings around matches of this regular expression.
*
* @param limit Non-negative value specifying the maximum number of substrings the string can be split to.
* Zero by default means no limit is set.
* @sample samples.text.Regexps.splitToSequence
*/
@SinceKotlin("1.6")
@WasExperimental(ExperimentalStdlibApi::class)
@Suppress("ACTUAL_FUNCTION_WITH_DEFAULT_ARGUMENTS")
public actual fun splitToSequence(input: CharSequence, limit: Int = 0): Sequence<String> {
requireNonNegativeLimit(limit)
return sequence {
var match = find(input)
if (match == null || limit == 1) {
yield(input.toString())
return@sequence
}
var nextStart = 0
var splitCount = 0
do {
val foundMatch = match!!
yield(input.substring(nextStart, foundMatch.range.first))
nextStart = foundMatch.range.endInclusive + 1
match = foundMatch.next()
} while (++splitCount != limit - 1 && match != null)
yield(input.substring(nextStart, input.length))
}
}
/**
* Returns the string representation of this regular expression, namely the [pattern] of this regular expression.
*
* Note that another regular expression constructed from the same pattern string may have different [options]
* and may match strings differently.
*/
override fun toString(): String = nativePattern.toString()
}
// The same code from K/JS regex.kt
private fun substituteGroupRefs(match: MatchResult, replacement: String): String {
var index = 0
val result = StringBuilder(replacement.length)
while (index < replacement.length) {
val char = replacement[index++]
if (char == '\\') {
if (index == replacement.length)
throw IllegalArgumentException("The Char to be escaped is missing")
result.append(replacement[index++])
} else if (char == '$') {
if (index == replacement.length)
throw IllegalArgumentException("Capturing group index is missing")
if (replacement[index] == '{')
throw IllegalArgumentException("Named capturing group reference currently is not supported")
if (replacement[index] !in '0'..'9')
throw IllegalArgumentException("Invalid capturing group reference")
val endIndex = replacement.readGroupIndex(index, match.groupValues.size)
val groupIndex = replacement.substring(index, endIndex).toInt()
if (groupIndex >= match.groupValues.size)
throw IndexOutOfBoundsException("Group with index $groupIndex does not exist")
result.append(match.groupValues[groupIndex])
index = endIndex
} else {
result.append(char)
}
}
return result.toString()
}
private fun String.readGroupIndex(startIndex: Int, groupCount: Int): Int {
// at least one digit after '$' is always captured
var index = startIndex + 1
var groupIndex = this[startIndex] - '0'
// capture the largest valid group index
while (index < length && this[index] in '0'..'9') {
val newGroupIndex = (groupIndex * 10) + (this[index] - '0')
if (newGroupIndex in 0 until groupCount) {
groupIndex = newGroupIndex
index++
} else {
break
}
}
return index
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/AbstractCharClass.kt
+586
View File
@@ -0,0 +1,586 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
@file:Suppress("DEPRECATION") // Char.toInt()
package kotlin.text.regex
import kotlin.collections.associate
import kotlin.native.concurrent.AtomicReference
import kotlin.native.concurrent.freeze
import kotlin.native.BitSet
/**
* Unicode category (i.e. Ll, Lu).
*/
internal open class UnicodeCategory(protected val category: Int) : AbstractCharClass() {
override fun contains(ch: Int): Boolean = alt xor (ch.toChar().category.value == category)
}
/**
* Unicode category scope (i.e IsL, IsM, ...)
*/
internal class UnicodeCategoryScope(category: Int) : UnicodeCategory(category) {
override fun contains(ch: Int): Boolean {
return alt xor (((category shr ch.toChar().category.value) and 1) != 0)
}
}
/**
* This class represents character classes, i.e. sets of character either predefined or user defined.
* Note: this class represent a token, not node, so being constructed by lexer.
*/
internal abstract class AbstractCharClass : SpecialToken() {
/**
* Show if the class has alternative meaning:
* if the class contains character 'a' and alt == true then the class will contains all characters except 'a'.
*/
internal var alt: Boolean = false
internal var altSurrogates: Boolean = false
internal val lowHighSurrogates = BitSet(SURROGATE_CARDINALITY) // Bit set for surrogates?
/*
* Indicates if this class may contain supplementary Unicode codepoints.
* If this flag is specified it doesn't mean that this class contains supplementary characters but may contain.
*/
var mayContainSupplCodepoints = false
protected set
/** Returns true if this char class contains character specified. */
abstract operator fun contains(ch: Int): Boolean
open fun contains(ch: Char): Boolean = contains(ch.toInt())
/**
* Returns BitSet representing this character class or `null`
* if this character class does not have character representation;
*/
open internal val bits: BitSet?
get() = null
fun hasLowHighSurrogates(): Boolean {
return if (altSurrogates)
lowHighSurrogates.nextClearBit(0) != -1
else
lowHighSurrogates.nextSetBit(0) != -1
}
override val type: Type = Type.CHARCLASS
open val instance: AbstractCharClass
get() = this
private val surrogates_ = AtomicReference<AbstractCharClass?>(null)
fun classWithSurrogates(): AbstractCharClass {
surrogates_.value?.let {
return it
}
val surrogates = lowHighSurrogates
val result = object : AbstractCharClass() {
override fun contains(ch: Int): Boolean {
val index = ch - Char.MIN_SURROGATE.toInt()
return if (index >= 0 && index < AbstractCharClass.SURROGATE_CARDINALITY) {
this.altSurrogates xor surrogates[index]
} else {
false
}
}
}
result.setNegative(this.altSurrogates)
surrogates_.compareAndSet(null, result.freeze())
return surrogates_.value!!
}
// We cannot cache this class as we've done with surrogates above because
// here is a circular reference between it and AbstractCharClass.
fun classWithoutSurrogates(): AbstractCharClass {
val result = object : AbstractCharClass() {
override fun contains(ch: Int): Boolean {
val index = ch - Char.MIN_SURROGATE.toInt()
val containslHS = if (index >= 0 && index < AbstractCharClass.SURROGATE_CARDINALITY)
this.altSurrogates xor this@AbstractCharClass.lowHighSurrogates.get(index)
else
false
return this@AbstractCharClass.contains(ch) && !containslHS
}
}
result.setNegative(isNegative())
result.mayContainSupplCodepoints = mayContainSupplCodepoints
return result
}
/**
* Sets this CharClass to negative form, i.e. if they will add some characters and after that set this
* class to negative it will accept all the characters except previously set ones.
*
* Although this method will not alternate all the already set characters,
* just overall meaning of the class.
*/
fun setNegative(value: Boolean): AbstractCharClass {
if (alt xor value) {
alt = !alt
altSurrogates = !altSurrogates
}
if (!mayContainSupplCodepoints) {
mayContainSupplCodepoints = true
}
return this
}
fun isNegative(): Boolean {
return alt
}
internal abstract class CachedCharClass {
lateinit private var posValue: AbstractCharClass
lateinit private var negValue: AbstractCharClass
// Somewhat ugly init sequence, as computeValue() may depend on fields, initialized in subclass ctor.
protected fun initValues() {
posValue = computeValue()
negValue = computeValue().setNegative(true)
}
fun getValue(negative: Boolean): AbstractCharClass = if (!negative) posValue else negValue
protected abstract fun computeValue(): AbstractCharClass
}
internal class CachedDigit : CachedCharClass() {
init {
initValues()
}
override fun computeValue(): AbstractCharClass = CharClass().add('0', '9')
}
internal class CachedNonDigit : CachedCharClass() {
init {
initValues()
}
override fun computeValue(): AbstractCharClass =
CharClass().add('0', '9').setNegative(true).apply { mayContainSupplCodepoints = true }
}
internal class CachedSpace : CachedCharClass() {
init {
initValues()
}
/* 9-13 - \t\n\x0B\f\r; 32 - ' ' */
override fun computeValue(): AbstractCharClass = CharClass().add(9, 13).add(32)
}
internal class CachedNonSpace : CachedCharClass() {
init {
initValues()
}
override fun computeValue(): AbstractCharClass =
CachedSpace().getValue(negative = true).apply { mayContainSupplCodepoints = true }
}
internal class CachedWord : CachedCharClass() {
init {
initValues()
}
override fun computeValue(): AbstractCharClass = CharClass().add('a', 'z').add('A', 'Z').add('0', '9').add('_')
}
internal class CachedNonWord : CachedCharClass() {
init {
initValues()
}
override fun computeValue(): AbstractCharClass =
CachedWord().getValue(negative = true).apply { mayContainSupplCodepoints = true }
}
internal class CachedLower : CachedCharClass() {
init {
initValues()
}
override fun computeValue(): AbstractCharClass = CharClass().add('a', 'z')
}
internal class CachedUpper : CachedCharClass() {
init {
initValues()
}
override fun computeValue(): AbstractCharClass = CharClass().add('A', 'Z')
}
internal class CachedASCII : CachedCharClass() {
init {
initValues()
}
override fun computeValue(): AbstractCharClass = CharClass().add(0x00, 0x7F)
}
internal class CachedAlpha : CachedCharClass() {
init {
initValues()
}
override fun computeValue(): AbstractCharClass = CharClass().add('a', 'z').add('A', 'Z')
}
internal class CachedAlnum : CachedCharClass() {
init {
initValues()
}
override fun computeValue(): AbstractCharClass =
(CachedAlpha().getValue(negative = false) as CharClass).add('0', '9')
}
internal class CachedPunct : CachedCharClass() {
init {
initValues()
}
/* Punctuation !"#$%&'()*+,-./:;<=>?@ [\]^_` {|}~ */
override fun computeValue(): AbstractCharClass = CharClass().add(0x21, 0x40).add(0x5B, 0x60).add(0x7B, 0x7E)
}
internal class CachedGraph : CachedCharClass() {
init {
initValues()
}
/* plus punctuation */
override fun computeValue(): AbstractCharClass =
(CachedAlnum().getValue(negative = false) as CharClass)
.add(0x21, 0x40)
.add(0x5B, 0x60)
.add(0x7B, 0x7E)
}
internal class CachedPrint : CachedCharClass() {
init {
initValues()
}
override fun computeValue(): AbstractCharClass =
(CachedGraph().getValue(negative = true) as CharClass).add(0x20)
}
internal class CachedBlank : CachedCharClass() {
init {
initValues()
}
override fun computeValue(): AbstractCharClass = CharClass().add(' ').add('\t')
}
internal class CachedCntrl : CachedCharClass() {
init {
initValues()
}
override fun computeValue(): AbstractCharClass = CharClass().add(0x00, 0x1F).add(0x7F)
}
internal class CachedXDigit : CachedCharClass() {
init {
initValues()
}
override fun computeValue(): AbstractCharClass = CharClass().add('0', '9').add('a', 'f').add('A', 'F')
}
internal class CachedRange(var start: Int, var end: Int) : CachedCharClass() {
init {
initValues()
}
override fun computeValue(): AbstractCharClass =
object: AbstractCharClass() {
override fun contains(ch: Int): Boolean = alt xor (ch in start..end)
}.apply {
if (end >= Char.MIN_SUPPLEMENTARY_CODE_POINT) {
mayContainSupplCodepoints = true
}
val minSurrogate = Char.MIN_SURROGATE.toInt()
val maxSurrogate = Char.MAX_SURROGATE.toInt()
// There is an intersection with surrogate characters.
if (end >= minSurrogate && start <= maxSurrogate && start <= end) {
val surrogatesStart = maxOf(start, minSurrogate) - minSurrogate
val surrogatesEnd = minOf(end, maxSurrogate) - minSurrogate
lowHighSurrogates.set(surrogatesStart..surrogatesEnd)
}
}
}
internal class CachedSpecialsBlock : CachedCharClass() {
init {
initValues()
}
public override fun computeValue(): AbstractCharClass = CharClass().add(0xFEFF, 0xFEFF).add(0xFFF0, 0xFFFD)
}
internal class CachedCategoryScope(
val category: Int,
val mayContainSupplCodepoints: Boolean,
val containsAllSurrogates: Boolean = false) : CachedCharClass() {
init {
initValues()
}
override fun computeValue(): AbstractCharClass {
val result = UnicodeCategoryScope(category)
if (containsAllSurrogates) {
result.lowHighSurrogates.set(0, SURROGATE_CARDINALITY)
}
result.mayContainSupplCodepoints = mayContainSupplCodepoints
return result
}
}
internal class CachedCategory(
val category: Int,
val mayContainSupplCodepoints: Boolean,
val containsAllSurrogates: Boolean = false) : CachedCharClass() {
init {
initValues()
}
override fun computeValue(): AbstractCharClass {
val result = UnicodeCategory(category)
if (containsAllSurrogates) {
result.lowHighSurrogates.set(0, SURROGATE_CARDINALITY)
}
result.mayContainSupplCodepoints = mayContainSupplCodepoints
return result
}
}
companion object {
//Char.MAX_SURROGATE - Char.MIN_SURROGATE + 1
const val SURROGATE_CARDINALITY = 2048
/**
* Character classes.
* See http://www.unicode.org/reports/tr18/, http://www.unicode.org/Public/4.1.0/ucd/Blocks.txt
*/
enum class CharClasses(val regexName : String, val factory: () -> CachedCharClass) {
LOWER("Lower", ::CachedLower),
UPPER("Upper", ::CachedUpper),
ASCII("ASCII", ::CachedASCII),
ALPHA("Alpha", ::CachedAlpha),
DIGIT("Digit", ::CachedDigit),
ALNUM("Alnum", :: CachedAlnum),
PUNCT("Punct", ::CachedPunct),
GRAPH("Graph", ::CachedGraph),
PRINT("Print", ::CachedPrint),
BLANK("Blank", ::CachedBlank),
CNTRL("Cntrl", ::CachedCntrl),
XDIGIT("XDigit", ::CachedXDigit),
SPACE("Space", ::CachedSpace),
WORD("w", ::CachedWord),
NON_WORD("W", ::CachedNonWord),
SPACE_SHORT("s", ::CachedSpace),
NON_SPACE("S", ::CachedNonSpace),
DIGIT_SHORT("d", ::CachedDigit),
NON_DIGIT("D", ::CachedNonDigit),
BASIC_LATIN("BasicLatin", { CachedRange(0x0000, 0x007F) }),
LATIN1_SUPPLEMENT("Latin-1Supplement", { CachedRange(0x0080, 0x00FF) }),
LATIN_EXTENDED_A("LatinExtended-A", { CachedRange(0x0100, 0x017F) }),
LATIN_EXTENDED_B("LatinExtended-B", { CachedRange(0x0180, 0x024F) }),
IPA_EXTENSIONS("IPAExtensions", { CachedRange(0x0250, 0x02AF) }),
SPACING_MODIFIER_LETTERS("SpacingModifierLetters", { CachedRange(0x02B0, 0x02FF) }),
COMBINING_DIACRITICAL_MARKS("CombiningDiacriticalMarks", { CachedRange(0x0300, 0x036F) }),
GREEK("Greek", { CachedRange(0x0370, 0x03FF) }),
CYRILLIC("Cyrillic", { CachedRange(0x0400, 0x04FF) }),
CYRILLIC_SUPPLEMENT("CyrillicSupplement", { CachedRange(0x0500, 0x052F) }),
ARMENIAN("Armenian", { CachedRange(0x0530, 0x058F) }),
HEBREW("Hebrew", { CachedRange(0x0590, 0x05FF) }),
ARABIC("Arabic", { CachedRange(0x0600, 0x06FF) }),
SYRIAC("Syriac", { CachedRange(0x0700, 0x074F) }),
ARABICSUPPLEMENT("ArabicSupplement", { CachedRange(0x0750, 0x077F) }),
THAANA("Thaana", { CachedRange(0x0780, 0x07BF) }),
DEVANAGARI("Devanagari", { CachedRange(0x0900, 0x097F) }),
BENGALI("Bengali", { CachedRange(0x0980, 0x09FF) }),
GURMUKHI("Gurmukhi", { CachedRange(0x0A00, 0x0A7F) }),
GUJARATI("Gujarati", { CachedRange(0x0A80, 0x0AFF) }),
ORIYA("Oriya", { CachedRange(0x0B00, 0x0B7F) }),
TAMIL("Tamil", { CachedRange(0x0B80, 0x0BFF) }),
TELUGU("Telugu", { CachedRange(0x0C00, 0x0C7F) }),
KANNADA("Kannada", { CachedRange(0x0C80, 0x0CFF) }),
MALAYALAM("Malayalam", { CachedRange(0x0D00, 0x0D7F) }),
SINHALA("Sinhala", { CachedRange(0x0D80, 0x0DFF) }),
THAI("Thai", { CachedRange(0x0E00, 0x0E7F) }),
LAO("Lao", { CachedRange(0x0E80, 0x0EFF) }),
TIBETAN("Tibetan", { CachedRange(0x0F00, 0x0FFF) }),
MYANMAR("Myanmar", { CachedRange(0x1000, 0x109F) }),
GEORGIAN("Georgian", { CachedRange(0x10A0, 0x10FF) }),
HANGULJAMO("HangulJamo", { CachedRange(0x1100, 0x11FF) }),
ETHIOPIC("Ethiopic", { CachedRange(0x1200, 0x137F) }),
ETHIOPICSUPPLEMENT("EthiopicSupplement", { CachedRange(0x1380, 0x139F) }),
CHEROKEE("Cherokee", { CachedRange(0x13A0, 0x13FF) }),
UNIFIEDCANADIANABORIGINALSYLLABICS("UnifiedCanadianAboriginalSyllabics", { CachedRange(0x1400, 0x167F) }),
OGHAM("Ogham", { CachedRange(0x1680, 0x169F) }),
RUNIC("Runic", { CachedRange(0x16A0, 0x16FF) }),
TAGALOG("Tagalog", { CachedRange(0x1700, 0x171F) }),
HANUNOO("Hanunoo", { CachedRange(0x1720, 0x173F) }),
BUHID("Buhid", { CachedRange(0x1740, 0x175F) }),
TAGBANWA("Tagbanwa", { CachedRange(0x1760, 0x177F) }),
KHMER("Khmer", { CachedRange(0x1780, 0x17FF) }),
MONGOLIAN("Mongolian", { CachedRange(0x1800, 0x18AF) }),
LIMBU("Limbu", { CachedRange(0x1900, 0x194F) }),
TAILE("TaiLe", { CachedRange(0x1950, 0x197F) }),
NEWTAILUE("NewTaiLue", { CachedRange(0x1980, 0x19DF) }),
KHMERSYMBOLS("KhmerSymbols", { CachedRange(0x19E0, 0x19FF) }),
BUGINESE("Buginese", { CachedRange(0x1A00, 0x1A1F) }),
PHONETICEXTENSIONS("PhoneticExtensions", { CachedRange(0x1D00, 0x1D7F) }),
PHONETICEXTENSIONSSUPPLEMENT("PhoneticExtensionsSupplement", { CachedRange(0x1D80, 0x1DBF) }),
COMBININGDIACRITICALMARKSSUPPLEMENT("CombiningDiacriticalMarksSupplement", { CachedRange(0x1DC0, 0x1DFF) }),
LATINEXTENDEDADDITIONAL("LatinExtendedAdditional", { CachedRange(0x1E00, 0x1EFF) }),
GREEKEXTENDED("GreekExtended", { CachedRange(0x1F00, 0x1FFF) }),
GENERALPUNCTUATION("GeneralPunctuation", { CachedRange(0x2000, 0x206F) }),
SUPERSCRIPTSANDSUBSCRIPTS("SuperscriptsandSubscripts", { CachedRange(0x2070, 0x209F) }),
CURRENCYSYMBOLS("CurrencySymbols", { CachedRange(0x20A0, 0x20CF) }),
COMBININGMARKSFORSYMBOLS("CombiningMarksforSymbols", { CachedRange(0x20D0, 0x20FF) }),
LETTERLIKESYMBOLS("LetterlikeSymbols", { CachedRange(0x2100, 0x214F) }),
NUMBERFORMS("NumberForms", { CachedRange(0x2150, 0x218F) }),
ARROWS("Arrows", { CachedRange(0x2190, 0x21FF) }),
MATHEMATICALOPERATORS("MathematicalOperators", { CachedRange(0x2200, 0x22FF) }),
MISCELLANEOUSTECHNICAL("MiscellaneousTechnical", { CachedRange(0x2300, 0x23FF) }),
CONTROLPICTURES("ControlPictures", { CachedRange(0x2400, 0x243F) }),
OPTICALCHARACTERRECOGNITION("OpticalCharacterRecognition", { CachedRange(0x2440, 0x245F) }),
ENCLOSEDALPHANUMERICS("EnclosedAlphanumerics", { CachedRange(0x2460, 0x24FF) }),
BOXDRAWING("BoxDrawing", { CachedRange(0x2500, 0x257F) }),
BLOCKELEMENTS("BlockElements", { CachedRange(0x2580, 0x259F) }),
GEOMETRICSHAPES("GeometricShapes", { CachedRange(0x25A0, 0x25FF) }),
MISCELLANEOUSSYMBOLS("MiscellaneousSymbols", { CachedRange(0x2600, 0x26FF) }),
DINGBATS("Dingbats", { CachedRange(0x2700, 0x27BF) }),
MISCELLANEOUSMATHEMATICALSYMBOLS_A("MiscellaneousMathematicalSymbols-A", { CachedRange(0x27C0, 0x27EF) }),
SUPPLEMENTALARROWS_A("SupplementalArrows-A", { CachedRange(0x27F0, 0x27FF) }),
BRAILLEPATTERNS("BraillePatterns", { CachedRange(0x2800, 0x28FF) }),
SUPPLEMENTALARROWS_B("SupplementalArrows-B", { CachedRange(0x2900, 0x297F) }),
MISCELLANEOUSMATHEMATICALSYMBOLS_B("MiscellaneousMathematicalSymbols-B", { CachedRange(0x2980, 0x29FF) }),
SUPPLEMENTALMATHEMATICALOPERATORS("SupplementalMathematicalOperators", { CachedRange(0x2A00, 0x2AFF) }),
MISCELLANEOUSSYMBOLSANDARROWS("MiscellaneousSymbolsandArrows", { CachedRange(0x2B00, 0x2BFF) }),
GLAGOLITIC("Glagolitic", { CachedRange(0x2C00, 0x2C5F) }),
COPTIC("Coptic", { CachedRange(0x2C80, 0x2CFF) }),
GEORGIANSUPPLEMENT("GeorgianSupplement", { CachedRange(0x2D00, 0x2D2F) }),
TIFINAGH("Tifinagh", { CachedRange(0x2D30, 0x2D7F) }),
ETHIOPICEXTENDED("EthiopicExtended", { CachedRange(0x2D80, 0x2DDF) }),
SUPPLEMENTALPUNCTUATION("SupplementalPunctuation", { CachedRange(0x2E00, 0x2E7F) }),
CJKRADICALSSUPPLEMENT("CJKRadicalsSupplement", { CachedRange(0x2E80, 0x2EFF) }),
KANGXIRADICALS("KangxiRadicals", { CachedRange(0x2F00, 0x2FDF) }),
IDEOGRAPHICDESCRIPTIONCHARACTERS("IdeographicDescriptionCharacters", { CachedRange(0x2FF0, 0x2FFF) }),
CJKSYMBOLSANDPUNCTUATION("CJKSymbolsandPunctuation", { CachedRange(0x3000, 0x303F) }),
HIRAGANA("Hiragana", { CachedRange(0x3040, 0x309F) }),
KATAKANA("Katakana", { CachedRange(0x30A0, 0x30FF) }),
BOPOMOFO("Bopomofo", { CachedRange(0x3100, 0x312F) }),
HANGULCOMPATIBILITYJAMO("HangulCompatibilityJamo", { CachedRange(0x3130, 0x318F) }),
KANBUN("Kanbun", { CachedRange(0x3190, 0x319F) }),
BOPOMOFOEXTENDED("BopomofoExtended", { CachedRange(0x31A0, 0x31BF) }),
CJKSTROKES("CJKStrokes", { CachedRange(0x31C0, 0x31EF) }),
KATAKANAPHONETICEXTENSIONS("KatakanaPhoneticExtensions", { CachedRange(0x31F0, 0x31FF) }),
ENCLOSEDCJKLETTERSANDMONTHS("EnclosedCJKLettersandMonths", { CachedRange(0x3200, 0x32FF) }),
CJKCOMPATIBILITY("CJKCompatibility", { CachedRange(0x3300, 0x33FF) }),
CJKUNIFIEDIDEOGRAPHSEXTENSIONA("CJKUnifiedIdeographsExtensionA", { CachedRange(0x3400, 0x4DB5) }),
YIJINGHEXAGRAMSYMBOLS("YijingHexagramSymbols", { CachedRange(0x4DC0, 0x4DFF) }),
CJKUNIFIEDIDEOGRAPHS("CJKUnifiedIdeographs", { CachedRange(0x4E00, 0x9FFF) }),
YISYLLABLES("YiSyllables", { CachedRange(0xA000, 0xA48F) }),
YIRADICALS("YiRadicals", { CachedRange(0xA490, 0xA4CF) }),
MODIFIERTONELETTERS("ModifierToneLetters", { CachedRange(0xA700, 0xA71F) }),
SYLOTINAGRI("SylotiNagri", { CachedRange(0xA800, 0xA82F) }),
HANGULSYLLABLES("HangulSyllables", { CachedRange(0xAC00, 0xD7A3) }),
HIGHSURROGATES("HighSurrogates", { CachedRange(0xD800, 0xDB7F) }),
HIGHPRIVATEUSESURROGATES("HighPrivateUseSurrogates", { CachedRange(0xDB80, 0xDBFF) }),
LOWSURROGATES("LowSurrogates", { CachedRange(0xDC00, 0xDFFF) }),
PRIVATEUSEAREA("PrivateUseArea", { CachedRange(0xE000, 0xF8FF) }),
CJKCOMPATIBILITYIDEOGRAPHS("CJKCompatibilityIdeographs", { CachedRange(0xF900, 0xFAFF) }),
ALPHABETICPRESENTATIONFORMS("AlphabeticPresentationForms", { CachedRange(0xFB00, 0xFB4F) }),
ARABICPRESENTATIONFORMS_A("ArabicPresentationForms-A", { CachedRange(0xFB50, 0xFDFF) }),
VARIATIONSELECTORS("VariationSelectors", { CachedRange(0xFE00, 0xFE0F) }),
VERTICALFORMS("VerticalForms", { CachedRange(0xFE10, 0xFE1F) }),
COMBININGHALFMARKS("CombiningHalfMarks", { CachedRange(0xFE20, 0xFE2F) }),
CJKCOMPATIBILITYFORMS("CJKCompatibilityForms", { CachedRange(0xFE30, 0xFE4F) }),
SMALLFORMVARIANTS("SmallFormVariants", { CachedRange(0xFE50, 0xFE6F) }),
ARABICPRESENTATIONFORMS_B("ArabicPresentationForms-B", { CachedRange(0xFE70, 0xFEFF) }),
HALFWIDTHANDFULLWIDTHFORMS("HalfwidthandFullwidthForms", { CachedRange(0xFF00, 0xFFEF) }),
ALL("all", { CachedRange(0x00, 0x10FFFF) }),
SPECIALS("Specials", ::CachedSpecialsBlock),
CN("Cn", { CachedCategory(CharCategory.UNASSIGNED.value, true) }),
ISL("IsL", { CachedCategoryScope(0x3E, true) }),
LU("Lu", { CachedCategory(CharCategory.UPPERCASE_LETTER.value, true) }),
LL("Ll", { CachedCategory(CharCategory.LOWERCASE_LETTER.value, true) }),
LT("Lt", { CachedCategory(CharCategory.TITLECASE_LETTER.value, false) }),
LM("Lm", { CachedCategory(CharCategory.MODIFIER_LETTER.value, false) }),
LO("Lo", { CachedCategory(CharCategory.OTHER_LETTER.value, true) }),
ISM("IsM", { CachedCategoryScope(0x1C0, true) }),
MN("Mn", { CachedCategory(CharCategory.NON_SPACING_MARK.value, true) }),
ME("Me", { CachedCategory(CharCategory.ENCLOSING_MARK.value, false) }),
MC("Mc", { CachedCategory(CharCategory.COMBINING_SPACING_MARK.value, true) }),
N("N", { CachedCategoryScope(0xE00, true) }),
ND("Nd", { CachedCategory(CharCategory.DECIMAL_DIGIT_NUMBER.value, true) }),
NL("Nl", { CachedCategory(CharCategory.LETTER_NUMBER.value, true) }),
NO("No", { CachedCategory(CharCategory.OTHER_NUMBER.value, true) }),
ISZ("IsZ", { CachedCategoryScope(0x7000, false) }),
ZS("Zs", { CachedCategory(CharCategory.SPACE_SEPARATOR.value, false) }),
ZL("Zl", { CachedCategory(CharCategory.LINE_SEPARATOR.value, false) }),
ZP("Zp", { CachedCategory(CharCategory.PARAGRAPH_SEPARATOR.value, false) }),
ISC("IsC", { CachedCategoryScope(0xF0000, true, true) }),
CC("Cc", { CachedCategory(CharCategory.CONTROL.value, false) }),
CF("Cf", { CachedCategory(CharCategory.FORMAT.value, true) }),
CO("Co", { CachedCategory(CharCategory.PRIVATE_USE.value, true) }),
CS("Cs", { CachedCategory(CharCategory.SURROGATE.value, false, true) }),
ISP("IsP", { CachedCategoryScope((1 shl CharCategory.DASH_PUNCTUATION.value)
or (1 shl CharCategory.START_PUNCTUATION.value)
or (1 shl CharCategory.END_PUNCTUATION.value)
or (1 shl CharCategory.CONNECTOR_PUNCTUATION.value)
or (1 shl CharCategory.OTHER_PUNCTUATION.value)
or (1 shl CharCategory.INITIAL_QUOTE_PUNCTUATION.value)
or (1 shl CharCategory.FINAL_QUOTE_PUNCTUATION.value), true) }),
PD("Pd", { CachedCategory(CharCategory.DASH_PUNCTUATION.value, false) }),
PS("Ps", { CachedCategory(CharCategory.START_PUNCTUATION.value, false) }),
PE("Pe", { CachedCategory(CharCategory.END_PUNCTUATION.value, false) }),
PC("Pc", { CachedCategory(CharCategory.CONNECTOR_PUNCTUATION.value, false) }),
PO("Po", { CachedCategory(CharCategory.OTHER_PUNCTUATION.value, true) }),
ISS("IsS", { CachedCategoryScope(0x7E000000, true) }),
SM("Sm", { CachedCategory(CharCategory.MATH_SYMBOL.value, true) }),
SC("Sc", { CachedCategory(CharCategory.CURRENCY_SYMBOL.value, false) }),
SK("Sk", { CachedCategory(CharCategory.MODIFIER_SYMBOL.value, false) }),
SO("So", { CachedCategory(CharCategory.OTHER_SYMBOL.value, true) }),
PI("Pi", { CachedCategory(CharCategory.INITIAL_QUOTE_PUNCTUATION.value, false) }),
PF("Pf", { CachedCategory(CharCategory.FINAL_QUOTE_PUNCTUATION.value, false) })
}
private val classCache = Array<AtomicReference<CachedCharClass?>>(CharClasses.values().size, {
AtomicReference<CachedCharClass?>(null)
})
private val classCacheMap = CharClasses.values().associate { it -> it.regexName to it }
fun intersects(ch1: Int, ch2: Int): Boolean = ch1 == ch2
fun intersects(cc: AbstractCharClass, ch: Int): Boolean = cc.contains(ch)
fun intersects(cc1: AbstractCharClass, cc2: AbstractCharClass): Boolean {
if (cc1.bits == null || cc2.bits == null) {
return true
}
return cc1.bits!!.intersects(cc2.bits!!)
}
fun getPredefinedClass(name: String, negative: Boolean): AbstractCharClass {
val charClass = classCacheMap[name] ?: throw PatternSyntaxException("No such character class")
val cachedClass = classCache[charClass.ordinal].value ?: run {
classCache[charClass.ordinal].compareAndSwap(null, charClass.factory().freeze())
classCache[charClass.ordinal].value!!
}
return cachedClass.getValue(negative)
}
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/AbstractLineTerminator.kt
+76
View File
@@ -0,0 +1,76 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
@file:Suppress("DEPRECATION") // Char.toInt()
package kotlin.text.regex
private object unixLT : AbstractLineTerminator() {
override fun isLineTerminator(codepoint: Int): Boolean = (codepoint == '\n'.toInt())
override fun isLineTerminatorPair(char1: Char, char2: Char): Boolean = false
override fun isAfterLineTerminator(previous: Char, checked: Char): Boolean = (previous == '\n')
}
private object unicodeLT : AbstractLineTerminator() {
override fun isLineTerminatorPair(char1: Char, char2: Char): Boolean {
return char1 == '\r' && char2 == '\n'
}
override fun isLineTerminator(codepoint: Int): Boolean {
return codepoint == '\n'.toInt()
|| codepoint == '\r'.toInt()
|| codepoint == '\u0085'.toInt()
|| codepoint or 1 == '\u2029'.toInt()
}
override fun isAfterLineTerminator(previous: Char, checked: Char): Boolean {
return previous == '\n' || previous == '\u0085' || previous.toInt() or 1 == '\u2029'.toInt()
|| previous == '\r' && checked != '\n'
}
}
/**
* Line terminator factory
*/
internal abstract class AbstractLineTerminator {
/** Checks if the single character is a line terminator or not. */
open fun isLineTerminator(char: Char): Boolean = isLineTerminator(char.toInt())
/** Checks if the codepoint is a line terminator or not */
abstract fun isLineTerminator(codepoint: Int): Boolean
/** Checks if the pair of symbols is a line terminator (e.g. for \r\n case) */
abstract fun isLineTerminatorPair(char1: Char, char2: Char): Boolean
/** Checks if a [checked] character is after a line terminator using the [previous] character.*/
abstract fun isAfterLineTerminator(previous: Char, checked: Char): Boolean
companion object {
fun getInstance(flag: Int): AbstractLineTerminator {
if (flag and Pattern.UNIX_LINES != 0) {
return unixLT
} else {
return unicodeLT
}
}
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/CharClass.kt
+541
View File
@@ -0,0 +1,541 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
@file:Suppress("DEPRECATION") // Char.toInt()
package kotlin.text.regex
import kotlin.native.BitSet
/**
* User defined character classes (e.g. [abef]).
*/
// TODO: replace the implementation with one using BitSet for first 256 symbols and a hash table / tree for the rest of UTF.
internal class CharClass(val ignoreCase: Boolean = false, negative: Boolean = false) : AbstractCharClass() {
var invertedSurrogates = false
/**
* Shows if the alt flags was inverted during the range construction process.
* E.g. consider the following range: [\D3]. Here we firstly add the \D char class (which has the alt flag set) into a
* resulting char set. After that the resulting char also has the alt flag set. But then we need to add the '3' character
* into this class with positive sense. So we set the inverted flag to show that the range is not negative ([^..])
* by itself but was inverted during some transformations.
*/
var inverted = false
var hideBits = false
internal var bits_ = BitSet()
override val bits: BitSet?
get() {
if (hideBits)
return null
return bits_
}
var nonBitSet: AbstractCharClass? = null
private val Int.asciiSupplement: Int
get() = when {
this in 'a'.toInt()..'z'.toInt() -> this - 32
this in 'A'.toInt()..'Z'.toInt() -> this + 32
else -> this
}
private val Int.isSurrogate: Boolean
get() = this in Char.MIN_SURROGATE.toInt()..Char.MAX_SURROGATE.toInt()
init {
setNegative(negative)
}
/*
* We can use this method safely even if nonBitSet != null
* due to specific of range constructions in regular expressions.
*/
fun add(ch: Int): CharClass {
var character = ch
if (ignoreCase) {
if (character.toChar() in 'a'..'z' || character.toChar() in 'A'..'Z') {
bits_.set(character.asciiSupplement, !inverted)
} else if (character > 128) {
character = Char.toLowerCase(Char.toUpperCase(character))
}
}
if (character.toChar().isSurrogate()) {
lowHighSurrogates.set(character - Char.MIN_SURROGATE.toInt(), !invertedSurrogates)
}
bits_.set(character, !inverted)
if (!mayContainSupplCodepoints && Char.isSupplementaryCodePoint(ch)) {
mayContainSupplCodepoints = true
}
return this
}
fun add(ch: Char): CharClass = add(ch.toInt())
/*
* The difference between add(AbstractCharClass) and union(AbstractCharClass)
* is that add() is used for constructions like "[^abc\\d]"
* (this pattern doesn't match "1")
* while union is used for constructions like "[^abc[\\d]]"
* (this pattern matches "1").
*/
fun add(another: AbstractCharClass): CharClass {
if (!mayContainSupplCodepoints && another.mayContainSupplCodepoints) {
mayContainSupplCodepoints = true
}
// Process surrogates.
if (!invertedSurrogates) {
//A | !B = ! ((A ^ B) & B)
if (another.altSurrogates) {
lowHighSurrogates.xor(another.lowHighSurrogates)
lowHighSurrogates.and(another.lowHighSurrogates)
altSurrogates = !altSurrogates
invertedSurrogates = true
//A | B
} else {
lowHighSurrogates.or(another.lowHighSurrogates)
}
} else {
//!A | !B = !(A & B)
if (another.altSurrogates) {
lowHighSurrogates.and(another.lowHighSurrogates)
//!A | B = !(A & !B)
} else {
lowHighSurrogates.andNot(another.lowHighSurrogates)
}
}
val anotherBits = another.bits
if (!hideBits && anotherBits != null) {
if (!inverted) {
//A | !B = ! ((A ^ B) & B)
if (another.isNegative()) {
bits_.xor(anotherBits)
bits_.and(anotherBits)
alt = !alt
inverted = true
//A | B
} else {
bits_.or(anotherBits)
}
} else {
//!A | !B = !(A & B)
if (another.isNegative()) {
bits_.and(anotherBits)
//!A | B = !(A & !B)
} else {
bits_.andNot(anotherBits)
}
}
// Some of charclasses hides its bits
} else {
val curAlt = alt
if (nonBitSet == null) {
if (curAlt && !inverted && bits_.isEmpty) {
nonBitSet = object : AbstractCharClass() {
override operator fun contains(ch: Int): Boolean {
return another.contains(ch)
}
}
} else {
/*
* We keep the value of alt unchanged for
* constructions like [^[abc]fgb] by using
* the formula a ^ b == !a ^ !b.
*/
if (curAlt) {
nonBitSet = object : AbstractCharClass() {
override operator fun contains(ch: Int): Boolean {
return !(curAlt xor bits_.get(ch) || curAlt xor inverted xor another.contains(ch))
}
}
} else {
nonBitSet = object : AbstractCharClass() {
override operator fun contains(ch: Int): Boolean {
return curAlt xor bits_.get(ch) || curAlt xor inverted xor another.contains(ch)
}
}
}
}
hideBits = true
} else {
val nb = nonBitSet
if (curAlt) {
nonBitSet = object : AbstractCharClass() {
override operator fun contains(ch: Int): Boolean {
return !(curAlt xor (nb!!.contains(ch) || another.contains(ch)))
}
}
} else {
nonBitSet = object : AbstractCharClass() {
override operator fun contains(ch: Int): Boolean {
return curAlt xor (nb!!.contains(ch) || another.contains(ch))
}
}
}
}
}
return this
}
fun add(start: Int, end: Int): CharClass {
if (start > end)
throw IllegalArgumentException("Incorrect range of symbols (start > end)")
val minSurrogate = Char.MIN_SURROGATE.toInt()
val maxSurrogate = Char.MAX_SURROGATE.toInt()
if (ignoreCase) {
// TODO: Make a faster implementation.
for (i in start..end) {
add(i)
}
} else {
// No intersection with surrogate characters.
if (end < minSurrogate || start > maxSurrogate) {
bits_.set(start, end + 1, !inverted)
} else {
val surrogatesStart = maxOf(start, minSurrogate)
val surrogatesEnd = minOf(end, maxSurrogate)
bits_.set(start, end + 1, !inverted)
lowHighSurrogates.set(surrogatesStart - minSurrogate,
surrogatesEnd - minSurrogate + 1,
!invertedSurrogates)
if (!mayContainSupplCodepoints && end >= Char.MIN_SUPPLEMENTARY_CODE_POINT) {
mayContainSupplCodepoints = true
}
}
}
return this
}
fun add(start: Char, end: Char): CharClass = add(start.toInt(), end.toInt())
// OR operation
fun union(another: AbstractCharClass) {
if (!mayContainSupplCodepoints && another.mayContainSupplCodepoints) {
mayContainSupplCodepoints = true
}
if (altSurrogates xor another.altSurrogates) {
//!A | B = !(A & !B)
if (altSurrogates) {
lowHighSurrogates.andNot(another.lowHighSurrogates)
//A | !B = !((A ^ B) & B)
} else {
lowHighSurrogates.xor(another.lowHighSurrogates)
lowHighSurrogates.and(another.lowHighSurrogates)
altSurrogates = true
}
} else {
//!A | !B = !(A & B)
if (altSurrogates) {
lowHighSurrogates.and(another.lowHighSurrogates)
//A | B
} else {
lowHighSurrogates.or(another.lowHighSurrogates)
}
}
val anotherBits = another.bits
if (!hideBits && anotherBits != null) {
if (alt xor another.isNegative()) {
//!A | B = !(A & !B)
if (alt) {
bits_.andNot(anotherBits)
//A | !B = !((A ^ B) & B)
} else {
bits_.xor(anotherBits)
bits_.and(anotherBits)
alt = true
}
} else {
//!A | !B = !(A & B)
if (alt) {
bits_.and(anotherBits)
//A | B
} else {
bits_.or(anotherBits)
}
}
} else {
val curAlt = alt
if (nonBitSet == null) {
if (!inverted && bits_.isEmpty) {
if (curAlt) {
nonBitSet = object : AbstractCharClass() {
override operator fun contains(ch: Int): Boolean {
return !another.contains(ch)
}
}
} else {
nonBitSet = object : AbstractCharClass() {
override operator fun contains(ch: Int): Boolean {
return another.contains(ch)
}
}
}
} else {
if (curAlt) {
nonBitSet = object : AbstractCharClass() {
override operator fun contains(ch: Int): Boolean {
return !(another.contains(ch) || curAlt xor bits_.get(ch))
}
}
} else {
nonBitSet = object : AbstractCharClass() {
override operator fun contains(ch: Int): Boolean {
return another.contains(ch) || curAlt xor bits_.get(ch)
}
}
}
}
hideBits = true
} else {
val nb = nonBitSet
if (curAlt) {
nonBitSet = object : AbstractCharClass() {
override operator fun contains(ch: Int): Boolean {
return !(curAlt xor nb!!.contains(ch) || another.contains(ch))
}
}
} else {
nonBitSet = object : AbstractCharClass() {
override operator fun contains(ch: Int): Boolean {
return curAlt xor nb!!.contains(ch) || another.contains(ch)
}
}
}
}
}
}
// AND operation
fun intersection(another: AbstractCharClass) {
if (!mayContainSupplCodepoints && another.mayContainSupplCodepoints) {
mayContainSupplCodepoints = true
}
if (altSurrogates xor another.altSurrogates) {
//!A & B = ((A ^ B) & B)
if (altSurrogates) {
lowHighSurrogates.xor(another.lowHighSurrogates)
lowHighSurrogates.and(another.lowHighSurrogates)
altSurrogates = false
//A & !B
} else {
lowHighSurrogates.andNot(another.lowHighSurrogates)
}
} else {
//!A & !B = !(A | B)
if (altSurrogates) {
lowHighSurrogates.or(another.lowHighSurrogates)
//A & B
} else {
lowHighSurrogates.and(another.lowHighSurrogates)
}
}
val anotherBits = another.bits
if (!hideBits && anotherBits != null) {
if (alt xor another.isNegative()) {
//!A & B = ((A ^ B) & B)
if (alt) {
bits_.xor(anotherBits)
bits_.and(anotherBits)
alt = false
//A & !B
} else {
bits_.andNot(anotherBits)
}
} else {
//!A & !B = !(A | B)
if (alt) {
bits_.or(anotherBits)
//A & B
} else {
bits_.and(anotherBits)
}
}
} else {
val curAlt = alt
if (nonBitSet == null) {
if (!inverted && bits_.isEmpty) {
if (curAlt) {
nonBitSet = object : AbstractCharClass() {
override operator fun contains(ch: Int): Boolean {
return !another.contains(ch)
}
}
} else {
nonBitSet = object : AbstractCharClass() {
override operator fun contains(ch: Int): Boolean {
return another.contains(ch)
}
}
}
} else {
if (curAlt) {
nonBitSet = object : AbstractCharClass() {
override operator fun contains(ch: Int): Boolean {
return !(another.contains(ch) && curAlt xor bits_.get(ch))
}
}
} else {
nonBitSet = object : AbstractCharClass() {
override operator fun contains(ch: Int): Boolean {
return another.contains(ch) && curAlt xor bits_.get(ch)
}
}
}
}
hideBits = true
} else {
val nb = nonBitSet
if (curAlt) {
nonBitSet = object : AbstractCharClass() {
override operator fun contains(ch: Int): Boolean {
return !(curAlt xor nb!!.contains(ch) && another.contains(ch))
}
}
} else {
nonBitSet = object : AbstractCharClass() {
override operator fun contains(ch: Int): Boolean {
return curAlt xor nb!!.contains(ch) && another.contains(ch)
}
}
}
}
}
}
/**
* Returns `true` if character class contains symbol specified,
* `false` otherwise. Note: #setNegative() method changes the
* meaning of contains method;
* @param ch
* *
* @return `true` if character class contains symbol specified;
* */
override operator fun contains(ch: Int): Boolean {
if (nonBitSet == null) {
return alt xor bits_.get(ch)
} else {
return alt xor nonBitSet!!.contains(ch)
}
}
@OptIn(ExperimentalStdlibApi::class)
override val instance: AbstractCharClass
get() {
if (nonBitSet == null) {
val bs = bits
val res = object : AbstractCharClass() {
override operator fun contains(ch: Int): Boolean {
return this.alt xor bs!!.get(ch)
}
override fun toString(): String {
val temp = StringBuilder()
var i = bs!!.nextSetBit(0)
while (i >= 0) {
temp.append(Char.toChars(i))
temp.append('|')
i = bs.nextSetBit(i + 1)
}
if (temp.length > 0)
temp.deleteAt(temp.length - 1)
return temp.toString()
}
}
return res.setNegative(isNegative())
} else {
return this
}
}
@OptIn(ExperimentalStdlibApi::class)
//for debugging purposes only
override fun toString(): String {
val temp = StringBuilder()
var i = bits_.nextSetBit(0)
while (i >= 0) {
temp.append(Char.toChars(i))
temp.append('|')
i = bits_.nextSetBit(i + 1)
}
if (temp.length > 0)
temp.deleteAt(temp.length - 1)
return temp.toString()
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/Lexer.kt
+922
View File
@@ -0,0 +1,922 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
// TODO: Licenses.
@file:Suppress("DEPRECATION") // Char.toInt()
package kotlin.text.regex
/**
* This is base class for special tokens like character classes and quantifiers.
*/
internal abstract class SpecialToken {
/**
* Returns the type of the token, may return following values:
* TOK_CHARCLASS - token representing character class;
* TOK_QUANTIFIER - token representing quantifier;
*/
abstract val type: Type
enum class Type {
CHARCLASS,
QUANTIFIER
}
}
internal class Lexer(val patternString: String, flags: Int) {
// The property is set in the init block after some transformations over the pattern string.
private val pattern: CharArray
var flags = flags
private set
// Modes ===========================================================================================================
enum class Mode {
PATTERN,
RANGE,
ESCAPE
}
/**
* Mode: whether the lexer processes character range ([Mode.RANGE]), escaped sequence ([Mode.RANGE])
* or any other part of a regex ([PATTERN]).
*/
var mode = Mode.PATTERN
private set
/** When in [Mode.ESCAPE] mode, this field will save the previous one */
private var savedMode = Mode.PATTERN
fun setModeWithReread(value: Mode) {
if(value == Mode.PATTERN || value == Mode.RANGE) {
mode = value
}
if (mode == Mode.PATTERN) {
reread()
}
}
// Tokens ==========================================================================================================
internal var lookBack: Int = 0 // Previous char read.
private set
internal var currentChar: Int = 0 // Current character read. Returns 0 if there is no more characters.
private set
internal var lookAhead: Int = 0 // Next character.
private set
internal var curSpecialToken: SpecialToken? = null // Current special token (e.g. quantifier)
private set
internal var lookAheadSpecialToken: SpecialToken? = null // Next special token
private set
// Indices in the pattern.
var index = 0 // Current char being processed index.
private set
var prevNonWhitespaceIndex = 0 // Previous non-whitespace character index.
private set
var curTokenIndex = 0 // Current token start index.
private set
var lookAheadTokenIndex = 0 // Next token index.
private set
init {
var processedPattern = patternString
if (flags and Pattern.LITERAL > 0) {
processedPattern = Pattern.quote(patternString)
} else if (flags and Pattern.CANON_EQ > 0) {
processedPattern = Lexer.normalize(patternString)
}
this.pattern = processedPattern.toCharArray().copyOf(processedPattern.length + 2)
this.pattern[this.pattern.size - 1] = 0.toChar()
this.pattern[this.pattern.size - 2] = 0.toChar()
// Read first two tokens.
movePointer()
movePointer()
}
// Character checks ================================================================================================
/** Returns true, if current token is special, i.e. quantifier, or other compound token. */
val isSpecial: Boolean get() = curSpecialToken != null
val isQuantifier: Boolean get() = isSpecial && curSpecialToken!!.type == SpecialToken.Type.QUANTIFIER
val isNextSpecial: Boolean get() = lookAheadSpecialToken != null
private fun Int.isSurrogatePair() : Boolean {
val high = (this ushr 16).toChar()
val low = this.toChar()
return high.isHighSurrogate() && low.isLowSurrogate()
}
private fun Char.isLineSeparator(): Boolean =
this == '\n' || this == '\r' || this == '\u0085' || this.toInt() or 1 == '\u2029'.toInt()
/** Checks if there are any characters in the pattern. */
fun isEmpty(): Boolean =
currentChar == 0 && lookAhead == 0 && index >= pattern.size && !isSpecial
/** Return true if the current character is letter, false otherwise .*/
fun isLetter(): Boolean =
!isEmpty() && !isSpecial && isLetter(currentChar)
/** Check if the current char is high/low surrogate. */
fun isHighSurrogate(): Boolean = currentChar in 0xDBFF..0xD800
fun isLowSurrogate(): Boolean = currentChar in 0xDFFF..0xDC00
fun isSurrogate(): Boolean = isHighSurrogate() || isLowSurrogate()
/**
* Restores flags for Lexer
* @param flags
*/
fun restoreFlags(flags: Int) {
this.flags = flags
lookAhead = currentChar
lookAheadSpecialToken = curSpecialToken
// curTokenIndex is an index of closing bracket ')'
index = curTokenIndex + 1
lookAheadTokenIndex = curTokenIndex
movePointer()
}
override fun toString(): String {
return patternString
}
// Processing index moving =========================================================================================
/** Returns current character and moves string index to the next one. */
operator fun next(): Int {
movePointer()
return lookBack
}
/** Returns current special token and moves string index to the next one */
fun nextSpecial(): SpecialToken? {
val res = curSpecialToken
movePointer()
return res
}
/**
* Reread current character. May be required if a previous token changes mode
* to one with different character interpretation.
*/
private fun reread() {
lookAhead = currentChar
lookAheadSpecialToken = curSpecialToken
index = lookAheadTokenIndex
lookAheadTokenIndex = curTokenIndex
movePointer()
}
/**
* Returns the next character index to read and moves pointer to the next one.
* If comments flag is on this method will skip comments and whitespaces.
*
* The following actions are equivalent if comments flag is off:
* currentChar = pattern[index++] == currentChar = pattern[nextIndex]
*/
private fun nextIndex(): Int {
prevNonWhitespaceIndex = index
if (flags and Pattern.COMMENTS != 0) {
skipComments()
} else {
index++
}
return prevNonWhitespaceIndex
}
/** Skips comments and whitespaces */
private fun skipComments(): Int {
val length = pattern.size - 2
index++
do {
while (index < length && pattern[index].isWhitespace()) {
index++
}
if (index < length && pattern[index] == '#') {
index++
while (index < length && !pattern[index].isLineSeparator()) {
index++
}
} else {
return index
}
} while (true)
}
/**
* Returns the next code point in the pattern string.
*/
private fun nextCodePoint(): Int {
val high = pattern[nextIndex()] // nextIndex skips comments and whitespaces if comments flag is on.
if (high.isHighSurrogate()) {
// Low and high chars may be delimited by spaces.
val lowExpectedIndex = prevNonWhitespaceIndex + 1
if (lowExpectedIndex < pattern.size) {
val low = pattern[lowExpectedIndex]
if (low.isLowSurrogate()) {
nextIndex()
return Char.toCodePoint(high, low)
}
}
}
return high.toInt()
}
/**
* Moves pointer one position right. Saves the current character to [lookBack],
* [lookAhead] to the current one and finally read one more to [lookAhead].
*/
private fun movePointer() {
// swap pointers
lookBack = currentChar
currentChar = lookAhead
curSpecialToken = lookAheadSpecialToken
curTokenIndex = lookAheadTokenIndex
lookAheadTokenIndex = index
var reread: Boolean
do {
// Read the next character, analyze it and construct a token.
lookAhead = if (index < pattern.size) nextCodePoint() else 0
lookAheadSpecialToken = null
if (mode == Mode.ESCAPE) {
processInEscapeMode()
}
reread = when (mode) {
Mode.PATTERN -> processInPatternMode()
Mode.RANGE -> processInRangeMode()
else -> false
}
} while (reread)
}
// Special functions called from [movePointer] function to process chars in different modes ========================
/**
* Processing an escaped sequence like "\Q foo \E". Just skip a character if it is not \E.
* Returns whether we need to reread the character or not
*/
private fun processInEscapeMode(): Boolean {
if (lookAhead == '\\'.toInt()) {
// Need not care about supplementary code points here.
val lookAheadChar: Char = if (index < pattern.size) pattern[nextIndex()] else '\u0000'
lookAhead = lookAheadChar.toInt()
if (lookAheadChar == 'E') {
// If \E found - change the mode to the previous one and shift to the next char.
mode = savedMode
lookAhead = if (index <= pattern.size - 2) nextCodePoint() else 0
} else {
// If \ have no E - make a step back and return.
lookAhead = '\\'.toInt()
index = prevNonWhitespaceIndex
}
}
return false
}
/** Processes a next character in [Mode.PATTERN] mode. Returns whether we need to reread the character or not */
private fun processInPatternMode(): Boolean {
if (lookAhead.isSurrogatePair()) {
return false
}
val lookAheadChar = lookAhead.toChar()
if (lookAheadChar == '\\') {
return processEscapedChar()
}
// TODO: Look like we can create a quantifier here.
when (lookAheadChar) {
// Quantifier (*, +, ?).
'+', '*', '?' -> {
val mode = if (index < pattern.size) pattern[index] else '*'
// look at the next character to determine if the mode is greedy, reluctant or possessive.
when (mode) {
'+' -> { lookAhead = lookAhead or Lexer.QMOD_POSSESSIVE; nextIndex() }
'?' -> { lookAhead = lookAhead or Lexer.QMOD_RELUCTANT; nextIndex() }
else -> lookAhead = lookAhead or Lexer.QMOD_GREEDY
}
}
// Quantifier ({x,y}).
'{' -> lookAheadSpecialToken = processQuantifier()
// $.
'$' -> lookAhead = CHAR_DOLLAR
// A group or a special construction.
'(' -> {
if (pattern[index] != '?') {
// Group
lookAhead = CHAR_LEFT_PARENTHESIS
} else {
// Special constructs (non-capturing groups, look ahead/look behind etc).
nextIndex()
var char = pattern[index]
var isLookBehind = false
do {
if (!isLookBehind) {
when (char) {
// Look ahead or an atomic group.
'!' -> { lookAhead = CHAR_NEG_LOOKAHEAD; nextIndex() }
'=' -> { lookAhead = CHAR_POS_LOOKAHEAD; nextIndex() }
'>' -> { lookAhead = CHAR_ATOMIC_GROUP; nextIndex() }
// Positive / negaitve look behind - need to check the next char.
'<' -> {
nextIndex()
char = pattern[index]
isLookBehind = true
}
// Flags.
else -> {
lookAhead = readFlags()
// We return `res = res or 1 shl 8` from readFlags() if we read (?idmsux-idmsux)
if (lookAhead >= 256) {
// Just flags (no non-capturing group with them). Erase auxiliary bit.
lookAhead = lookAhead and 0xff
flags = lookAhead
lookAhead = lookAhead shl 16
lookAhead = CHAR_FLAGS or lookAhead
} else {
// A non-capturing group with flags: (?<flags>:Foo)
flags = lookAhead
lookAhead = lookAhead shl 16
lookAhead = CHAR_NONCAP_GROUP or lookAhead
}
}
}
} else {
// Process the second char for look behind construction.
isLookBehind = false
when (char) {
'!' -> { lookAhead = CHAR_NEG_LOOKBEHIND; nextIndex() }
'=' -> { lookAhead = CHAR_POS_LOOKBEHIND; nextIndex() }
else -> throw PatternSyntaxException("Unknown look behind", patternString, curTokenIndex)
}
}
} while (isLookBehind)
}
}
')' -> lookAhead = CHAR_RIGHT_PARENTHESIS
'[' -> { lookAhead = CHAR_LEFT_SQUARE_BRACKET; mode = Mode.RANGE }
'^' -> lookAhead = CHAR_CARET
'|' -> lookAhead = CHAR_VERTICAL_BAR
'.' -> lookAhead = CHAR_DOT
}
return false
}
/** Processes a character inside a range. Returns whether we need to reread the character or not */
private fun processInRangeMode(): Boolean {
if (lookAhead.isSurrogatePair()) {
return false
}
val lookAheadChar = lookAhead.toChar()
when (lookAheadChar) {
'\\' -> return processEscapedChar()
'[' -> lookAhead = CHAR_LEFT_SQUARE_BRACKET
']' -> lookAhead = CHAR_RIGHT_SQUARE_BRACKET
'^' -> lookAhead = CHAR_CARET
'&' -> lookAhead = CHAR_AMPERSAND
'-' -> lookAhead = CHAR_HYPHEN
}
return false
}
/** Processes an escaped (\x) character in any mode. Returns whether we need to reread the character or not */
private fun processEscapedChar() : Boolean {
lookAhead = if (index < pattern.size - 2) {
nextCodePoint()
} else {
throw PatternSyntaxException("Trailing \\", patternString, curTokenIndex)
}
// The current code point cannot be a surrogate pair because it is an escaped special one.
// Cast it to char or just skip it as if we pass through the else branch of the when below.
if (lookAhead.isSurrogatePair()) {
return false
}
val lookAheadChar = lookAhead.toChar()
when (lookAheadChar) {
// Character class.
'P', 'p' -> {
val cs = parseCharClassName()
val negative = lookAheadChar == 'P'
lookAheadSpecialToken = AbstractCharClass.getPredefinedClass(cs, negative)
lookAhead = 0
}
// Word/whitespace/digit.
'w', 's', 'd', 'W', 'S', 'D' -> {
lookAheadSpecialToken = AbstractCharClass.getPredefinedClass(
pattern.concatToString(prevNonWhitespaceIndex, prevNonWhitespaceIndex + 1),
false
)
lookAhead = 0
}
// Enter in ESCAPE mode. Skip this \Q symbol.
'Q' -> {
savedMode = mode
mode = Mode.ESCAPE
return true
}
// Special characters like tab, new line etc.
't' -> lookAhead = '\t'.toInt()
'n' -> lookAhead = '\n'.toInt()
'r' -> lookAhead = '\r'.toInt()
'f' -> lookAhead = '\u000C'.toInt()
'a' -> lookAhead = '\u0007'.toInt()
'e' -> lookAhead = '\u001B'.toInt()
// Back references to capturing groups.
'1', '2', '3', '4', '5', '6', '7', '8', '9' -> {
if (mode == Mode.PATTERN) {
lookAhead = 0x80000000.toInt() or lookAhead // Captured group reference is 0x80...<group number>
}
}
// A literal: octal, hex, or hex unicode.
'0' -> lookAhead = readOctals()
'x' -> lookAhead = readHex("hexadecimal", 2)
'u' -> lookAhead = readHex("Unicode", 4)
// Special characters like EOL, EOI etc
'b' -> lookAhead = CHAR_WORD_BOUND
'B' -> lookAhead = CHAR_NONWORD_BOUND
'A' -> lookAhead = CHAR_START_OF_INPUT
'G' -> lookAhead = CHAR_PREVIOUS_MATCH
'Z' -> lookAhead = CHAR_END_OF_LINE
'z' -> lookAhead = CHAR_END_OF_INPUT
// \cx - A control character corresponding to x.
'c' -> {
if (index < pattern.size - 2) {
// Need not care about supplementary codepoints here.
lookAhead = pattern[nextIndex()].toInt() and 0x1f
} else {
throw PatternSyntaxException("Illegal control sequence", patternString, curTokenIndex)
}
}
'C', 'E', 'F', 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', 'R', 'T', 'U', 'V', 'X', 'Y', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'o', 'q', 'y' ->
throw PatternSyntaxException("Illegal escape sequence", patternString, curTokenIndex)
}
return false
}
/** Process [lookAhead] in assumption that it's quantifier. */
private fun processQuantifier(): Quantifier {
assert(lookAhead == '{'.toInt())
val sb = StringBuilder(4)
var min = -1
var max = -1
// Obtain a min value.
var char: Char = if (index < pattern.size) {
pattern[nextIndex()]
} else {
throw PatternSyntaxException("Incorrect Quantifier Syntax", patternString, curTokenIndex)
}
while (char != '}') {
if (char == ',' && min < 0) {
try {
val minParsed = sb.toString().toInt()
min = if (minParsed >= 0) minParsed else throw PatternSyntaxException("Incorrect Quantifier Syntax", patternString, curTokenIndex)
sb.setLength(0)
} catch (nfe: NumberFormatException) {
throw PatternSyntaxException("Incorrect Quantifier Syntax", patternString, curTokenIndex)
}
} else {
sb.append(char)
}
char = if (index < pattern.size) pattern[nextIndex()] else break
}
if (char != '}') {
throw PatternSyntaxException("Incorrect Quantifier Syntax", patternString, curTokenIndex)
}
// Obtain a max value, if it exists
if (sb.isNotEmpty()) {
try {
val maxParsed = sb.toString().toInt()
max = if (maxParsed >= 0) maxParsed else throw PatternSyntaxException("Incorrect Quantifier Syntax", patternString, curTokenIndex)
if (min < 0) {
min = max
}
} catch (nfe: NumberFormatException) {
throw PatternSyntaxException("Incorrect Quantifier Syntax", patternString, curTokenIndex)
}
}
if (min < 0 || max >=0 && max < min) {
throw PatternSyntaxException("Incorrect Quantifier Syntax", patternString, curTokenIndex)
}
val mod = if (index < pattern.size) pattern[index] else '*'
when (mod) {
'+' -> { lookAhead = Lexer.QUANT_COMP_P; nextIndex() }
'?' -> { lookAhead = Lexer.QUANT_COMP_R; nextIndex() }
else -> lookAhead = Lexer.QUANT_COMP
}
return Quantifier(min, max)
}
// Reading methods for specific tokens =============================================================================
/** Process expression flags given with (?idmsux-idmsux). Returns the flags processed. */
private fun readFlags(): Int {
var positive = true
var result = flags
while (index < pattern.size) {
val char = pattern[index]
when (char) {
'-' -> {
if (!positive) {
throw PatternSyntaxException("Illegal inline construct", patternString, curTokenIndex)
}
positive = false
}
'i' -> result = if (positive)
result or Pattern.CASE_INSENSITIVE
else
result xor Pattern.CASE_INSENSITIVE and result
'd' -> result = if (positive)
result or Pattern.UNIX_LINES
else
result xor Pattern.UNIX_LINES and result
'm' -> result = if (positive)
result or Pattern.MULTILINE
else
result xor Pattern.MULTILINE and result
's' -> result = if (positive)
result or Pattern.DOTALL
else
result xor Pattern.DOTALL and result
// We don't support UNICODE_CASE.
/*'u' -> result = if (positive)
result or Pattern.UNICODE_CASE
else
result xor Pattern.UNICODE_CASE and result*/
'x' -> result = if (positive)
result or Pattern.COMMENTS
else
result xor Pattern.COMMENTS and result
':' -> {
nextIndex()
return result
}
')' -> {
nextIndex()
return result or (1 shl 8)
}
}
nextIndex()
}
throw PatternSyntaxException("Illegal inline construct", patternString, curTokenIndex)
}
/** Parse character classes names and verifies correction of the syntax */
private fun parseCharClassName(): String {
val sb = StringBuilder(10)
if (index < pattern.size - 2) {
// one symbol family
if (pattern[index] != '{') {
return "Is${pattern[nextIndex()]}"
}
nextIndex() // Skip '{'
var char = pattern[nextIndex()]
while (index < pattern.size - 2 && char != '}') {
sb.append(char)
char = pattern[nextIndex()]
}
if (char != '}') throw PatternSyntaxException("Unclosed character family", patternString, curTokenIndex)
}
if (sb.isEmpty()) throw PatternSyntaxException("Empty character family", patternString, curTokenIndex)
val res = sb.toString()
return when {
res.length == 1 -> "Is$res"
res.length > 3 && (res.startsWith("Is") || res.startsWith("In")) -> res.substring(2)
else -> res
}
}
/** Process hexadecimal integer. */
private fun readHex(radixName: String, max: Int): Int {
val builder = StringBuilder(max)
val length = pattern.size - 2
var i = 0
while (i < max && index < length) {
builder.append(pattern[nextIndex()])
i++
}
if (i == max) {
try {
return builder.toString().toInt(16)
} catch (e: NumberFormatException) {}
}
throw PatternSyntaxException("Invalid $radixName escape sequence", patternString, curTokenIndex)
}
/** Process octal integer. */
private fun readOctals(): Int {
val length = pattern.size - 2
var result = 0
var digit = digitOf(pattern[index], 8)
if (digit == -1) {
throw PatternSyntaxException("Invalid octal escape sequence", patternString, curTokenIndex)
}
val max = if (digit > 3) 2 else 3
var i = 0
while (i < max && index < length && digit != -1) {
result *= 8
result += digit
nextIndex()
digit = digitOf(pattern[index], 8)
i++
}
return result
}
companion object {
// Special characters.
val CHAR_DOLLAR = 0xe0000000.toInt() or '$'.toInt()
val CHAR_RIGHT_PARENTHESIS = 0xe0000000.toInt() or ')'.toInt()
val CHAR_LEFT_SQUARE_BRACKET = 0xe0000000.toInt() or '['.toInt()
val CHAR_RIGHT_SQUARE_BRACKET = 0xe0000000.toInt() or ']'.toInt()
val CHAR_CARET = 0xe0000000.toInt() or '^'.toInt()
val CHAR_VERTICAL_BAR = 0xe0000000.toInt() or '|'.toInt()
val CHAR_AMPERSAND = 0xe0000000.toInt() or '&'.toInt()
val CHAR_HYPHEN = 0xe0000000.toInt() or '-'.toInt()
val CHAR_DOT = 0xe0000000.toInt() or '.'.toInt()
val CHAR_LEFT_PARENTHESIS = 0x80000000.toInt() or '('.toInt()
val CHAR_NONCAP_GROUP = 0xc0000000.toInt() or '('.toInt()
val CHAR_POS_LOOKAHEAD = 0xe0000000.toInt() or '('.toInt()
val CHAR_NEG_LOOKAHEAD = 0xf0000000.toInt() or '('.toInt()
val CHAR_POS_LOOKBEHIND = 0xf8000000.toInt() or '('.toInt()
val CHAR_NEG_LOOKBEHIND = 0xfc000000.toInt() or '('.toInt()
val CHAR_ATOMIC_GROUP = 0xfe000000.toInt() or '('.toInt()
val CHAR_FLAGS = 0xff000000.toInt() or '('.toInt()
val CHAR_START_OF_INPUT = 0x80000000.toInt() or 'A'.toInt()
val CHAR_WORD_BOUND = 0x80000000.toInt() or 'b'.toInt()
val CHAR_NONWORD_BOUND = 0x80000000.toInt() or 'B'.toInt()
val CHAR_PREVIOUS_MATCH = 0x80000000.toInt() or 'G'.toInt()
val CHAR_END_OF_INPUT = 0x80000000.toInt() or 'z'.toInt()
val CHAR_END_OF_LINE = 0x80000000.toInt() or 'Z'.toInt()
// Quantifier modes.
val QMOD_GREEDY = 0xe0000000.toInt()
val QMOD_RELUCTANT = 0xc0000000.toInt()
val QMOD_POSSESSIVE = 0x80000000.toInt()
// Quantifiers.
val QUANT_STAR = QMOD_GREEDY or '*'.toInt()
val QUANT_STAR_P = QMOD_POSSESSIVE or '*'.toInt()
val QUANT_STAR_R = QMOD_RELUCTANT or '*'.toInt()
val QUANT_PLUS = QMOD_GREEDY or '+'.toInt()
val QUANT_PLUS_P = QMOD_POSSESSIVE or '+'.toInt()
val QUANT_PLUS_R = QMOD_RELUCTANT or '+'.toInt()
val QUANT_ALT = QMOD_GREEDY or '?'.toInt()
val QUANT_ALT_P = QMOD_POSSESSIVE or '?'.toInt()
val QUANT_ALT_R = QMOD_RELUCTANT or '?'.toInt()
val QUANT_COMP = QMOD_GREEDY or '{'.toInt()
val QUANT_COMP_P = QMOD_POSSESSIVE or '{'.toInt()
val QUANT_COMP_R = QMOD_RELUCTANT or '{'.toInt()
/** Returns true if [ch] is a plain token. */
fun isLetter(ch: Int): Boolean {
// All supplementary codepoints have integer value that is >= 0.
return ch >= 0
}
private fun String.codePointAt(index: Int): Int {
val high = this[index]
if (high.isHighSurrogate() && index + 1 < this.length) {
val low = this[index + 1]
if (low.isLowSurrogate()) {
return Char.toCodePoint(high, low)
}
}
return high.toInt()
}
// Decomposition ===============================================================================================
// Maximum length of decomposition.
val MAX_DECOMPOSITION_LENGTH = 4
// Maximum length of Hangul decomposition. Note that MAX_HANGUL_DECOMPOSITION_LENGTH <= MAX_DECOMPOSITION_LENGTH.
val MAX_HANGUL_DECOMPOSITION_LENGTH = 3
/*
* Following constants are needed for Hangul canonical decomposition.
* Hangul decomposition algorithm and constants are taken according
* to description at http://www.unicode.org/versions/Unicode4.0.0/ch03.pdf
* "3.12 Conjoining Jamo Behavior"
*/
const val SBase = 0xAC00
const val LBase = 0x1100
const val VBase = 0x1161
const val TBase = 0x11A7
const val SCount = 11172
const val LCount = 19
const val VCount = 21
const val TCount = 28
const val NCount = 588
// Access to the decomposition tables. =========================================================================
/** Gets canonical class for given codepoint from decomposition mappings table. */
fun getCanonicalClass(ch: Int): Int = getCanonicalClassInternal(ch)
/** Tests Unicode codepoint if it is a boundary of decomposed Unicode codepoint. */
fun isDecomposedCharBoundary(ch: Int): Boolean = getCanonicalClass(ch) == 0
/** Tests if given codepoint is a canonical decomposition of another codepoint. */
fun hasSingleCodepointDecomposition(ch: Int): Boolean = hasSingleCodepointDecompositionInternal(ch)
/** Tests if given codepoint has canonical decomposition and given codepoint's canonical class is not 0. */
fun hasDecompositionNonNullCanClass(ch: Int): Boolean =
(ch == 0x0340) or (ch == 0x0341) or (ch == 0x0343) or (ch == 0x0344)
/** Gets decomposition for given codepoint from decomposition mappings table. */
fun getDecomposition(ch: Int): IntArray? = getDecompositionInternal(ch)
// =============================================================================================================
/**
* Normalize given string.
*/
fun normalize(input: String): String {
val inputChars = input.toCharArray()
val inputLength = inputChars.size
var inputCodePointsIndex = 0
var decompHangulIndex = 0
//codePoints of input
val inputCodePoints = IntArray(inputLength)
//result of canonical decomposition of input
var resCodePoints = IntArray(inputLength * MAX_DECOMPOSITION_LENGTH)
//current symbol's codepoint
var ch: Int
//current symbol's decomposition
var decomp: IntArray?
//result of canonical and Hangul decomposition of input
val decompHangul: IntArray
//result of canonical decomposition of input in UTF-16 encoding
val result = StringBuilder()
var i = 0
while (i < inputLength) {
ch = input.codePointAt(i)
inputCodePoints[inputCodePointsIndex++] = ch
i += if (Char.isSupplementaryCodePoint(ch)) 2 else 1
}
// Canonical decomposition based on mappings in decomposition table.
var resCodePointsIndex = decomposeString(inputCodePoints, inputCodePointsIndex, resCodePoints)
// Canonical ordering.
// See http://www.unicode.org/reports/tr15/#Decomposition for details
resCodePoints = Lexer.getCanonicalOrder(resCodePoints, resCodePointsIndex)
// Decomposition for Hangul syllables.
// See http://www.unicode.org/reports/tr15/#Hangul for details
decompHangul = IntArray(resCodePoints.size)
@Suppress("NAME_SHADOWING")
for (i in 0..resCodePointsIndex - 1) {
val curSymb = resCodePoints[i]
decomp = getHangulDecomposition(curSymb)
if (decomp == null) {
decompHangul[decompHangulIndex++] = curSymb
} else {
// Note that Hangul decompositions have length that is equal 2 or 3.
decompHangul[decompHangulIndex++] = decomp[0]
decompHangul[decompHangulIndex++] = decomp[1]
if (decomp.size == 3) {
decompHangul[decompHangulIndex++] = decomp[2]
}
}
}
// Translating into UTF-16 encoding
@Suppress("NAME_SHADOWING")
for (i in 0..decompHangulIndex - 1) {
result.append(Char.toChars(decompHangul[i]))
}
return result.toString()
}
/**
* Rearrange codepoints in [inputInts] according to canonical order. Return an array with rearranged codepoints.
*/
fun getCanonicalOrder(inputInts: IntArray, length: Int): IntArray {
val inputLength = if (length < inputInts.size)
length
else
inputInts.size
/*
* Simple bubble-sort algorithm. Note that many codepoints have 0 canonical class, so this algorithm works
* almost lineary in overwhelming majority of cases. This is due to specific of Unicode combining
* classes and codepoints.
*/
for (i in 1..inputLength - 1) {
var j = i - 1
val iCanonicalClass = getCanonicalClass(inputInts[i])
val ch: Int
if (iCanonicalClass == 0) {
continue
}
while (j > -1) {
if (getCanonicalClass(inputInts[j]) > iCanonicalClass) {
j = j - 1
} else {
break
}
}
ch = inputInts[i]
for (k in i downTo j + 1 + 1) {
inputInts[k] = inputInts[k - 1]
}
inputInts[j + 1] = ch
}
return inputInts
}
/**
* Gets decomposition for given Hangul syllable.
* This is an implementation of Hangul decomposition algorithm
* according to http://www.unicode.org/versions/Unicode4.0.0/ch03.pdf "3.12 Conjoining Jamo Behavior".
*/
fun getHangulDecomposition(ch: Int): IntArray? {
val SIndex = ch - SBase
if (SIndex < 0 || SIndex >= SCount) {
return null
} else {
val L = LBase + SIndex / NCount
val V = VBase + SIndex % NCount / TCount
var T = SIndex % TCount
val decomp: IntArray
if (T == 0) {
decomp = intArrayOf(L, V)
} else {
T = TBase + T
decomp = intArrayOf(L, V, T)
}
return decomp
}
}
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/MatchResultImpl.kt
+251
View File
@@ -0,0 +1,251 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Match result implementation
*/
internal class MatchResultImpl
/**
* @param input an input sequence for matching/searching.
* @param regex a [Regex] instance used for matching/searching.
* @param rightBound index in the [input] used as a right bound for matching/searching. Exclusive.
*/
constructor (internal val input: CharSequence,
internal val regex: Regex) : MatchResult {
// Harmony's implementation ========================================================================================
private val nativePattern = regex.nativePattern
private val groupCount = nativePattern.capturingGroupCount
private val groupBounds = IntArray(groupCount * 2) { -1 }
private val consumers = IntArray(nativePattern.consumersCount + 1) { -1 }
// Used by quantifiers to store a count of a quantified expression occurrences.
val enterCounters: IntArray = IntArray( maxOf(nativePattern.groupQuantifierCount, 0) )
var startIndex: Int = 0
set (startIndex: Int) {
field = startIndex
if (previousMatch < 0) {
previousMatch = startIndex
}
}
var previousMatch = -1
var mode = Regex.Mode.MATCH
private data class MatchResultState(val groupBounds: IntArray, val consumers: IntArray, val enterCounters: IntArray,
val startIndex: Int, val previousMatch: Int)
private var state: MatchResultState? = null
internal fun saveState() {
state = MatchResultState(groupBounds.copyOf(), consumers.copyOf(), enterCounters.copyOf(), startIndex, previousMatch)
}
internal fun rollbackState(): Boolean {
return state?.let {
it.groupBounds.copyInto(groupBounds)
it.consumers.copyInto(consumers)
it.enterCounters.copyInto(enterCounters)
startIndex = it.startIndex
previousMatch = it.previousMatch
true
} ?: false
}
// MatchResult interface ===========================================================================================
/** The range of indices in the original string where match was captured. */
override val range: IntRange
get() = getStart(0) until getEnd(0)
/** The substring from the input string captured by this match. */
override val value: String
get() = group(0) ?: throw AssertionError("No groupIndex #0 in the match result.")
/**
* A collection of groups matched by the regular expression.
*
* This collection has size of `groupCount + 1` where `groupCount` is the count of groups in the regular expression.
* Groups are indexed from 1 to `groupCount` and group with the index 0 corresponds to the entire match.
*/
// Create one object or several ones?
override val groups: MatchGroupCollection = object: MatchGroupCollection, AbstractCollection<MatchGroup?>() {
override val size: Int
get() = this@MatchResultImpl.groupCount
override fun iterator(): Iterator<MatchGroup?> {
return object: Iterator<MatchGroup?> {
var nextIndex: Int = 0
override fun hasNext(): Boolean {
return nextIndex < size
}
override fun next(): MatchGroup? {
if (!hasNext()) {
throw NoSuchElementException()
}
return get(nextIndex++)
}
}
}
override fun get(index: Int): MatchGroup? {
val value = group(index) ?: return null
return MatchGroup(value, getStart(index) until getEnd(index))
}
}
/**
* A list of matched indexed group values.
*
* This list has size of `groupCount + 1` where `groupCount` is the count of groups in the regular expression.
* Groups are indexed from 1 to `groupCount` and group with the index 0 corresponds to the entire match.
*
* If the group in the regular expression is optional and there were no match captured by that group,
* corresponding item in [groupValues] is an empty string.
*
* @sample: samples.text.Regexps.matchDestructuringToGroupValues
*/
override val groupValues: List<String>
get() = mutableListOf<String>().apply {
for (i in 0 until groupCount) {
this.add(group(i) ?: "")
}
}
override fun next(): MatchResult? {
var nextStart = range.endInclusive + 1
// If the current match is empty - shift by 1.
if (nextStart == range.start) {
nextStart++
}
if (nextStart > input.length) {
return null
}
return regex.find(input, nextStart)
}
// =================================================================================================================
// Harmony's implementation ========================================================================================
fun setConsumed(counter: Int, value: Int) {
this.consumers[counter] = value
}
fun getConsumed(counter: Int): Int {
return this.consumers[counter]
}
fun isCaptured(group: Int): Boolean = getStart(group) >= 0
// Setters and getters for starts and ends of groups ===============================================================
internal fun setStart(group: Int, offset: Int) {
checkGroup(group)
groupBounds[group * 2] = offset
}
internal fun setEnd(group: Int, offset: Int) {
checkGroup(group)
groupBounds[group * 2 + 1] = offset
}
/**
* Returns the index of the first character of the text that matched a given group.
*
* @param group the group, ranging from 0 to groupCount() - 1, with 0 representing the whole pattern.
* @return the character index.
*/
fun getStart(group: Int = 0): Int {
checkGroup(group)
return groupBounds[group * 2]
}
/**
* Returns the index of the first character following the text that matched a given group.
*
* @param group the group, ranging from 0 to groupCount() - 1, with 0 representing the whole pattern.
* @return the character index.
*/
fun getEnd(group: Int = 0): Int {
checkGroup(group)
return groupBounds[group * 2 + 1]
}
// ==================================================================================================
/**
* Returns the text that matched a given group of the regular expression.
*
* @param group the group, ranging from 0 to groupCount() - 1, with 0 representing the whole pattern.
* @return the text that matched the group.
*/
fun group(group: Int = 0): String? {
val start = getStart(group)
val end = getEnd(group)
if (start < 0 || end < 0) {
return null
}
return input.subSequence(getStart(group), getEnd(group)).toString()
}
/**
* Returns the number of groups in the result, which is always equal to
* the number of groups in the original regular expression.
*
* @return the number of groups.
*/
fun groupCount(): Int {
return groupCount - 1
}
/*
* This method being called after any successful match; For now it's being
* used to check zero group for empty match;
*/
fun finalizeMatch() {
if (this.groupBounds[0] == -1) {
this.groupBounds[0] = this.startIndex
this.groupBounds[1] = this.startIndex
}
previousMatch = getEnd()
}
private fun checkGroup(group: Int) {
if (group < 0 || group > groupCount) {
throw IndexOutOfBoundsException("Group index out of bounds: $group")
}
}
fun updateGroup(index: Int, srtOffset: Int, endOffset: Int) {
checkGroup(index)
groupBounds[index * 2] = srtOffset
groupBounds[index * 2 + 1] = endOffset
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/Pattern.kt
+879
View File
@@ -0,0 +1,879 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
@file:Suppress("DEPRECATION") // Char.toInt()
package kotlin.text.regex
/** Represents a compiled pattern used by [Regex] for matching, searching, or replacing strings. */
internal class Pattern(val pattern: String, flags: Int = 0) {
var flags = flags
private set
/** A lexer instance used to get tokens from the pattern. */
private val lexemes = Lexer(pattern, flags)
/* All back references that may be used in pattern. */
private val backRefs = arrayOfNulls<FSet>(BACK_REF_NUMBER)
/** Is true if back referenced sets replacement by second compilation pass is needed.*/
private var needsBackRefReplacement = false
/** Global count of found capturing groups. */
var capturingGroupCount = 0
private set
/** A number of group quantifiers in the pattern */
var groupQuantifierCount = 0
private set
/**
* A number of consumers found in the pattern.
* Consumer is any expression ending with an FSet except capturing groups (they are counted by [capturingGroupCount])
*/
var consumersCount = 0
private set
/** A node to start a matching/searching process by call startNode.matches/startNode.find. */
internal val startNode: AbstractSet
/** Compiles the given pattern */
init {
if (flags != 0 && flags or flagsBitMask != flagsBitMask) {
throw IllegalArgumentException("Invalid match flags value")
}
startNode = processExpression(-1, this.flags, null)
if (!lexemes.isEmpty()) {
throw PatternSyntaxException("Trailing characters", pattern, lexemes.curTokenIndex)
}
// Finalize compilation
if (needsBackRefReplacement) {
startNode.processSecondPass()
}
}
override fun toString(): String = pattern
/** Return true if the pattern has the specified flag */
private fun hasFlag(flag: Int): Boolean = flags and flag == flag
// Compilation methods. ============================================================================================
/** A->(a|)+ */
private fun processAlternations(last: AbstractSet): AbstractSet {
val auxRange = CharClass(hasFlag(Pattern.CASE_INSENSITIVE))
while (!lexemes.isEmpty() && lexemes.isLetter()
&& (lexemes.lookAhead == 0
|| lexemes.lookAhead == Lexer.CHAR_VERTICAL_BAR
|| lexemes.lookAhead == Lexer.CHAR_RIGHT_PARENTHESIS)) {
auxRange.add(lexemes.next())
if (lexemes.currentChar == Lexer.CHAR_VERTICAL_BAR) {
lexemes.next()
}
}
val rangeSet = processRangeSet(auxRange)
rangeSet.next = last
return rangeSet
}
/** E->AE; E->S|E; E->S; A->(a|)+ E->S(|S)* */
private fun processExpression(ch: Int, newFlags: Int, last: AbstractSet?): AbstractSet {
val children = ArrayList<AbstractSet>()
val savedFlags = flags
var saveChangedFlags = false
if (newFlags != flags) {
flags = newFlags
}
// Create a right finalizing set.
val fSet: FSet
when (ch) {
// Special groups: non-capturing, look ahead/behind etc.
Lexer.CHAR_NONCAP_GROUP -> fSet = NonCapFSet(consumersCount++)
Lexer.CHAR_POS_LOOKAHEAD,
Lexer.CHAR_NEG_LOOKAHEAD -> fSet = AheadFSet()
Lexer.CHAR_POS_LOOKBEHIND,
Lexer.CHAR_NEG_LOOKBEHIND -> fSet = BehindFSet(consumersCount++)
Lexer.CHAR_ATOMIC_GROUP -> fSet = AtomicFSet(consumersCount++)
// A Capturing group.
else -> {
if (last == null) {
// Whole pattern - group #0.
fSet = FinalSet()
saveChangedFlags = true
} else {
fSet = FSet(capturingGroupCount)
}
if (capturingGroupCount < BACK_REF_NUMBER) {
backRefs[capturingGroupCount] = fSet
}
capturingGroupCount++
}
}
//Process to EOF or ')'
do {
val child: AbstractSet
when {
// a|...
lexemes.isLetter() && lexemes.lookAhead == Lexer.CHAR_VERTICAL_BAR -> child = processAlternations(fSet)
// ..|.., e.g. in "a||||b"
lexemes.currentChar == Lexer.CHAR_VERTICAL_BAR -> {
child = EmptySet(fSet)
lexemes.next()
}
else -> {
child = processSubExpression(fSet)
if (lexemes.currentChar == Lexer.CHAR_VERTICAL_BAR) {
lexemes.next()
}
}
}
children.add(child)
} while (!(lexemes.isEmpty() || lexemes.currentChar == Lexer.CHAR_RIGHT_PARENTHESIS))
// |) or |<EOF> - add an empty node.
if (lexemes.lookBack == Lexer.CHAR_VERTICAL_BAR) {
children.add(EmptySet(fSet))
}
// Restore flags.
if (flags != savedFlags && !saveChangedFlags) {
flags = savedFlags
lexemes.restoreFlags(flags)
}
when (ch) {
Lexer.CHAR_NONCAP_GROUP -> return NonCapturingJointSet(children, fSet)
Lexer.CHAR_POS_LOOKAHEAD -> return PositiveLookAheadSet(children, fSet)
Lexer.CHAR_NEG_LOOKAHEAD -> return NegativeLookAheadSet(children, fSet)
Lexer.CHAR_POS_LOOKBEHIND -> return PositiveLookBehindSet(children, fSet)
Lexer.CHAR_NEG_LOOKBEHIND -> return NegativeLookBehindSet(children, fSet)
Lexer.CHAR_ATOMIC_GROUP -> return AtomicJointSet(children, fSet)
else -> when (children.size) {
0 -> return EmptySet(fSet)
1 -> return SingleSet(children[0], fSet)
else -> return JointSet(children, fSet)
}
}
}
/**
* T->aaa
*/
private fun processSequence(): AbstractSet {
val substring = StringBuilder()
while (!lexemes.isEmpty()
&& lexemes.isLetter()
&& !lexemes.isSurrogate()
&& (!lexemes.isNextSpecial && lexemes.lookAhead == 0 // End of a pattern.
|| !lexemes.isNextSpecial && Lexer.isLetter(lexemes.lookAhead)
|| lexemes.lookAhead == Lexer.CHAR_RIGHT_PARENTHESIS
|| lexemes.lookAhead and 0x8000ffff.toInt() == Lexer.CHAR_LEFT_PARENTHESIS
|| lexemes.lookAhead == Lexer.CHAR_VERTICAL_BAR
|| lexemes.lookAhead == Lexer.CHAR_DOLLAR)) {
val ch = lexemes.next()
if (Char.isSupplementaryCodePoint(ch)) {
substring.append(Char.toChars(ch))
} else {
substring.append(ch.toChar())
}
}
return SequenceSet(substring, hasFlag(CASE_INSENSITIVE))
}
/**
* D->a
*/
private fun processDecomposedChar(): AbstractSet {
val codePoints = IntArray(Lexer.MAX_DECOMPOSITION_LENGTH)
val codePointsHangul: CharArray
var readCodePoints = 0
var curSymb = -1
var curSymbIndex = -1
if (!lexemes.isEmpty() && lexemes.isLetter()) {
curSymb = lexemes.next()
codePoints[readCodePoints] = curSymb
curSymbIndex = curSymb - Lexer.LBase
}
/*
* We process decomposed Hangul syllable LV or LVT or process jamo L.
* See http://www.unicode.org/versions/Unicode4.0.0/ch03.pdf
* "3.12 Conjoining Jamo Behavior"
*/
if (curSymbIndex >= 0 && curSymbIndex < Lexer.LCount) {
codePointsHangul = CharArray(Lexer.MAX_HANGUL_DECOMPOSITION_LENGTH)
codePointsHangul[readCodePoints++] = curSymb.toChar()
curSymb = lexemes.currentChar
curSymbIndex = curSymb - Lexer.VBase
if (curSymbIndex >= 0 && curSymbIndex < Lexer.VCount) {
codePointsHangul[readCodePoints++] = curSymb.toChar()
lexemes.next()
curSymb = lexemes.currentChar
curSymbIndex = curSymb - Lexer.TBase
if (curSymbIndex >= 0 && curSymbIndex < Lexer.TCount) {
codePointsHangul[@Suppress("UNUSED_CHANGED_VALUE")readCodePoints++] = curSymb.toChar()
lexemes.next()
//LVT syllable
return HangulDecomposedCharSet(codePointsHangul, 3)
} else {
//LV syllable
return HangulDecomposedCharSet(codePointsHangul, 2)
}
} else {
//L jamo
return CharSet(codePointsHangul[0], hasFlag(CASE_INSENSITIVE))
}
/*
* We process single codepoint or decomposed codepoint.
* We collect decomposed codepoint and obtain
* one DecomposedCharSet.
*/
} else {
readCodePoints++
while (readCodePoints < Lexer.MAX_DECOMPOSITION_LENGTH
&& !lexemes.isEmpty() && lexemes.isLetter()
&& !Lexer.isDecomposedCharBoundary(lexemes.currentChar)) {
codePoints[readCodePoints++] = lexemes.next()
}
/*
* We have read an ordinary symbol.
*/
if (readCodePoints == 1 && !Lexer.hasSingleCodepointDecomposition(codePoints[0])) {
return processCharSet(codePoints[0])
} else {
return DecomposedCharSet(codePoints, readCodePoints)
}
}
}
/**
* S->BS; S->QS; S->Q; B->a+
*/
private fun processSubExpression(last: AbstractSet): AbstractSet {
var cur: AbstractSet
when {
lexemes.isLetter() && !lexemes.isNextSpecial && Lexer.isLetter(lexemes.lookAhead) -> {
when {
hasFlag(Pattern.CANON_EQ) -> {
cur = processDecomposedChar()
if (!lexemes.isEmpty()
&& (lexemes.currentChar != Lexer.CHAR_RIGHT_PARENTHESIS || last is FinalSet)
&& lexemes.currentChar != Lexer.CHAR_VERTICAL_BAR
&& !lexemes.isLetter()) {
cur = processQuantifier(last, cur)
}
}
lexemes.isHighSurrogate() || lexemes.isLowSurrogate() -> {
val term = processTerminal(last)
cur = processQuantifier(last, term)
}
else -> {
cur = processSequence()
}
}
}
lexemes.currentChar == Lexer.CHAR_RIGHT_PARENTHESIS -> {
if (last is FinalSet) {
throw PatternSyntaxException("unmatched )", pattern, lexemes.curTokenIndex)
}
cur = EmptySet(last)
}
else -> {
val term = processTerminal(last)
cur = processQuantifier(last, term)
}
}
if (!lexemes.isEmpty()
&& (lexemes.currentChar != Lexer.CHAR_RIGHT_PARENTHESIS || last is FinalSet)
&& lexemes.currentChar != Lexer.CHAR_VERTICAL_BAR) {
val next = processSubExpression(last)
if (cur is LeafQuantifierSet
// '*' or '{0,}' quantifier
&& cur.max == Quantifier.INF
&& cur.min == 0
&& !next.first(cur.innerSet)) {
// An Optimizer node for the case where there is no intersection with the next node
cur = UnifiedQuantifierSet(cur)
}
cur.next = next
} else {
cur.next = last
}
return cur
}
/**
* Q->T(*|+|?...) also do some optimizations.
*/
private fun processQuantifier(last: AbstractSet, term: AbstractSet): AbstractSet {
val quant = lexemes.currentChar
if (term !is LeafSet) {
return when (quant) {
Lexer.QUANT_STAR, Lexer.QUANT_PLUS -> {
val q: QuantifierSet
lexemes.next()
if (term.type == AbstractSet.TYPE_DOTSET) {
q = DotQuantifierSet(term, last, quant, AbstractLineTerminator.getInstance(flags), hasFlag(Pattern.DOTALL))
} else {
q = GroupQuantifierSet(Quantifier.fromLexerToken(quant), term, last, quant, groupQuantifierCount++)
}
term.next = q
q
}
Lexer.QUANT_ALT -> {
lexemes.next()
val q = GroupQuantifierSet(Quantifier.fromLexerToken(quant), term, last, quant, groupQuantifierCount++)
term.next = q
q
}
Lexer.QUANT_STAR_R, Lexer.QUANT_PLUS_R, Lexer.QUANT_ALT_R -> {
lexemes.next()
val q = ReluctantGroupQuantifierSet(Quantifier.fromLexerToken(quant), term, last, quant, groupQuantifierCount++)
term.next = q
q
}
Lexer.QUANT_PLUS_P, Lexer.QUANT_STAR_P, Lexer.QUANT_ALT_P -> {
lexemes.next()
PossessiveGroupQuantifierSet(Quantifier.fromLexerToken(quant), term, last, quant, groupQuantifierCount++)
}
Lexer.QUANT_COMP -> {
val q = GroupQuantifierSet(lexemes.nextSpecial() as Quantifier, term, last, Lexer.QUANT_ALT, groupQuantifierCount++)
term.next = q
q
}
Lexer.QUANT_COMP_R -> {
val q = ReluctantGroupQuantifierSet(lexemes.nextSpecial() as Quantifier, term, last, Lexer.QUANT_ALT, groupQuantifierCount++)
term.next = q
q
}
Lexer.QUANT_COMP_P -> {
return PossessiveGroupQuantifierSet(lexemes.nextSpecial() as Quantifier, term, last, Lexer.QUANT_ALT, groupQuantifierCount++)
}
else -> term
}
} else {
val leaf: LeafSet = term
return when (quant) {
Lexer.QUANT_STAR, Lexer.QUANT_PLUS -> {
lexemes.next()
val q = LeafQuantifierSet(Quantifier.fromLexerToken(quant), leaf, last, quant)
leaf.next = q
q
}
Lexer.QUANT_STAR_R, Lexer.QUANT_PLUS_R -> {
lexemes.next()
val q = ReluctantLeafQuantifierSet(Quantifier.fromLexerToken(quant), leaf, last, quant)
leaf.next = q
q
}
Lexer.QUANT_PLUS_P, Lexer.QUANT_STAR_P -> {
lexemes.next()
val q = PossessiveLeafQuantifierSet(Quantifier.fromLexerToken(quant), leaf, last, quant)
leaf.next = q
q
}
Lexer.QUANT_ALT -> {
lexemes.next()
LeafQuantifierSet(Quantifier.altQuantifier, leaf, last, Lexer.QUANT_ALT)
}
Lexer.QUANT_ALT_R -> {
lexemes.next()
ReluctantLeafQuantifierSet(Quantifier.altQuantifier, leaf, last, Lexer.QUANT_ALT_R)
}
Lexer.QUANT_ALT_P -> {
lexemes.next()
PossessiveLeafQuantifierSet(Quantifier.altQuantifier, leaf, last, Lexer.QUANT_ALT_P)
}
Lexer.QUANT_COMP -> {
LeafQuantifierSet(lexemes.nextSpecial() as Quantifier, leaf, last, Lexer.QUANT_COMP)
}
Lexer.QUANT_COMP_R -> {
ReluctantLeafQuantifierSet(lexemes.nextSpecial() as Quantifier, leaf, last, Lexer.QUANT_COMP_R)
}
Lexer.QUANT_COMP_P -> {
ReluctantLeafQuantifierSet(lexemes.nextSpecial() as Quantifier, leaf, last, Lexer.QUANT_COMP_P)
}
else -> term
}
}
}
/**
* T-> letter|[range]|{char-class}|(E)
*/
private fun processTerminal(last: AbstractSet): AbstractSet {
val term: AbstractSet
var char = lexemes.currentChar
// Process flags: (?...)(?...)...
while (char and 0xff00ffff.toInt() == Lexer.CHAR_FLAGS) {
lexemes.next()
flags = (char shr 16) and flagsBitMask
char = lexemes.currentChar
}
// The terminal is some kind of group: (E). Call processExpression for it.
if (char and 0x8000ffff.toInt() == Lexer.CHAR_LEFT_PARENTHESIS) {
lexemes.next()
var newFlags = flags
if (char and 0xff00ffff.toInt() == Lexer.CHAR_NONCAP_GROUP) {
newFlags = (char shr 16) and flagsBitMask
}
term = processExpression(char and 0xff00ffff.toInt(), newFlags, last) // Remove flags from the token.
if (lexemes.currentChar != Lexer.CHAR_RIGHT_PARENTHESIS) {
throw PatternSyntaxException("unmatched (", pattern, lexemes.curTokenIndex)
}
lexemes.next()
} else {
// Other terminals.
when (char) {
Lexer.CHAR_LEFT_SQUARE_BRACKET -> { // Range: [...]
lexemes.next()
var negative = false
if (lexemes.currentChar == Lexer.CHAR_CARET) {
negative = true
lexemes.next()
}
term = processRange(negative, last)
if (lexemes.currentChar != Lexer.CHAR_RIGHT_SQUARE_BRACKET) {
throw PatternSyntaxException("unmatched [", pattern, lexemes.curTokenIndex)
}
lexemes.setModeWithReread(Lexer.Mode.PATTERN)
lexemes.next()
}
Lexer.CHAR_DOT -> { // Dot: .
lexemes.next()
term = DotSet(AbstractLineTerminator.getInstance(flags), hasFlag(DOTALL))
}
Lexer.CHAR_CARET -> { // Beginning of the string: ^
lexemes.next()
term = SOLSet(AbstractLineTerminator.getInstance(flags), hasFlag(MULTILINE))
consumersCount++
}
Lexer.CHAR_DOLLAR -> { // End of the string: $
lexemes.next()
term = EOLSet(consumersCount++, AbstractLineTerminator.getInstance(flags), hasFlag(MULTILINE))
}
// Word / non-word boundary.
Lexer.CHAR_WORD_BOUND -> {
lexemes.next()
term = WordBoundarySet(true)
}
Lexer.CHAR_NONWORD_BOUND -> {
lexemes.next()
term = WordBoundarySet(false)
}
Lexer.CHAR_END_OF_INPUT -> { // End of an input: \z
lexemes.next()
term = EOISet()
}
Lexer.CHAR_END_OF_LINE -> { // End of a line: \Z
lexemes.next()
term = EOLSet(consumersCount++, AbstractLineTerminator.getInstance(flags))
}
Lexer.CHAR_START_OF_INPUT -> { // Start if an input: \A
lexemes.next()
term = SOLSet(AbstractLineTerminator.getInstance(flags))
}
Lexer.CHAR_PREVIOUS_MATCH -> { // A previous match: \G
lexemes.next()
term = PreviousMatchSet()
}
// Back references: \1, \2 etc.
0x80000000.toInt() or '1'.toInt(),
0x80000000.toInt() or '2'.toInt(),
0x80000000.toInt() or '3'.toInt(),
0x80000000.toInt() or '4'.toInt(),
0x80000000.toInt() or '5'.toInt(),
0x80000000.toInt() or '6'.toInt(),
0x80000000.toInt() or '7'.toInt(),
0x80000000.toInt() or '8'.toInt(),
0x80000000.toInt() or '9'.toInt() -> {
val number = (char and 0x7FFFFFFF) - '0'.toInt()
if (number < capturingGroupCount) { // All is ok - the group exists.
lexemes.next()
term = BackReferenceSet(number, consumersCount++, hasFlag(CASE_INSENSITIVE))
// backRefs[number] is proved to be not null because the group is already created (number < capturingGroupCount)
backRefs[number]!!.isBackReferenced = true
needsBackRefReplacement = true // And process back references in the second pass.
} else {
throw PatternSyntaxException("No such group yet exists at this point in the pattern", pattern, lexemes.curTokenIndex)
}
}
// A special token (\D, \w etc), 'u0000' or the end of the pattern.
0 -> {
val cc: AbstractCharClass? = lexemes.curSpecialToken as AbstractCharClass?
when {
cc != null -> {
term = processRangeSet(cc)
lexemes.next()
}
!lexemes.isEmpty() -> {
term = CharSet(char.toChar())
lexemes.next()
}
else -> term = EmptySet(last)
}
}
else -> {
when {
// A regular character.
char >= 0 && !lexemes.isSpecial -> {
term = processCharSet(char)
lexemes.next()
}
char == Lexer.CHAR_VERTICAL_BAR -> {
term = EmptySet(last)
}
char == Lexer.CHAR_RIGHT_PARENTHESIS -> {
if (last is FinalSet) {
throw PatternSyntaxException("unmatched )", pattern, lexemes.curTokenIndex)
}
term = EmptySet(last)
}
else -> {
val current = if (lexemes.isSpecial) lexemes.curSpecialToken.toString() else char.toString()
throw PatternSyntaxException("Dangling meta construction: $current", pattern, lexemes.curTokenIndex)
}
}
}
}
}
return term
}
/**
* Process [...] ranges
*/
private fun processRange(negative: Boolean, last: AbstractSet): AbstractSet {
val res = processRangeExpression(negative)
val rangeSet = processRangeSet(res)
rangeSet.next = last
return rangeSet
}
private fun processRangeExpression(alt: Boolean): CharClass {
var result = CharClass(hasFlag(Pattern.CASE_INSENSITIVE), alt)
var buffer = -1
var intersection = false
var firstInClass = true
var notClosed = lexemes.currentChar != Lexer.CHAR_RIGHT_SQUARE_BRACKET
while (!lexemes.isEmpty() && (notClosed || firstInClass)) {
when (lexemes.currentChar) {
Lexer.CHAR_RIGHT_SQUARE_BRACKET -> {
if (buffer >= 0) {
result.add(buffer)
}
buffer = ']'.toInt()
lexemes.next()
}
Lexer.CHAR_LEFT_SQUARE_BRACKET -> {
if (buffer >= 0) {
result.add(buffer)
buffer = -1
}
lexemes.next()
var negative = false
if (lexemes.currentChar == Lexer.CHAR_CARET) {
lexemes.next()
negative = true
}
if (intersection)
result.intersection(processRangeExpression(negative))
else
result.union(processRangeExpression(negative))
intersection = false
lexemes.next()
}
Lexer.CHAR_AMPERSAND -> {
if (buffer >= 0) {
result.add(buffer)
}
buffer = lexemes.next() // buffer == Lexer.CHAR_AMPERSAND since next() returns currentChar.
/*
* If there is a start for subrange we will do an intersection
* otherwise treat '&' as a normal character
*/
if (lexemes.currentChar == Lexer.CHAR_AMPERSAND) {
if (lexemes.lookAhead == Lexer.CHAR_LEFT_SQUARE_BRACKET) {
lexemes.next()
intersection = true
buffer = -1
} else {
lexemes.next()
if (firstInClass) {
// Skip "&&" at "[&&...]" or "[^&&...]"
result = processRangeExpression(false)
} else {
// Ignore "&&" at "[X&&]" ending where X != empty string
if (lexemes.currentChar != Lexer.CHAR_RIGHT_SQUARE_BRACKET) {
result.intersection(processRangeExpression(false))
}
}
}
} else {
//treat '&' as a normal character
buffer = '&'.toInt()
}
}
Lexer.CHAR_HYPHEN -> {
if (firstInClass
|| lexemes.lookAhead == Lexer.CHAR_RIGHT_SQUARE_BRACKET
|| lexemes.lookAhead == Lexer.CHAR_LEFT_SQUARE_BRACKET
|| buffer < 0) {
// Treat the hypen as a normal character.
if (buffer >= 0) {
result.add(buffer)
}
buffer = '-'.toInt()
lexemes.next()
} else {
// A range.
lexemes.next()
var cur = lexemes.currentChar
if (!lexemes.isSpecial
&& (cur >= 0
|| lexemes.lookAhead == Lexer.CHAR_RIGHT_SQUARE_BRACKET
|| lexemes.lookAhead == Lexer.CHAR_LEFT_SQUARE_BRACKET
|| buffer < 0)) {
try {
if (!Lexer.isLetter(cur)) {
cur = cur and 0xFFFF
}
result.add(buffer, cur)
} catch (e: Exception) {
throw PatternSyntaxException("Illegal character range", pattern, lexemes.curTokenIndex)
}
lexemes.next()
buffer = -1
} else {
throw PatternSyntaxException("Illegal character range", pattern, lexemes.curTokenIndex)
}
}
}
Lexer.CHAR_CARET -> {
if (buffer >= 0) {
result.add(buffer)
}
buffer = '^'.toInt()
lexemes.next()
}
0 -> {
if (buffer >= 0) {
result.add(buffer)
}
val cs = lexemes.curSpecialToken as AbstractCharClass?
if (cs != null) {
result.add(cs)
buffer = -1
} else {
buffer = 0
}
lexemes.next()
}
else -> {
if (buffer >= 0) {
result.add(buffer)
}
buffer = lexemes.next()
}
}
firstInClass = false
notClosed = lexemes.currentChar != Lexer.CHAR_RIGHT_SQUARE_BRACKET
}
if (notClosed) {
throw PatternSyntaxException("Missing ']'", pattern, lexemes.curTokenIndex)
}
if (buffer >= 0) {
result.add(buffer)
}
return result
}
private fun processRangeSet(charClass: AbstractCharClass): AbstractSet {
if (charClass.hasLowHighSurrogates()) {
val lowHighSurrRangeSet = SurrogateRangeSet(charClass.classWithSurrogates())
if (charClass.mayContainSupplCodepoints) {
return CompositeRangeSet(SupplementaryRangeSet(charClass.classWithoutSurrogates(), hasFlag(CASE_INSENSITIVE)), lowHighSurrRangeSet)
}
return CompositeRangeSet(RangeSet(charClass.classWithoutSurrogates(), hasFlag(CASE_INSENSITIVE)), lowHighSurrRangeSet)
}
if (charClass.mayContainSupplCodepoints) {
return SupplementaryRangeSet(charClass, hasFlag(CASE_INSENSITIVE))
}
return RangeSet(charClass, hasFlag(CASE_INSENSITIVE))
}
private fun processCharSet(ch: Int): AbstractSet {
val isSupplCodePoint = Char.isSupplementaryCodePoint(ch)
return when {
isSupplCodePoint -> SequenceSet(Char.toChars(ch).concatToString(0, 2), hasFlag(CASE_INSENSITIVE))
ch.toChar().isLowSurrogate() -> LowSurrogateCharSet(ch.toChar())
ch.toChar().isHighSurrogate() -> HighSurrogateCharSet(ch.toChar())
else -> CharSet(ch.toChar(), hasFlag(CASE_INSENSITIVE))
}
}
companion object {
//TODO: Use RegexOption enum here.
// Flags.
/**
* This constant specifies that a pattern matches Unix line endings ('\n')
* only against the '.', '^', and '$' meta characters.
*/
val UNIX_LINES = 1 shl 0
/**
* This constant specifies that a `Pattern` is matched
* case-insensitively. That is, the patterns "a+" and "A+" would both match
* the string "aAaAaA".
*/
val CASE_INSENSITIVE = 1 shl 1
/**
* This constant specifies that a `Pattern` may contain whitespace or
* comments. Otherwise comments and whitespace are taken as literal
* characters.
*/
val COMMENTS = 1 shl 2
/**
* This constant specifies that the meta characters '^' and '$' match only
* the beginning and end end of an input line, respectively. Normally, they
* match the beginning and the end of the complete input.
*/
val MULTILINE = 1 shl 3
/**
* This constant specifies that the whole `Pattern` is to be taken
* literally, that is, all meta characters lose their meanings.
*/
val LITERAL = 1 shl 4
/**
* This constant specifies that the '.' meta character matches arbitrary
* characters, including line endings, which is normally not the case.
*/
val DOTALL = 1 shl 5
/**
* This constant specifies that a character in a `Pattern` and a
* character in the input string only match if they are canonically
* equivalent.
*/
val CANON_EQ = 1 shl 6
/** Max number of back references supported. */
internal val BACK_REF_NUMBER = 10
/** A bit mask that includes all defined match flags */
internal val flagsBitMask = Pattern.UNIX_LINES or
Pattern.CASE_INSENSITIVE or
Pattern.COMMENTS or
Pattern.MULTILINE or
Pattern.LITERAL or
Pattern.DOTALL or
Pattern.CANON_EQ
/**
* Quotes a given string using "\Q" and "\E", so that all other meta-characters lose their special meaning.
* If the string is used for a `Pattern` afterwards, it can only be matched literally.
*/
fun quote(s: String): String {
return StringBuilder()
.append("\\Q")
.append(s.replace("\\E", "\\E\\\\E\\Q"))
.append("\\E").toString()
}
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/Quantifier.kt
+58
View File
@@ -0,0 +1,58 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
import kotlin.IllegalArgumentException
/**
* Represents RE quantifier; contains two fields responsible for min and max number of repetitions.
* -1 as a maximum number of repetition represents infinity(i.e. +,*).
*/
internal class Quantifier(val min: Int, val max: Int = min) : SpecialToken() {
init {
if (min < 0 || max < -1) {
throw IllegalArgumentException("Incorrect quantifier value: $this")
}
}
override fun toString() = "{$min, ${if (max == INF) "" else max}}"
override val type: Type = SpecialToken.Type.QUANTIFIER
companion object {
val starQuantifier = Quantifier(0, -1)
val plusQuantifier = Quantifier(1, -1)
val altQuantifier = Quantifier(0, 1)
val INF = -1
fun fromLexerToken(token: Int) = when(token) {
Lexer.QUANT_STAR, Lexer.QUANT_STAR_P, Lexer.QUANT_STAR_R -> starQuantifier
Lexer.QUANT_ALT, Lexer.QUANT_ALT_P, Lexer.QUANT_ALT_R -> altQuantifier
Lexer.QUANT_PLUS, Lexer.QUANT_PLUS_P, Lexer.QUANT_PLUS_R -> plusQuantifier
else -> throw IllegalArgumentException("Unknown quantifier token: $token")
}
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/AbstractSet.kt
+160
View File
@@ -0,0 +1,160 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
import kotlin.AssertionError
/** Basic class for sets which have no complex next node handling. */
internal abstract class SimpleSet : AbstractSet {
override var next: AbstractSet = dummyNext
constructor()
constructor(type : Int): super(type)
}
/**
* Basic class for nodes, representing given regular expression.
* Note: (Almost) All the classes representing nodes has 'set' suffix.
*/
internal abstract class AbstractSet(val type: Int = 0) {
companion object {
const val TYPE_LEAF = 1 shl 0
const val TYPE_FSET = 1 shl 1
const val TYPE_QUANT = 1 shl 3
@Suppress("DEPRECATION")
const val TYPE_DOTSET = 0x80000000.toInt() or '.'.toInt()
val dummyNext = object : AbstractSet() {
override var next: AbstractSet
get() = throw AssertionError("This method is not expected to be called.")
@Suppress("UNUSED_PARAMETER")
set(value) {}
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl) =
throw AssertionError("This method is not expected to be called.")
override fun hasConsumed(matchResult: MatchResultImpl): Boolean =
throw AssertionError("This method is not expected to be called.")
override fun processSecondPassInternal(): AbstractSet = this
override fun processSecondPass(): AbstractSet = this
}
}
var secondPassVisited = false
abstract var next: AbstractSet
protected open val name: String
get() = ""
/**
* Checks if this node matches in given position and recursively call
* next node matches on positive self match. Returns positive integer if
* entire match succeed, negative otherwise.
* @param startIndex - string index to start from.
* @param testString - input string.
* @param matchResult - MatchResult to sore result into.
* @return -1 if match fails or n > 0;
*/
abstract fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int
/**
* Attempts to apply pattern starting from this set/startIndex; returns
* index this search was started from, if value is negative, this means that
* this search didn't succeed, additional information could be obtained via
* matchResult.
*
* Note: this is default implementation for find method, it's based on
* matches, subclasses do not have to override find method unless
* more effective find method exists for a particular node type
* (sequence, i.e. substring, for example). Same applies for find back
* method.
*
* @param startIndex - starting index.
* @param testString - string to search in.
* @param matchResult - result of the match.
* @return last searched index.
*/
open fun find(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int
= (startIndex..testString.length).firstOrNull { index -> matches(index, testString, matchResult) >= 0 }
?: -1
/**
* @param leftLimit - an index, to finish search back (left limit).
* @param rightLimit - an index to start search from (right limit).
* @param testString - test string.
* @param matchResult - match result.
* @return an index to start back search next time if this search fails(new left bound);
* if this search fails the value is negative.
*/
open fun findBack(leftLimit: Int, rightLimit: Int, testString: CharSequence, matchResult: MatchResultImpl): Int
= (rightLimit downTo leftLimit).firstOrNull { index -> matches(index, testString, matchResult) >= 0 }
?: -1
/**
* Returns true, if this node has consumed any characters during
* positive match attempt, for example node representing character always
* consumes one character if it matches. If particular node matches
* empty sting this method will return false.
*
* @param matchResult - match result;
* @return true if the node consumes any character and false otherwise.
*/
abstract fun hasConsumed(matchResult: MatchResultImpl): Boolean
/**
* Returns true if the given node intersects with this one, false otherwise.
* This method is being used for quantifiers construction, lets consider the
* following regular expression (a|b)*ccc. (a|b) does not intersects with "ccc"
* and thus can be quantified greedily (w/o kickbacks), like *+ instead of *.
* @param set - A node the intersection is checked for. Usually a previous node.
* @return true if the given node intersects with this one, false otherwise.
*/
open fun first(set: AbstractSet): Boolean = true
/**
* This method is used for replacement backreferenced sets.
*
* @return null if current node need not to be replaced,
* [JointSet] which is replacement of current node otherwise.
*/
open fun processBackRefReplacement(): JointSet? {
return null
}
/**
* This method performs the second pass without checking if it's already performed or not.
*/
protected open fun processSecondPassInternal(): AbstractSet {
if (!next.secondPassVisited) {
this.next = next.processSecondPass()
}
return processBackRefReplacement() ?: this
}
/**
* This method is used for traversing nodes after the first stage of compilation.
*/
open fun processSecondPass(): AbstractSet {
secondPassVisited = true
return processSecondPassInternal()
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/AtomicJointSet.kt
+50
View File
@@ -0,0 +1,50 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* This class represent atomic group (?>X), once X matches, this match become unchangeable till the end of the match.
*/
open internal class AtomicJointSet(children: List<AbstractSet>, fSet: FSet) : NonCapturingJointSet(children, fSet) {
/** Returns startIndex+shift, the next position to match */
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
val start = matchResult.getConsumed(groupIndex)
matchResult.setConsumed(groupIndex, startIndex)
children.forEach {
val shift = it.matches(startIndex, testString, matchResult)
if (shift >= 0) {
// AtomicFset always returns true, but saves the index to run this next.match() from;
return next.matches((fSet as AtomicFSet).index, testString, matchResult)
}
}
matchResult.setConsumed(groupIndex, start)
return -1
}
override val name: String
get() = "AtomicJointSet"
override var next: AbstractSet = dummyNext
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/BackReferenceSet.kt
+97
View File
@@ -0,0 +1,97 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Back reference node;
*/
open internal class BackReferenceSet(val referencedGroup: Int, val consCounter: Int, val ignoreCase: Boolean = false)
: SimpleSet() {
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
val groupValue = getReferencedGroupValue(matchResult)
if (groupValue == null || startIndex + groupValue.length > testString.length) {
return -1
}
if (testString.startsWith(groupValue, startIndex, ignoreCase)) {
matchResult.setConsumed(consCounter, groupValue.length)
return next.matches(startIndex + groupValue.length, testString, matchResult)
}
return -1
}
override fun find(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
val groupValue = getReferencedGroupValue(matchResult)
if (groupValue == null || startIndex + groupValue.length > testString.length) {
return -1
}
var index = startIndex
while (index <= testString.length) {
index = testString.indexOf(groupValue, index, ignoreCase)
if (index < 0) {
return -1
}
if (index < testString.length
&& next.matches(index + groupValue.length, testString, matchResult) >=0) {
return index
}
index++
}
return -1
}
override fun findBack(leftLimit: Int, rightLimit: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
val groupValue = getReferencedGroupValue(matchResult)
if (groupValue == null || leftLimit + groupValue.length > rightLimit) {
return -1
}
var index = rightLimit
while (index >= leftLimit) {
index = testString.lastIndexOf(groupValue, index, ignoreCase)
if (index < 0) {
return -1
}
if (index >= 0 && next.matches(index + groupValue.length, testString, matchResult) >= 0) {
return index
}
index--
}
return -1
}
protected fun getReferencedGroupValue(matchResult: MatchResultImpl) = matchResult.group(referencedGroup)
override val name: String
get() = "back reference: $referencedGroup"
override fun hasConsumed(matchResult: MatchResultImpl): Boolean {
val result = matchResult.getConsumed(consCounter) != 0
matchResult.setConsumed(consCounter, -1)
return result
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/CharSet.kt
+88
View File
@@ -0,0 +1,88 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Represents node accepting single character.
*/
open internal class CharSet(char: Char, val ignoreCase: Boolean = false) : LeafSet() {
// We use only low case characters when working in case insensitive mode.
val char: Char = if (ignoreCase) char.lowercaseChar() else char
// Overrides =======================================================================================================
override fun accepts(startIndex: Int, testString: CharSequence): Int {
if (ignoreCase) {
return if (this.char == testString[startIndex].lowercaseChar()) 1 else -1
} else {
return if (this.char == testString[startIndex]) 1 else -1
}
}
override fun find(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
var index = startIndex
while (index < testString.length) {
index = testString.indexOf(char, index, ignoreCase)
if (index < 0) {
return -1
}
if (next.matches(index + charCount, testString, matchResult) >= 0) {
return index
}
index++
}
return -1
}
override fun findBack(leftLimit: Int, rightLimit: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
var index = rightLimit
while (index >= leftLimit) {
index = testString.lastIndexOf(char, index, ignoreCase)
if (index < 0) {
return -1
}
if (next.matches(index + charCount, testString, matchResult) >= 0) {
return index
}
index--
}
return -1
}
override val name: String
get()= char.toString()
override fun first(set: AbstractSet): Boolean {
if (ignoreCase) {
return super.first(set)
}
return when (set) {
is CharSet -> set.char == char
is RangeSet -> set.accepts(0, char.toString()) > 0
is SupplementaryCharSet -> false
is SupplementaryRangeSet -> set.contains(char)
else -> true
}
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/CompositeRangeSet.kt
+122
View File
@@ -0,0 +1,122 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/*
*
* Portions, Copyright © 1991-2005 Unicode, Inc. The following applies to Unicode.
*
* COPYRIGHT AND PERMISSION NOTICE
*
* Copyright © 1991-2005 Unicode, Inc. All rights reserved. Distributed under
* the Terms of Use in http://www.unicode.org/copyright.html. Permission is
* hereby granted, free of charge, to any person obtaining a copy of the
* Unicode data files and any associated documentation (the "Data Files")
* or Unicode software and any associated documentation (the "Software")
* to deal in the Data Files or Software without restriction, including without
* limitation the rights to use, copy, modify, merge, publish, distribute,
* and/or sell copies of the Data Files or Software, and to permit persons
* to whom the Data Files or Software are furnished to do so, provided that
* (a) the above copyright notice(s) and this permission notice appear with
* all copies of the Data Files or Software, (b) both the above copyright
* notice(s) and this permission notice appear in associated documentation,
* and (c) there is clear notice in each modified Data File or in the Software
* as well as in the documentation associated with the Data File(s) or Software
* that the data or software has been modified.
* THE DATA FILES AND SOFTWARE ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY
* KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT
* OF THIRD PARTY RIGHTS. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR HOLDERS
* INCLUDED IN THIS NOTICE BE LIABLE FOR ANY CLAIM, OR ANY SPECIAL INDIRECT
* OR CONSEQUENTIAL DAMAGES, OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS
* OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR
* OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
* PERFORMANCE OF THE DATA FILES OR SOFTWARE.
*
* Except as contained in this notice, the name of a copyright holder shall
* not be used in advertising or otherwise to promote the sale, use or other
* dealings in these Data Files or Software without prior written
* authorization of the copyright holder.
*
* 2. Additional terms from the Database:
*
* Copyright © 1995-1999 Unicode, Inc. All Rights reserved.
*
* Disclaimer
*
* The Unicode Character Database is provided as is by Unicode, Inc.
* No claims are made as to fitness for any particular purpose. No warranties
* of any kind are expressed or implied. The recipient agrees to determine
* applicability of information provided. If this file has been purchased
* on magnetic or optical media from Unicode, Inc., the sole remedy for any claim
* will be exchange of defective media within 90 days of receipt. This disclaimer
* is applicable for all other data files accompanying the Unicode Character Database,
* some of which have been compiled by the Unicode Consortium, and some of which
* have been supplied by other sources.
*
* Limitations on Rights to Redistribute This Data
*
* Recipient is granted the right to make copies in any form for internal
* distribution and to freely use the information supplied in the creation of
* products supporting the UnicodeTM Standard. The files in
* the Unicode Character Database can be redistributed to third parties or other
* organizations (whether for profit or not) as long as this notice and the disclaimer
* notice are retained. Information can be extracted from these files and used
* in documentation or programs, as long as there is an accompanying notice
* indicating the source.
*/
package kotlin.text.regex
/**
* This class is used to split the range that contains surrogate characters into two ranges:
* the first consisting of these surrogate characters and the second consisting of all others characters
* from the parent range. This class represents the parent range split in such a manner.
*/
internal class CompositeRangeSet(/* range without surrogates */ val withoutSurrogates: AbstractSet,
/* range containing surrogates only */ val surrogates: AbstractSet) : SimpleSet() {
override var next: AbstractSet = dummyNext
get() = field
set(next) {
field = next
surrogates.next = next
withoutSurrogates.next = next
}
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
var result = withoutSurrogates.matches(startIndex, testString, matchResult)
if (result < 0) {
result = surrogates.matches(startIndex, testString, matchResult)
}
return result
}
override val name: String
get() = "CompositeRangeSet: " + " <nonsurrogate> " + withoutSurrogates + " <surrogate> " + surrogates
override fun hasConsumed(matchResult: MatchResultImpl): Boolean = true
override fun first(set: AbstractSet): Boolean {
return true
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/DecomposedCharSet.kt
+155
View File
@@ -0,0 +1,155 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/** Represents canonical decomposition of Unicode character. Is used when CANON_EQ flag of Pattern class is specified. */
open internal class DecomposedCharSet(
/** Decomposition of the Unicode codepoint */
private val decomposedChar: IntArray,
/** Length of useful part of decomposedChar decomposedCharLength <= decomposedChar.length */
private val decomposedCharLength: Int) : SimpleSet() {
/** Contains information about number of chars that were read for a codepoint last time */
private var readCharsForCodePoint = 1
/** UTF-16 encoding of decomposedChar */
private val decomposedCharUTF16: String by lazy {
val strBuff = StringBuilder()
for (i in 0..decomposedCharLength - 1) {
strBuff.append(Char.toChars(decomposedChar[i]))
}
return@lazy strBuff.toString()
}
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
var strIndex = startIndex
val rightBound = testString.length
if (strIndex >= rightBound) {
return -1
}
// We read testString and decompose it gradually to compare with this decomposedChar at position strIndex
var curChar = codePointAt(strIndex, testString, rightBound)
strIndex += readCharsForCodePoint
var readCodePoints = 0
var i = 0
// All decompositions have length that is less or equal Lexer.MAX_DECOMPOSITION_LENGTH
var decomposedCodePoint: IntArray = IntArray(Lexer.MAX_DECOMPOSITION_LENGTH)
readCodePoints += decomposeCodePoint(curChar, decomposedCodePoint, readCodePoints)
if (strIndex < rightBound) {
curChar = codePointAt(strIndex, testString, rightBound)
// Read testString until we met a decomposed char boundary and decompose obtained portion of testString.
while (readCodePoints < Lexer.MAX_DECOMPOSITION_LENGTH && !Lexer.isDecomposedCharBoundary(curChar)) {
if (!Lexer.hasDecompositionNonNullCanClass(curChar)) {
decomposedCodePoint[readCodePoints++] = curChar
} else {
/*
* A few codepoints have decompositions and non null canonical classes, we have to take them into
* consideration, but general rule is: if canonical class != 0 then no decomposition
*/
readCodePoints += decomposeCodePoint(curChar, decomposedCodePoint, readCodePoints)
}
strIndex += readCharsForCodePoint
if (strIndex < rightBound) {
curChar = codePointAt(strIndex, testString, rightBound)
} else {
break
}
}
}
// Some optimization since length of decomposed char is <= 3 usually
when (readCodePoints) {
0, 1, 2 -> {}
3 -> {
var i1 = Lexer.getCanonicalClass(decomposedCodePoint[1])
val i2 = Lexer.getCanonicalClass(decomposedCodePoint[2])
if (i2 != 0 && i1 > i2) {
i1 = decomposedCodePoint[1]
decomposedCodePoint[1] = decomposedCodePoint[2]
decomposedCodePoint[2] = i1
}
}
else -> decomposedCodePoint = Lexer.getCanonicalOrder(decomposedCodePoint, readCodePoints)
}
// Compare decomposedChar with decomposed char that was just read from testString
if (readCodePoints != decomposedCharLength) {
return -1
}
if ((0 until readCodePoints).firstOrNull { decomposedCodePoint[i] != decomposedChar[i] } != null) {
return -1
}
return next.matches(strIndex, testString, matchResult)
}
override val name: String
get() = "decomposed char: $decomposedChar"
/** Reads Unicode codepoint from [testString] starting from [strIndex] until [rightBound]. */
fun codePointAt(strIndex: Int, testString: CharSequence, rightBound: Int): Int {
var index = strIndex
// We store information about number of codepoints we read at variable readCharsForCodePoint.
val curChar: Int
readCharsForCodePoint = 1
if (index < rightBound - 1) {
val high = testString[index++]
val low = testString[index]
if (Char.isSurrogatePair(high, low)) {
curChar = Char.toCodePoint(high, low)
readCharsForCodePoint = 2
} else {
@Suppress("DEPRECATION")
curChar = high.toInt()
}
} else {
@Suppress("DEPRECATION")
curChar = testString[index].toInt()
}
return curChar
}
override fun first(set: AbstractSet): Boolean {
return if (set is DecomposedCharSet)
set.decomposedChar.contentEquals(decomposedChar)
else
true
}
override fun hasConsumed(matchResult: MatchResultImpl): Boolean = true
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/DotQuantifierSet.kt
+103
View File
@@ -0,0 +1,103 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Special node for ".*" construction.
* The main idea here is to find line terminator and try to find the rest of the construction from this point.
*/
// TODO: Add optimized implementation for '.+' case
internal class DotQuantifierSet(
innerSet: AbstractSet,
next: AbstractSet,
type: Int,
val lineTerminator: AbstractLineTerminator,
val matchLineTerminator: Boolean = false
) : QuantifierSet(innerSet, next, type) {
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
val rightBound = testString.length
val startSearch = if (matchLineTerminator) rightBound else testString.findLineTerminator(startIndex, rightBound)
if (startSearch <= startIndex) {
return if (type.toChar() == '+') {
-1
} else {
next.matches(startIndex, testString, matchResult)
}
}
val result = next.findBack(startIndex, startSearch, testString, matchResult)
if (type.toChar() == '+' && result == startIndex) {
return -1
}
return result
}
override fun find(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
val rightBound = testString.length
if (matchLineTerminator) {
val foundIndex = next.findBack(startIndex, rightBound, testString, matchResult)
if (foundIndex >= 0 && !(type.toChar() == '+' && foundIndex == startIndex)) {
return startIndex
} else {
return -1
}
} else {
// 1. find first occurrence of the searched pattern.
var nextFound = next.find(startIndex, testString, matchResult)
if (nextFound < 0) {
return -1
}
// 2. Check if we have other occurrences till the end of line (because .* is greedy and we need the last one).
val nextFoundLast = next.findBack(nextFound,
testString.findLineTerminator(nextFound, rightBound),
testString, matchResult)
nextFound = maxOf(nextFound, nextFoundLast)
// 3. Find the left boundary of this search.
val leftBound = findBackLineTerminator(startIndex, nextFound, testString)
if (type.toChar() == '+' && leftBound + 1 == nextFound) {
return -1
}
return leftBound + 1
}
}
/**
* Find the first line terminator between [from] (inclusive) and [to] (exclusive) indices.
* Returns [to] if no terminator found.
*/
private fun CharSequence.findLineTerminator(from: Int, to: Int): Int =
(from until to).firstOrNull { lineTerminator.isLineTerminator(this[it]) } ?: to
/**
* Find the first line terminator between [from] (inclusive) and [to] (exclusive) indices.
* Returns [from - 1] if no terminator found.
*/
private fun findBackLineTerminator(from: Int, to: Int, testString: CharSequence): Int =
(from until to).lastOrNull { lineTerminator.isLineTerminator(testString[it]) } ?: from - 1
override val name: String
get() = ".*"
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/DotSet.kt
+58
View File
@@ -0,0 +1,58 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Node accepting any character except line terminators.
*/
internal class DotSet(val lt: AbstractLineTerminator, val matchLineTerminator: Boolean)
: SimpleSet(AbstractSet.TYPE_DOTSET) {
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
val rightBound = testString.length
if (startIndex >= rightBound) {
return -1
}
val high = testString[startIndex]
if (high.isHighSurrogate() && startIndex + 2 <= rightBound) {
val low = testString[startIndex + 1]
if (Char.isSurrogatePair(high, low)) {
if (!matchLineTerminator && lt.isLineTerminator(Char.toCodePoint(high, low))) {
return -1
} else {
return next.matches(startIndex + 2, testString, matchResult)
}
}
}
if (!matchLineTerminator && lt.isLineTerminator(high)) {
return -1
} else {
return next.matches(startIndex + 1, testString, matchResult)
}
}
override fun hasConsumed(matchResult: MatchResultImpl): Boolean = true
override val name: String
get() = "."
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/EOISet.kt
+40
View File
@@ -0,0 +1,40 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Represents end of input '\z', i.e. matches only character after the last one;
*/
internal class EOISet : SimpleSet() {
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
if (startIndex < testString.length) {
return -1
}
return next.matches(startIndex, testString, matchResult)
}
override fun hasConsumed(matchResult: MatchResultImpl): Boolean = false
override val name: String
get() = "EOI"
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/EOLSet.kt
+56
View File
@@ -0,0 +1,56 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Represents a node for a '$' sign.
* Note: In Kotlin we use only the "anchoring bounds" mode when "$" matches the end of a match region.
* See: http://docs.oracle.com/javase/8/docs/api/java/util/regex/Matcher.html#useAnchoringBounds-boolean-
*/
internal class EOLSet(val consCounter: Int, val lt: AbstractLineTerminator, val multiline: Boolean = false) : SimpleSet() {
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
val rightBound = testString.length
val remainingChars = rightBound - startIndex
when {
startIndex >= rightBound ||
remainingChars == 1 && lt.isLineTerminator(testString[startIndex]) ||
remainingChars == 2 && lt.isLineTerminatorPair(testString[startIndex], testString[startIndex+1]) ||
multiline && lt.isLineTerminator(testString[startIndex]) -> {
matchResult.setConsumed(consCounter, 0)
return next.matches(startIndex, testString, matchResult)
}
}
return -1
}
override fun hasConsumed(matchResult: MatchResultImpl): Boolean {
val result = matchResult.getConsumed(consCounter) != 0
matchResult.setConsumed(consCounter, -1)
return result
}
override val name: String
get()= "<EOL>"
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/EmptySet.kt
+72
View File
@@ -0,0 +1,72 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Valid constant zero character match.
*/
internal class EmptySet(override var next: AbstractSet) : LeafSet() {
override val charCount = 0
override fun accepts(startIndex: Int, testString: CharSequence): Int = 0
override fun find(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
for (index in startIndex..testString.length) {
if (index < testString.length) {
if (testString[index].isLowSurrogate() &&
index > 0 && testString[index - 1].isHighSurrogate()) {
continue
}
}
if (next.matches(index, testString, matchResult) >= 0) {
return index
}
}
return -1
}
override fun findBack(leftLimit: Int, rightLimit: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
for (index in rightLimit downTo leftLimit) {
if (index < testString.length) {
if (testString[index].isLowSurrogate() &&
index > 0 && testString[index - 1].isHighSurrogate()) {
continue
}
}
if (next.matches(index, testString, matchResult) >= 0) {
return index
}
}
return -1
}
override val name: String
get()= "<Empty set>"
override fun hasConsumed(matchResult: MatchResultImpl): Boolean {
return false
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/FSets.kt
+159
View File
@@ -0,0 +1,159 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* The node which marks end of the particular group.
*/
open internal class FSet(val groupIndex: Int) : SimpleSet() {
var isBackReferenced = false
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
val oldEnd = matchResult.getEnd(groupIndex)
matchResult.setEnd(groupIndex, startIndex)
val shift = next.matches(startIndex, testString, matchResult)
if (shift < 0) {
matchResult.setEnd(groupIndex, oldEnd)
}
return shift
}
override fun hasConsumed(matchResult: MatchResultImpl): Boolean = false
override val name: String
get() = "fSet"
override fun processSecondPass(): FSet {
val result = super.processSecondPass()
assert(result == this)
return this
}
/**
* Marks the end of the particular group and not take into account possible
* kickbacks (required for atomic groups, for instance)
*/
internal class PossessiveFSet : SimpleSet() {
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
return startIndex
}
override fun hasConsumed(matchResult: MatchResultImpl): Boolean {
return false
}
override val name: String
get() = "possessiveFSet"
}
companion object {
val possessiveFSet = PossessiveFSet()
}
}
/**
* Special construction which marks end of pattern.
*/
internal class FinalSet : FSet(0) {
override fun matches(startIndex: Int, testString: CharSequence,
matchResult: MatchResultImpl): Int {
if (matchResult.mode == Regex.Mode.FIND || startIndex == testString.length) {
matchResult.setEnd(0, startIndex)
return startIndex
}
return -1
}
override val name: String
get() = "FinalSet"
}
/**
* Non-capturing group closing node.
*/
internal class NonCapFSet(groupIndex: Int) : FSet(groupIndex) {
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
matchResult.setConsumed(groupIndex, startIndex - matchResult.getConsumed(groupIndex))
return next.matches(startIndex, testString, matchResult)
}
override val name: String
get() = "NonCapFSet"
override fun hasConsumed(matchResult: MatchResultImpl): Boolean {
return false
}
}
/**
* LookAhead FSet, always returns true
*/
internal class AheadFSet : FSet(-1) {
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
return startIndex
}
override val name: String
get() = "AheadFSet"
}
/**
* FSet for lookbehind constructs. Checks if string index saved by corresponding
* jointSet in "consumers" equals to current index and return current string
* index, return -1 otherwise.
*/
internal class BehindFSet(groupIndex: Int) : FSet(groupIndex) {
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
val rightBound = matchResult.getConsumed(groupIndex)
return if (rightBound == startIndex) startIndex else -1
}
override val name: String
get() = "BehindFSet"
}
/**
* Represents an end of an atomic group.
*/
internal class AtomicFSet(groupIndex: Int) : FSet(groupIndex) {
var index: Int = 0
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
matchResult.setConsumed(groupIndex, startIndex - matchResult.getConsumed(groupIndex))
index = startIndex
return startIndex
}
override val name: String
get() = "AtomicFSet"
override fun hasConsumed(matchResult: MatchResultImpl): Boolean {
return false
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/GroupQuantifierSet.kt
+92
View File
@@ -0,0 +1,92 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Default quantifier over groups, in fact this type of quantifier is
* generally used for constructions we cant identify number of characters they
* consume.
*/
open internal class GroupQuantifierSet(
val quantifier: Quantifier,
innerSet: AbstractSet,
next: AbstractSet,
type: Int,
val groupQuantifierIndex: Int // It's used to remember a number of the innerSet occurrences during the recursive search.
) : QuantifierSet(innerSet, next, type) {
val max: Int get() = quantifier.max
val min: Int get() = quantifier.min
// We call innerSet.matches here, if it succeeds it call next.matches where next is this QuantifierSet.
// So we have a recursive searching procedure.
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
var enterCount = matchResult.enterCounters[groupQuantifierIndex]
fun matchNext(): Int {
matchResult.enterCounters[groupQuantifierIndex] = 0
val result = next.matches(startIndex, testString, matchResult)
matchResult.enterCounters[groupQuantifierIndex] = enterCount
return result
}
if (!innerSet.hasConsumed(matchResult)) {
return matchNext()
}
// Fast case: '*' or {0, } - no need to count occurrences.
if (min == 0 && max == Quantifier.INF) {
val nextIndex = innerSet.matches(startIndex, testString, matchResult)
return if (nextIndex < 0) {
matchNext()
} else {
nextIndex
}
}
// can't go inner set;
if (max != Quantifier.INF && enterCount >= max) {
return matchNext()
}
// go inner set;
matchResult.enterCounters[groupQuantifierIndex] = ++enterCount
val nextIndex = innerSet.matches(startIndex, testString, matchResult)
return if (nextIndex < 0) {
matchResult.enterCounters[groupQuantifierIndex] = --enterCount
if (enterCount >= min) {
matchNext()
} else {
-1
}
} else {
nextIndex
}
}
override val name: String
get() = quantifier.toString()
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/HangulDecomposedCharSet.kt
+180
View File
@@ -0,0 +1,180 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
@file:Suppress("DEPRECATION") // Char.toInt()
package kotlin.text.regex
import kotlin.text.*
/**
* Represents canonical decomposition of Hangul syllable. Is used when
* CANON_EQ flag of Pattern class is specified.
*/
internal class HangulDecomposedCharSet(
/**
* Decomposed Hangul syllable.
*/
private val decomposedChar: CharArray,
/**
* Length of useful part of decomposedChar
* decomposedCharLength <= decomposedChar.length
*/
private val decomposedCharLength: Int) : SimpleSet() {
/**
* String representing syllable
*/
private val decomposedCharUTF16: String by lazy {
decomposedChar.concatToString(0, decomposedChar.size)
}
override val name: String
get() = "decomposed Hangul syllable: $decomposedCharUTF16"
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
var index = startIndex
/*
* All decompositions for Hangul syllables have length that
* is less or equal Lexer.MAX_DECOMPOSITION_LENGTH
*/
val rightBound = testString.length
var SyllIndex = 0
val decompSyllable = IntArray(Lexer
.MAX_HANGUL_DECOMPOSITION_LENGTH)
val decompCurSymb: IntArray?
var curSymb: Char
/*
* For details about Hangul composition and decomposition see
* http://www.unicode.org/versions/Unicode4.0.0/ch03.pdf
* "3.12 Conjoining Jamo Behavior"
*/
var LIndex: Int
var VIndex = -1
var TIndex = -1
if (index >= rightBound) {
return -1
}
curSymb = testString[index++]
decompCurSymb = Lexer.getHangulDecomposition(curSymb.toInt())
if (decompCurSymb == null) {
/*
* We deal with ordinary letter or sequence of jamos
* at index at testString.
*/
decompSyllable[SyllIndex++] = curSymb.toInt()
LIndex = curSymb.toInt() - Lexer.LBase
if (LIndex < 0 || LIndex >= Lexer.LCount) {
/*
* Ordinary letter, that doesn't match this
*/
return -1
}
if (index < rightBound) {
curSymb = testString[index]
VIndex = curSymb.toInt() - Lexer.VBase
}
if (VIndex < 0 || VIndex >= Lexer.VCount) {
/*
* Single L jamo doesn't compose Hangul syllable,
* so doesn't match
*/
return -1
}
index++
decompSyllable[SyllIndex++] = curSymb.toInt()
if (index < rightBound) {
curSymb = testString[index]
TIndex = curSymb.toInt() - Lexer.TBase
}
if (TIndex < 0 || TIndex >= Lexer.TCount) {
/*
* We deal with LV syllable at testString, so
* compare it to this
*/
return if (decomposedCharLength == 2
&& decompSyllable[0] == decomposedChar[0].toInt()
&& decompSyllable[1] == decomposedChar[1].toInt())
next.matches(index, testString, matchResult)
else
-1
}
index++
decompSyllable[@Suppress("UNUSED_CHANGED_VALUE")SyllIndex++] = curSymb.toInt()
/*
* We deal with LVT syllable at testString, so
* compare it to this
*/
return if (decomposedCharLength == 3
&& decompSyllable[0] == decomposedChar[0].toInt()
&& decompSyllable[1] == decomposedChar[1].toInt()
&& decompSyllable[2] == decomposedChar[2].toInt())
next.matches(index, testString, matchResult)
else
-1
} else {
/*
* We deal with Hangul syllable at index at testString.
* So we decomposed it to compare with this.
*/
var i = 0
if (decompCurSymb.size != decomposedCharLength) {
return -1
}
while (i < decomposedCharLength) {
if (decompCurSymb[i] != decomposedChar[i].toInt()) {
return -1
}
i++
}
return next.matches(index, testString, matchResult)
}
}
override fun first(set: AbstractSet): Boolean {
return if (set is HangulDecomposedCharSet)
set.decomposedCharUTF16 == decomposedCharUTF16
else
true
}
override fun hasConsumed(matchResult: MatchResultImpl): Boolean {
return true
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/JointSet.kt
+82
View File
@@ -0,0 +1,82 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Represents group, which is alternation of other subexpression.
* One should think about "group" in this model as JointSet opening group and corresponding FSet closing group.
*/
open internal class JointSet(children: List<AbstractSet>, fSet: FSet) : AbstractSet() {
protected var children: MutableList<AbstractSet> = mutableListOf<AbstractSet>().apply { addAll(children) }
var fSet: FSet = fSet
protected set
var groupIndex: Int = fSet.groupIndex
protected set
/**
* Returns startIndex+shift, the next position to match
*/
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
if (children.isEmpty()) {
return -1
}
val oldStart = matchResult.getStart(groupIndex)
matchResult.setStart(groupIndex, startIndex)
children.forEach {
val shift = it.matches(startIndex, testString, matchResult)
if (shift >= 0) {
return shift
}
}
matchResult.setStart(groupIndex, oldStart)
return -1
}
override var next: AbstractSet
get() = fSet.next
set(next) {
fSet.next = next
}
override val name: String
get() = "JointSet"
override fun first(set: AbstractSet): Boolean = children.any { it.first(set) }
override fun hasConsumed(matchResult: MatchResultImpl): Boolean {
return !(matchResult.getEnd(groupIndex) >= 0 && matchResult.getStart(groupIndex) == matchResult.getEnd(groupIndex))
}
override fun processSecondPassInternal(): AbstractSet {
val fSet = this.fSet
if (!fSet.secondPassVisited) {
val newFSet = fSet.processSecondPass()
assert(newFSet == fSet)
}
children.replaceAll { child -> if (!child.secondPassVisited) child.processSecondPass() else child }
return super.processSecondPassInternal()
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/LeafQuantifierSet.kt
+94
View File
@@ -0,0 +1,94 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
import kotlin.RuntimeException
/**
* Generalized greedy quantifier node over the leaf nodes.
* - a{n,m};
* - a* == a{0, <inf>};
* - a? == a{0, 1};
* - a+ == a{1, <inf>};
*/
open internal class LeafQuantifierSet(var quantifier: Quantifier,
innerSet: LeafSet,
next: AbstractSet,
type: Int
) : QuantifierSet(innerSet, next, type) {
val leaf: LeafSet get() = super.innerSet as LeafSet
val min: Int get() = quantifier.min
val max: Int get() = quantifier.max
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
var index = startIndex
var occurrences = 0
// Process first <min> occurrences of the sequence being looked for.
while (occurrences < min) {
if (index + leaf.charCount > testString.length) {
return -1
}
val shift = leaf.accepts(index, testString)
if (shift < 1) {
return -1
}
index += shift
occurrences++
}
// Process occurrences between min and max.
while ((max == Quantifier.INF || occurrences < max) && index + leaf.charCount <= testString.length) {
val shift = leaf.accepts(index, testString)
if (shift < 1) {
break
}
index += shift
occurrences++
}
// Roll back if the next node does't match the remaining string.
while (occurrences >= min) {
val shift = next.matches(index, testString, matchResult)
if (shift >= 0) {
return shift
}
index -= leaf.charCount
occurrences--
}
return -1
}
override val name: String
get() = quantifier.toString()
override var innerSet: AbstractSet
get() = super.innerSet
set(innerSet) {
if (innerSet !is LeafSet)
throw RuntimeException("Internal Error")
super.innerSet = innerSet
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/LeafSet.kt
+55
View File
@@ -0,0 +1,55 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Base class for nodes representing leaf tokens of the RE, those who consumes fixed number of characters.
*/
internal abstract class LeafSet : SimpleSet(AbstractSet.TYPE_LEAF) {
open val charCount = 1
/** Returns "shift", the number of accepted chars. Commonly internal function, but called by quantifiers. */
abstract fun accepts(startIndex: Int, testString: CharSequence): Int
/**
* Checks if we can enter this state and pass the control to the next one.
* Return positive value if match succeeds, negative otherwise.
*/
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
if (startIndex + charCount > testString.length) {
return -1
}
val shift = accepts(startIndex, testString) // TODO: may be move the check above in accept function.
if (shift < 0) {
return -1
}
return next.matches(startIndex + shift, testString, matchResult)
}
override fun hasConsumed(matchResult: MatchResultImpl): Boolean {
return true
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/LookAheadSets.kt
+64
View File
@@ -0,0 +1,64 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Positive lookahead node.
*/
internal class PositiveLookAheadSet(children: List<AbstractSet>, fSet: FSet) : LookAroundSet(children, fSet) {
/** Returns startIndex+shift, the next position to match */
override fun tryToMatch(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
children.forEach {
val shift = it.matches(startIndex, testString, matchResult)
if (shift >= 0) {
// PosLookaheadFset always returns true, position remains the same next.match() from;
return next.matches(startIndex, testString, matchResult)
}
}
return -1
}
override val name: String
get() = "PositiveLookaheadJointSet"
}
/**
* Negative look ahead node.
*/
internal class NegativeLookAheadSet(children: List<AbstractSet>, fSet: FSet) : LookAroundSet(children, fSet) {
/** Returns startIndex+shift, the next position to match */
override fun tryToMatch(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
children.forEach {
if (it.matches(startIndex, testString, matchResult) >= 0) {
return -1
}
}
return next.matches(startIndex, testString, matchResult)
}
override val name: String
get() = "NegativeLookaheadJointSet"
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/LookAroundSet.kt
+22
View File
@@ -0,0 +1,22 @@
/*
* Copyright 2010-2021 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
package kotlin.text.regex
/**
* Abstract class for lookahead and lookbehind nodes.
*/
internal abstract class LookAroundSet(children: List<AbstractSet>, fSet: FSet) : AtomicJointSet(children, fSet) {
protected abstract fun tryToMatch(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int
/** Returns startIndex+shift, the next position to match */
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
matchResult.saveState()
return tryToMatch(startIndex, testString, matchResult).also { if (it < 0) matchResult.rollbackState() }
}
override fun hasConsumed(matchResult: MatchResultImpl): Boolean = true
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/LookBehindSets.kt
+68
View File
@@ -0,0 +1,68 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Positive lookbehind node.
*/
internal class PositiveLookBehindSet(children: List<AbstractSet>, fSet: FSet) : LookAroundSet(children, fSet) {
/** Returns startIndex+shift, the next position to match */
override fun tryToMatch(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
matchResult.setConsumed(groupIndex, startIndex)
children.forEach {
if (it.findBack(0, startIndex, testString, matchResult) >= 0) {
matchResult.setConsumed(groupIndex, -1)
return next.matches(startIndex, testString, matchResult)
}
}
return -1
}
override val name: String
get() = "PositiveBehindJointSet"
}
/**
* Negative look behind node.
*/
internal class NegativeLookBehindSet(children: List<AbstractSet>, fSet: FSet) : LookAroundSet(children, fSet) {
/** Returns startIndex+shift, the next position to match */
override fun tryToMatch(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
matchResult.setConsumed(groupIndex, startIndex)
children.forEach {
val shift = it.findBack(0, startIndex, testString, matchResult)
if (shift >= 0) {
return -1
}
}
return next.matches(startIndex, testString, matchResult)
}
override val name: String
get() = "NegativeBehindJointSet"
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/NonCapturingJointSet.kt
+54
View File
@@ -0,0 +1,54 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Node representing non-capturing group
*/
open internal class NonCapturingJointSet(children: List<AbstractSet>, fSet: FSet) : JointSet(children, fSet) {
/**
* Returns startIndex+shift, the next position to match
*/
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
val start = matchResult.getConsumed(groupIndex)
matchResult.setConsumed(groupIndex, startIndex)
children.forEach {
val shift = it.matches(startIndex, testString, matchResult)
if (shift >= 0) {
return shift
}
}
matchResult.setConsumed(groupIndex, start)
return -1
}
override val name: String
get() = "NonCapturingJointSet"
override fun hasConsumed(matchResult: MatchResultImpl): Boolean {
return matchResult.getConsumed(groupIndex) != 0
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/PossessiveGroupQuantifierSet.kt
+57
View File
@@ -0,0 +1,57 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Possessive quantifier set over groups.
*/
internal class PossessiveGroupQuantifierSet(
quantifier: Quantifier,
innerSet: AbstractSet,
next: AbstractSet,
type: Int,
setCounter: Int
): GroupQuantifierSet(quantifier, innerSet, next, type, setCounter) {
init {
innerSet.next = FSet.possessiveFSet
}
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
var index = startIndex
var nextIndex: Int = innerSet.matches(index, testString, matchResult)
var occurrences = 0
while (nextIndex > index && (max == Quantifier.INF || occurrences < max)) {
occurrences++
index = nextIndex
nextIndex = innerSet.matches(index, testString, matchResult)
}
if (occurrences < quantifier.min) {
return -1
} else {
return next.matches(index, testString, matchResult)
}
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/PossessiveLeafQuantifierSet.kt
+66
View File
@@ -0,0 +1,66 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Possessive quantifier node over a leaf node.
* - a{n,m}+;
* - a*+ == a{0, <inf>}+;
* - a?+ == a{0, 1}+;
* - a++ == a{1, <inf>}+;
*/
internal class PossessiveLeafQuantifierSet(
quant: Quantifier,
innerSet: LeafSet,
next: AbstractSet, type: Int
) : LeafQuantifierSet(quant, innerSet, next, type) {
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
var index = startIndex
var occurrences = 0
while (occurrences < min) {
if (index + leaf.charCount > testString.length) {
return -1
}
val shift = leaf.accepts(index, testString)
if (shift < 1) {
return -1
}
index += shift
occurrences++
}
while ((max == Quantifier.INF || occurrences < max) && index + leaf.charCount <= testString.length) {
val shift = leaf.accepts(index, testString)
if (shift < 1) {
break
}
index += shift
occurrences++
}
return next.matches(index, testString, matchResult)
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/PreviousMatchSet.kt
+43
View File
@@ -0,0 +1,43 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Node representing previous match (\G).
*/
internal class PreviousMatchSet : SimpleSet() {
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
if (startIndex == matchResult.previousMatch) {
return next.matches(startIndex, testString, matchResult)
}
return -1
}
override fun hasConsumed(matchResult: MatchResultImpl): Boolean = false
override val name: String
get() = "PreviousMatchSet"
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/QuantifierSet.kt
+44
View File
@@ -0,0 +1,44 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Base class for quantifiers.
*/
internal abstract class QuantifierSet(open var innerSet: AbstractSet, override var next: AbstractSet, type: Int)
: SimpleSet(type) {
override fun first(set: AbstractSet): Boolean =
innerSet.first(set) || next.first(set)
override fun hasConsumed(matchResult: MatchResultImpl): Boolean = true
override fun processSecondPassInternal(): AbstractSet {
val innerSet = this.innerSet
if (innerSet.secondPassVisited) {
this.innerSet = innerSet.processSecondPass()
}
return super.processSecondPassInternal()
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/RangeSet.kt
+54
View File
@@ -0,0 +1,54 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Represents node accepting single character from the given char class.
*/
open internal class RangeSet(charClass: AbstractCharClass, val ignoreCase: Boolean = false) : LeafSet() {
val chars: AbstractCharClass = charClass.instance
override fun accepts(startIndex: Int, testString: CharSequence): Int {
if (ignoreCase) {
val char = testString[startIndex]
return if (chars.contains(char.uppercaseChar()) || chars.contains(char.lowercaseChar())) 1 else -1
} else {
return if (chars.contains(testString[startIndex])) 1 else -1
}
}
override val name: String
get() = "range:" + (if (chars.alt) "^ " else " ") + chars.toString()
override fun first(set: AbstractSet): Boolean {
@Suppress("DEPRECATION")
return when (set) {
is CharSet -> AbstractCharClass.intersects(chars, set.char.toInt())
is RangeSet -> AbstractCharClass.intersects(chars, set.chars)
is SupplementaryCharSet -> false
is SupplementaryRangeSet -> AbstractCharClass.intersects(chars, set.chars)
else -> true
}
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/ReluctantGroupQuantifierSet.kt
+78
View File
@@ -0,0 +1,78 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Reluctant version of the group quantifier set.
*/
internal class ReluctantGroupQuantifierSet(
quantifier: Quantifier,
innerSet: AbstractSet,
next: AbstractSet,
type: Int,
setCounter: Int
) : GroupQuantifierSet(quantifier, innerSet, next, type, setCounter) {
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
var enterCount = matchResult.enterCounters[groupQuantifierIndex]
fun matchNext(): Int {
matchResult.enterCounters[groupQuantifierIndex] = 0
val result = next.matches(startIndex, testString, matchResult)
matchResult.enterCounters[groupQuantifierIndex] = enterCount
return result
}
if (!innerSet.hasConsumed(matchResult)) {
return matchNext()
}
// Fast case: '*' or {0, } - no need to count occurrences.
if (min == 0 && max == Quantifier.INF) {
val res = next.matches(startIndex, testString, matchResult)
return if (res < 0) {
innerSet.matches(startIndex, testString, matchResult)
} else {
res
}
}
// can't go inner set;
if (max != Quantifier.INF && enterCount >= max) {
return matchNext()
}
return if (enterCount >= min) {
val nextIndex = matchNext()
if (nextIndex < 0) {
matchResult.enterCounters[groupQuantifierIndex] = ++enterCount
innerSet.matches(startIndex, testString, matchResult)
} else {
nextIndex
}
} else {
matchResult.enterCounters[groupQuantifierIndex] = ++enterCount
innerSet.matches(startIndex, testString, matchResult)
}
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/ReluctantLeafQuantifierSet.kt
+72
View File
@@ -0,0 +1,72 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Reluctant quantifier node over a leaf node.
* - a{n,m}?;
* - a*? == a{0, <inf>}?;
* - a?? == a{0, 1}?;
* - a+? == a{1, <inf>}?;
*/
internal class ReluctantLeafQuantifierSet(
quant: Quantifier,
innerSet: LeafSet,
next: AbstractSet,
type: Int
) : LeafQuantifierSet(quant, innerSet, next, type) {
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
var index = startIndex
var occurrences = 0
while (occurrences < min) {
if (index + leaf.charCount > testString.length) {
return -1
}
val shift = leaf.accepts(index, testString)
if (shift < 1) {
return -1
}
index += shift
occurrences++
}
do {
var shift = next.matches(index, testString, matchResult)
if (shift >= 0) {
return shift
}
if (index + leaf.charCount <= testString.length) {
shift = leaf.accepts(index, testString)
index += shift
occurrences++
}
} while (shift >= 1 && (max == Quantifier.INF || occurrences <= max))
return -1
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/SOLSet.kt
+50
View File
@@ -0,0 +1,50 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* A node representing a '^' sign.
* Note: In Kotlin we use only the "anchoring bounds" mode when "^" matches beginning of a match region.
* See: http://docs.oracle.com/javase/8/docs/api/java/util/regex/Matcher.html#useAnchoringBounds-boolean-
*/
internal class SOLSet(val lt: AbstractLineTerminator, val multiline: Boolean = false) : SimpleSet() {
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
if (!multiline) {
if (startIndex == 0) {
return next.matches(startIndex, testString, matchResult)
}
} else {
if (startIndex != testString.length
&& (startIndex == 0
|| lt.isAfterLineTerminator(testString[startIndex - 1], testString[startIndex]))) {
return next.matches(startIndex, testString, matchResult)
}
}
return -1
}
override fun hasConsumed(matchResult: MatchResultImpl): Boolean = false
override val name: String
get() = "^"
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/SequenceSet.kt
+105
View File
@@ -0,0 +1,105 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* This class represents nodes constructed with character sequences. For
* example, lets consider regular expression: ".*word.*". During regular
* expression compilation phase character sequence w-o-r-d, will be represented
* with single node for the entire word.
*/
open internal class SequenceSet(substring: CharSequence, val ignoreCase: Boolean = false) : LeafSet() {
/** Represents a character sequence used for matching/searching. */
protected val patternString: String = substring.toString()
override val name: String= "sequence: " + patternString
override val charCount = substring.length
// Overrides =======================================================================================================
/** Returns true if [index] points to a low surrogate following a high surrogate */
private fun isLowSurrogateOfSupplement(string: CharSequence, index: Int): Boolean =
index < string.length && string[index].isLowSurrogate() && index > 0 && string[index - 1].isHighSurrogate()
override fun accepts(startIndex: Int, testString: CharSequence): Int {
return if (testString.startsWith(patternString, startIndex, ignoreCase)
&& !isLowSurrogateOfSupplement(testString, startIndex)
&& !isLowSurrogateOfSupplement(testString, startIndex + patternString.length)) {
charCount
} else {
-1
}
}
override fun find(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
var index = startIndex
while (index < testString.length) {
index = testString.indexOf(patternString, index, ignoreCase)
if (index < 0) {
return -1
}
// Check if we have a supplementary code point at the beginning or at the end of the string.
if (!isLowSurrogateOfSupplement(testString, index)
&& !isLowSurrogateOfSupplement(testString, index + patternString.length)
&& next.matches(index + charCount, testString, matchResult) >= 0) {
return index
}
index++
}
return -1
}
override fun findBack(leftLimit: Int, rightLimit: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
var index = rightLimit
while (index >= leftLimit) {
index = testString.lastIndexOf(patternString, index, ignoreCase)
if (index < 0 || index < leftLimit) {
return -1
}
// Check if we have a supplementary code point at the beginning or at the end of the string.
if (!isLowSurrogateOfSupplement(testString, index)
&& !isLowSurrogateOfSupplement(testString, index + patternString.length)
&& next.matches(index + charCount, testString, matchResult) >= 0) {
return index
}
index--
}
return -1
}
override fun first(set: AbstractSet): Boolean {
if (ignoreCase) {
return super.first(set)
}
return when (set) {
is CharSet -> set.char == patternString[0]
is RangeSet -> set.accepts(0, patternString.substring(0, 1)) > 0
is SupplementaryRangeSet -> set.contains(patternString[0]) || patternString.length > 1 && set.contains(Char.toCodePoint(patternString[0], patternString[1]))
is SupplementaryCharSet -> if (patternString.length > 1) set.codePoint == Char.toCodePoint(patternString[0], patternString[1])
else false
else -> true
}
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/SingleSet.kt
+151
View File
@@ -0,0 +1,151 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Group node over subexpression without alternations.
*/
open internal class SingleSet(var kid: AbstractSet, fSet: FSet) : JointSet(listOf(), fSet) {
var backReferencedSet: BackReferencedSingleSet? = null
// Overrides (API) =================================================================================================
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
val start = matchResult.getStart(groupIndex)
matchResult.setStart(groupIndex, startIndex)
val shift = kid.matches(startIndex, testString, matchResult)
if (shift >= 0) {
return shift
}
matchResult.setStart(groupIndex, start)
return -1
}
override fun find(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
val res = kid.find(startIndex, testString, matchResult)
if (res >= 0)
matchResult.setStart(groupIndex, res)
return res
}
override fun findBack(leftLimit: Int, rightLimit: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
val res = kid.findBack(leftLimit, rightLimit, testString, matchResult)
if (res >= 0)
matchResult.setStart(groupIndex, res)
return res
}
override fun first(set: AbstractSet): Boolean = kid.first(set)
// Second pass processing ==========================================================================================
override fun processBackRefReplacement(): JointSet? {
/**
* We will store a reference to created BackReferencedSingleSet
* in [backReferencedSet] field. This is needed to process replacement
* of sets correctly since sometimes we cannot renew all references to
* detachable set in the current point of traverse. See
* QuantifierSet and AbstractSet processSecondPass() methods for
* more details.
*/
val result = BackReferencedSingleSet(this)
backReferencedSet = result
return result
}
override fun processSecondPassInternal(): AbstractSet {
fSet = fSet.processSecondPass()
kid = kid.processSecondPass()
return processBackRefReplacement() ?: this
}
/**
* This method is used for traversing nodes after the first stage of compilation.
*/
override fun processSecondPass(): AbstractSet {
if (secondPassVisited) {
if (fSet.isBackReferenced) {
assert(backReferencedSet != null) // secondPassVisited
return backReferencedSet!!
}
}
secondPassVisited = true
return processSecondPassInternal()
}
// Backreferenced version of the class =============================================================================
/**
* Group node over subexpression without alternations.
* This node is used if current group is referenced via a backreference.
*/
internal class BackReferencedSingleSet(node: SingleSet) : SingleSet(node.kid, node.fSet) {
/*
* This class is needed only for overwriting find() and findBack() methods of SingleSet class, which is being
* back referenced. The following example explains the need for such substitution:
*
* Let's consider the pattern ".*(.)\\1".
* Leading .* works as follows: finds line terminator and runs findBack from that point.
* `findBack` method in its turn (in contrast to matches) sets group boundaries on the back trace.
* Thus at the point we try to match back reference(\\1) groups are not yet set.
*
* To fix this problem we replace backreferenced groups with instances of this class,
* which will use matches instead of find; this will affect performance, but ensure correctness of the match.
*/
override fun find(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
for (index in startIndex..testString.length) {
val oldStart = matchResult.getStart(groupIndex)
matchResult.setStart(groupIndex, index)
val res = kid.matches(index, testString, matchResult)
if (res >= 0) {
return index
} else {
matchResult.setStart(groupIndex, oldStart)
}
}
return -1
}
override fun findBack(leftLimit: Int, rightLimit: Int,
testString: CharSequence, matchResult: MatchResultImpl): Int {
for (index in rightLimit downTo leftLimit) {
val oldStart = matchResult.getStart(groupIndex)
matchResult.setStart(groupIndex, index)
val res = kid.matches(index, testString, matchResult)
if (res >= 0) {
return index
} else {
matchResult.setStart(groupIndex, oldStart)
}
}
return -1
}
override fun processBackRefReplacement(): JointSet? = null
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/SupplementaryCharSet.kt
+108
View File
@@ -0,0 +1,108 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/*
*
* Portions, Copyright © 1991-2005 Unicode, Inc. The following applies to Unicode.
*
* COPYRIGHT AND PERMISSION NOTICE
*
* Copyright © 1991-2005 Unicode, Inc. All rights reserved. Distributed under
* the Terms of Use in http://www.unicode.org/copyright.html. Permission is
* hereby granted, free of charge, to any person obtaining a copy of the
* Unicode data files and any associated documentation (the "Data Files")
* or Unicode software and any associated documentation (the "Software")
* to deal in the Data Files or Software without restriction, including without
* limitation the rights to use, copy, modify, merge, publish, distribute,
* and/or sell copies of the Data Files or Software, and to permit persons
* to whom the Data Files or Software are furnished to do so, provided that
* (a) the above copyright notice(s) and this permission notice appear with
* all copies of the Data Files or Software, (b) both the above copyright
* notice(s) and this permission notice appear in associated documentation,
* and (c) there is clear notice in each modified Data File or in the Software
* as well as in the documentation associated with the Data File(s) or Software
* that the data or software has been modified.
* THE DATA FILES AND SOFTWARE ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY
* KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT
* OF THIRD PARTY RIGHTS. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR HOLDERS
* INCLUDED IN THIS NOTICE BE LIABLE FOR ANY CLAIM, OR ANY SPECIAL INDIRECT
* OR CONSEQUENTIAL DAMAGES, OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS
* OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR
* OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
* PERFORMANCE OF THE DATA FILES OR SOFTWARE.
*
* Except as contained in this notice, the name of a copyright holder shall
* not be used in advertising or otherwise to promote the sale, use or other
* dealings in these Data Files or Software without prior written
* authorization of the copyright holder.
*
* 2. Additional terms from the Database:
*
* Copyright © 1995-1999 Unicode, Inc. All Rights reserved.
*
* Disclaimer
*
* The Unicode Character Database is provided as is by Unicode, Inc.
* No claims are made as to fitness for any particular purpose. No warranties
* of any kind are expressed or implied. The recipient agrees to determine
* applicability of information provided. If this file has been purchased
* on magnetic or optical media from Unicode, Inc., the sole remedy for any claim
* will be exchange of defective media within 90 days of receipt. This disclaimer
* is applicable for all other data files accompanying the Unicode Character Database,
* some of which have been compiled by the Unicode Consortium, and some of which
* have been supplied by other sources.
*
* Limitations on Rights to Redistribute This Data
*
* Recipient is granted the right to make copies in any form for internal
* distribution and to freely use the information supplied in the creation of
* products supporting the UnicodeTM Standard. The files in
* the Unicode Character Database can be redistributed to third parties or other
* organizations (whether for profit or not) as long as this notice and the disclaimer
* notice are retained. Information can be extracted from these files and used
* in documentation or programs, as long as there is an accompanying notice
* indicating the source.
*/
package kotlin.text.regex
/**
* Represents node accepting single supplementary codepoint.
*/
internal class SupplementaryCharSet(val codePoint: Int, ignoreCase: Boolean)
: SequenceSet(Char.toChars(codePoint).concatToString(0, 2), ignoreCase) {
override val name: String
get() = patternString
override fun first(set: AbstractSet): Boolean {
return when (set) {
is SupplementaryCharSet -> set.codePoint == codePoint
is SupplementaryRangeSet -> set.contains(codePoint)
is CharSet,
is RangeSet -> false
else -> true
}
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/SupplementaryRangeSet.kt
+151
View File
@@ -0,0 +1,151 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/*
*
* Portions, Copyright © 1991-2005 Unicode, Inc. The following applies to Unicode.
*
* COPYRIGHT AND PERMISSION NOTICE
*
* Copyright © 1991-2005 Unicode, Inc. All rights reserved. Distributed under
* the Terms of Use in http://www.unicode.org/copyright.html. Permission is
* hereby granted, free of charge, to any person obtaining a copy of the
* Unicode data files and any associated documentation (the "Data Files")
* or Unicode software and any associated documentation (the "Software")
* to deal in the Data Files or Software without restriction, including without
* limitation the rights to use, copy, modify, merge, publish, distribute,
* and/or sell copies of the Data Files or Software, and to permit persons
* to whom the Data Files or Software are furnished to do so, provided that
* (a) the above copyright notice(s) and this permission notice appear with
* all copies of the Data Files or Software, (b) both the above copyright
* notice(s) and this permission notice appear in associated documentation,
* and (c) there is clear notice in each modified Data File or in the Software
* as well as in the documentation associated with the Data File(s) or Software
* that the data or software has been modified.
* THE DATA FILES AND SOFTWARE ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY
* KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT
* OF THIRD PARTY RIGHTS. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR HOLDERS
* INCLUDED IN THIS NOTICE BE LIABLE FOR ANY CLAIM, OR ANY SPECIAL INDIRECT
* OR CONSEQUENTIAL DAMAGES, OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS
* OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR
* OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
* PERFORMANCE OF THE DATA FILES OR SOFTWARE.
*
* Except as contained in this notice, the name of a copyright holder shall
* not be used in advertising or otherwise to promote the sale, use or other
* dealings in these Data Files or Software without prior written
* authorization of the copyright holder.
*
* 2. Additional terms from the Database:
*
* Copyright © 1995-1999 Unicode, Inc. All Rights reserved.
*
* Disclaimer
*
* The Unicode Character Database is provided as is by Unicode, Inc.
* No claims are made as to fitness for any particular purpose. No warranties
* of any kind are expressed or implied. The recipient agrees to determine
* applicability of information provided. If this file has been purchased
* on magnetic or optical media from Unicode, Inc., the sole remedy for any claim
* will be exchange of defective media within 90 days of receipt. This disclaimer
* is applicable for all other data files accompanying the Unicode Character Database,
* some of which have been compiled by the Unicode Consortium, and some of which
* have been supplied by other sources.
*
* Limitations on Rights to Redistribute This Data
*
* Recipient is granted the right to make copies in any form for internal
* distribution and to freely use the information supplied in the creation of
* products supporting the UnicodeTM Standard. The files in
* the Unicode Character Database can be redistributed to third parties or other
* organizations (whether for profit or not) as long as this notice and the disclaimer
* notice are retained. Information can be extracted from these files and used
* in documentation or programs, as long as there is an accompanying notice
* indicating the source.
*/
package kotlin.text.regex
/**
* Represents node accepting single character from the given char class.
* This character can be supplementary (2 chars needed to represent) or from
* basic multilingual pane (1 needed char to represent it).
*/
open internal class SupplementaryRangeSet(charClass: AbstractCharClass, val ignoreCase: Boolean = false): SimpleSet() {
val chars = charClass.instance
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
val rightBound = testString.length
if (startIndex >= rightBound) {
return -1
}
var index = startIndex
val high = testString[index++]
if (contains(high)) {
val result = next.matches(index, testString, matchResult)
if (result >= 0) return result
}
if (index < rightBound) {
val low = testString[index++]
if (Char.isSurrogatePair(high, low) && contains(Char.toCodePoint(high, low))) {
return next.matches(index, testString, matchResult)
}
}
return -1
}
fun contains(char: Char): Boolean {
if (ignoreCase) {
return chars.contains(char.uppercaseChar()) || chars.contains(char.lowercaseChar())
} else {
return chars.contains(char)
}
}
fun contains(char: Int): Boolean {
return chars.contains(char)
}
override val name: String
get() = "range:" + (if (chars.alt) "^ " else " ") + chars.toString()
override fun first(set: AbstractSet): Boolean {
@Suppress("DEPRECATION")
return when(set) {
is SupplementaryCharSet -> AbstractCharClass.intersects(chars, set.codePoint)
is CharSet -> AbstractCharClass.intersects(chars, set.char.toInt())
is SupplementaryRangeSet -> AbstractCharClass.intersects(chars, set.chars)
is RangeSet -> AbstractCharClass.intersects(chars, set.chars)
else -> true
}
}
override fun hasConsumed(matchResult: MatchResultImpl): Boolean = true
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/SurrogateCharSets.kt
+219
View File
@@ -0,0 +1,219 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/*
*
* Portions, Copyright © 1991-2005 Unicode, Inc. The following applies to Unicode.
*
* COPYRIGHT AND PERMISSION NOTICE
*
* Copyright © 1991-2005 Unicode, Inc. All rights reserved. Distributed under
* the Terms of Use in http://www.unicode.org/copyright.html. Permission is
* hereby granted, free of charge, to any person obtaining a copy of the
* Unicode data files and any associated documentation (the "Data Files")
* or Unicode software and any associated documentation (the "Software")
* to deal in the Data Files or Software without restriction, including without
* limitation the rights to use, copy, modify, merge, publish, distribute,
* and/or sell copies of the Data Files or Software, and to permit persons
* to whom the Data Files or Software are furnished to do so, provided that
* (a) the above copyright notice(s) and this permission notice appear with
* all copies of the Data Files or Software, (b) both the above copyright
* notice(s) and this permission notice appear in associated documentation,
* and (c) there is clear notice in each modified Data File or in the Software
* as well as in the documentation associated with the Data File(s) or Software
* that the data or software has been modified.
* THE DATA FILES AND SOFTWARE ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY
* KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT
* OF THIRD PARTY RIGHTS. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR HOLDERS
* INCLUDED IN THIS NOTICE BE LIABLE FOR ANY CLAIM, OR ANY SPECIAL INDIRECT
* OR CONSEQUENTIAL DAMAGES, OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS
* OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR
* OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
* PERFORMANCE OF THE DATA FILES OR SOFTWARE.
*
* Except as contained in this notice, the name of a copyright holder shall
* not be used in advertising or otherwise to promote the sale, use or other
* dealings in these Data Files or Software without prior written
* authorization of the copyright holder.
*
* 2. Additional terms from the Database:
*
* Copyright © 1995-1999 Unicode, Inc. All Rights reserved.
*
* Disclaimer
*
* The Unicode Character Database is provided as is by Unicode, Inc.
* No claims are made as to fitness for any particular purpose. No warranties
* of any kind are expressed or implied. The recipient agrees to determine
* applicability of information provided. If this file has been purchased
* on magnetic or optical media from Unicode, Inc., the sole remedy for any claim
* will be exchange of defective media within 90 days of receipt. This disclaimer
* is applicable for all other data files accompanying the Unicode Character Database,
* some of which have been compiled by the Unicode Consortium, and some of which
* have been supplied by other sources.
*
* Limitations on Rights to Redistribute This Data
*
* Recipient is granted the right to make copies in any form for internal
* distribution and to freely use the information supplied in the creation of
* products supporting the UnicodeTM Standard. The files in
* the Unicode Character Database can be redistributed to third parties or other
* organizations (whether for profit or not) as long as this notice and the disclaimer
* notice are retained. Information can be extracted from these files and used
* in documentation or programs, as long as there is an accompanying notice
* indicating the source.
*/
package kotlin.text.regex
/**
* This class represents low surrogate character.
*
* Note that we can use high and low surrogate characters
* that don't combine into supplementary code point.
* See http://www.unicode.org/reports/tr18/#Supplementary_Characters
*/
internal class LowSurrogateCharSet(low: Char) : CharSet(low) {
override fun accepts(startIndex: Int, testString: CharSequence): Int {
val result = super.accepts(startIndex, testString)
if (result < 0 || testString.isHighSurrogate(startIndex - 1)) {
return -1
}
return result
}
private fun CharSequence.isHighSurrogate(index: Int, leftBound: Int = 0, rightBound: Int = length)
= (index in leftBound until rightBound && this[index].isHighSurrogate())
override fun find(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
var index = startIndex
while (index < testString.length) {
index = testString.indexOf(char, index, ignoreCase)
if (index < 0) {
return -1
}
if (!testString.isHighSurrogate(index - 1)
&& next.matches(index + charCount, testString, matchResult) >= 0) {
return index
}
index++
}
return -1
}
override fun findBack(leftLimit: Int, rightLimit: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
var index = rightLimit
while (index >= leftLimit) {
index = testString.lastIndexOf(char, index, ignoreCase)
if (index < 0) {
return -1
}
if (!testString.isHighSurrogate(index - 1, leftLimit, rightLimit)
&& next.matches(index + charCount, testString, matchResult) >= 0) {
return index
}
index--
}
return -1
}
override fun first(set: AbstractSet): Boolean {
return when(set) {
is LowSurrogateCharSet -> set.char == this.char
is CharSet,
is RangeSet,
is SupplementaryCharSet,
is SupplementaryRangeSet -> false
else -> true
}
}
override fun hasConsumed(matchResult: MatchResultImpl): Boolean = true
}
/**
* This class represents high surrogate character.
*/
internal class HighSurrogateCharSet(high: Char) : CharSet(high) {
override fun accepts(startIndex: Int, testString: CharSequence): Int {
val result = super.accepts(startIndex, testString)
if (result < 0 || testString.isLowSurrogate(startIndex + 1)) {
return -1
}
return result
}
private fun CharSequence.isLowSurrogate(index: Int, leftBound: Int = 0, rightBound: Int = length)
= (index in leftBound until rightBound && this[index].isLowSurrogate())
// TODO: We have a similar code here, in LowSurrogateCharSet and in CharSet. Reuse it somehow.
override fun find(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
var index = startIndex
while (index < testString.length) {
index = testString.indexOf(char, index, ignoreCase)
if (index < 0) {
return -1
}
// Remove params.
if (!testString.isLowSurrogate(index + 1)
&& next.matches(index + charCount, testString, matchResult) >= 0) {
return index
}
index++
}
return -1
}
override fun findBack(leftLimit: Int, rightLimit: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
var index = rightLimit
while (index >= leftLimit) {
index = testString.lastIndexOf(char, index, ignoreCase)
if (index < 0) {
return -1
}
if (!testString.isLowSurrogate(index + 1, leftLimit, rightLimit)
&& next.matches(index + charCount, testString, matchResult) >= 0) {
return index
}
index--
}
return -1
}
override fun first(set: AbstractSet): Boolean {
return when (set) {
is HighSurrogateCharSet -> set.char == this.char
is CharSet,
is RangeSet,
is SupplementaryCharSet,
is SupplementaryRangeSet -> false
else -> true
}
}
override fun hasConsumed(matchResult: MatchResultImpl): Boolean = true
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/SurrogateRangeSet.kt
+124
View File
@@ -0,0 +1,124 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/*
*
* Portions, Copyright © 1991-2005 Unicode, Inc. The following applies to Unicode.
*
* COPYRIGHT AND PERMISSION NOTICE
*
* Copyright © 1991-2005 Unicode, Inc. All rights reserved. Distributed under
* the Terms of Use in http://www.unicode.org/copyright.html. Permission is
* hereby granted, free of charge, to any person obtaining a copy of the
* Unicode data files and any associated documentation (the "Data Files")
* or Unicode software and any associated documentation (the "Software")
* to deal in the Data Files or Software without restriction, including without
* limitation the rights to use, copy, modify, merge, publish, distribute,
* and/or sell copies of the Data Files or Software, and to permit persons
* to whom the Data Files or Software are furnished to do so, provided that
* (a) the above copyright notice(s) and this permission notice appear with
* all copies of the Data Files or Software, (b) both the above copyright
* notice(s) and this permission notice appear in associated documentation,
* and (c) there is clear notice in each modified Data File or in the Software
* as well as in the documentation associated with the Data File(s) or Software
* that the data or software has been modified.
* THE DATA FILES AND SOFTWARE ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY
* KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT
* OF THIRD PARTY RIGHTS. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR HOLDERS
* INCLUDED IN THIS NOTICE BE LIABLE FOR ANY CLAIM, OR ANY SPECIAL INDIRECT
* OR CONSEQUENTIAL DAMAGES, OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS
* OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR
* OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
* PERFORMANCE OF THE DATA FILES OR SOFTWARE.
*
* Except as contained in this notice, the name of a copyright holder shall
* not be used in advertising or otherwise to promote the sale, use or other
* dealings in these Data Files or Software without prior written
* authorization of the copyright holder.
*
* 2. Additional terms from the Database:
*
* Copyright © 1995-1999 Unicode, Inc. All Rights reserved.
*
* Disclaimer
*
* The Unicode Character Database is provided as is by Unicode, Inc.
* No claims are made as to fitness for any particular purpose. No warranties
* of any kind are expressed or implied. The recipient agrees to determine
* applicability of information provided. If this file has been purchased
* on magnetic or optical media from Unicode, Inc., the sole remedy for any claim
* will be exchange of defective media within 90 days of receipt. This disclaimer
* is applicable for all other data files accompanying the Unicode Character Database,
* some of which have been compiled by the Unicode Consortium, and some of which
* have been supplied by other sources.
*
* Limitations on Rights to Redistribute This Data
*
* Recipient is granted the right to make copies in any form for internal
* distribution and to freely use the information supplied in the creation of
* products supporting the UnicodeTM Standard. The files in
* the Unicode Character Database can be redistributed to third parties or other
* organizations (whether for profit or not) as long as this notice and the disclaimer
* notice are retained. Information can be extracted from these files and used
* in documentation or programs, as long as there is an accompanying notice
* indicating the source.
*/
package kotlin.text.regex
/*
* This class is a range that contains only surrogate characters.
*/
internal class SurrogateRangeSet(surrChars: AbstractCharClass) : RangeSet(surrChars) {
override fun accepts(startIndex: Int, testString: CharSequence): Int {
val result = super.accepts(startIndex, testString)
when {
result < 0 ||
testString.isHighSurrogate(startIndex - 1) && testString.isLowSurrogate(startIndex) ||
testString.isHighSurrogate(startIndex) && testString.isLowSurrogate(startIndex + 1) ->
return -1
}
return result
}
private fun CharSequence.isHighSurrogate(index: Int, leftBound: Int = 0, rightBound: Int = length)
= (index in leftBound until rightBound && this[index].isHighSurrogate())
private fun CharSequence.isLowSurrogate(index: Int, leftBound: Int = 0, rightBound: Int = length)
= (index in leftBound until rightBound && this[index].isLowSurrogate())
override fun first(set: AbstractSet): Boolean {
return when (set) {
is SurrogateRangeSet -> true
is CharSet,
is RangeSet,
is SupplementaryCharSet,
is SupplementaryRangeSet -> false
else -> true
}
}
override fun hasConsumed(matchResult: MatchResultImpl): Boolean = true
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/UnifiedQuantifierSet.kt
+57
View File
@@ -0,0 +1,57 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Optimized greedy quantifier node ('*') for the case where there is no intersection with
* next node and normal quantifiers could be treated as greedy and possessive.
*/
internal class UnifiedQuantifierSet(quant: LeafQuantifierSet) : LeafQuantifierSet(Quantifier.starQuantifier, quant.leaf, quant.next, quant.type) {
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
var index = startIndex
while (index + leaf.charCount <= testString.length && leaf.accepts(index, testString) > 0) {
index += leaf.charCount
}
return next.matches(index, testString, matchResult)
}
override fun find(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
var startSearch = next.find(startIndex, testString, matchResult)
if (startSearch < 0)
return -1
var result = startSearch
var index = startSearch - leaf.charCount
while (index >= startIndex && leaf.accepts(index, testString) > 0) {
result = index
index -= leaf.charCount
}
return result
}
init {
innerSet.next = this
}
}
libraries/stdlib/native-wasm/src/kotlin/text/regex/sets/WordBoundarySet.kt
+67
View File
@@ -0,0 +1,67 @@
/*
* Copyright 2010-2018 JetBrains s.r.o. Use of this source code is governed by the Apache 2.0 license
* that can be found in the LICENSE file.
*/
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package kotlin.text.regex
/**
* Represents word boundary, checks current character and previous one. If they have different types returns true;
*/
internal class WordBoundarySet(var positive: Boolean) : SimpleSet() {
override fun matches(startIndex: Int, testString: CharSequence, matchResult: MatchResultImpl): Int {
val curChar = if (startIndex >= testString.length) ' ' else testString[startIndex]
val prevChar = if (startIndex == 0) ' ' else testString[startIndex - 1]
val right = curChar == ' ' || isSpace(curChar, startIndex, testString)
val left = prevChar == ' ' || isSpace(prevChar, startIndex - 1, testString)
return if (left xor right xor positive)
-1
else
next.matches(startIndex, testString, matchResult)
}
/** Returns false, because word boundary does not consumes any characters and do not move string index. */
override fun hasConsumed(matchResult: MatchResultImpl): Boolean = false
override val name: String
get() = "WordBoundarySet"
private fun isSpace(char: Char, startIndex: Int, testString: CharSequence): Boolean {
if (char.isLetterOrDigit() || char == '_') {
return false
}
if (char.category == CharCategory.NON_SPACING_MARK) {
var index = startIndex
while (--index >= 0) {
val ch = testString[index]
when {
ch.isLetterOrDigit() -> return false
char.category != CharCategory.NON_SPACING_MARK -> return true
}
}
}
return true
}
}