Redigera

Dela via


CompareOptions Enum

Definition

Defines the string comparison options to use with CompareInfo.

This enumeration supports a bitwise combination of its member values.

public enum class CompareOptions
[System.Flags]
public enum CompareOptions
[System.Flags]
[System.Serializable]
public enum CompareOptions
[System.Flags]
[System.Serializable]
[System.Runtime.InteropServices.ComVisible(true)]
public enum CompareOptions
[<System.Flags>]
type CompareOptions = 
[<System.Flags>]
[<System.Serializable>]
type CompareOptions = 
[<System.Flags>]
[<System.Serializable>]
[<System.Runtime.InteropServices.ComVisible(true)>]
type CompareOptions = 
Public Enum CompareOptions
Inheritance
CompareOptions
Attributes

Fields

Name Value Description
None 0

Indicates the default option settings for string comparisons.

IgnoreCase 1

Indicates that the string comparison ignores case differences.

IgnoreNonSpace 2

Indicates that the string comparison ignores nonspacing combining characters, such as diacritics. Nonspacing characters modify base characters without occupying their own space. The Unicode Standard defines combining characters as characters that are combined with base characters to produce a new character.

IgnoreSymbols 4

Indicates that the string comparison ignores symbols, including whitespace, punctuation, currency symbols, the percent sign, mathematical symbols, the ampersand, and similar characters.

IgnoreKanaType 8

Indicates that the string comparison ignores the kana type. Kana type refers to Japanese hiragana and katakana characters, which represent phonetic sounds. Hiragana is used for native Japanese words, while katakana is used for words borrowed from other languages. With this option, hiragana and katakana characters that represent the same sound are considered equal.

IgnoreWidth 16

Indicates that the string comparison ignores character width. For example, full-width and half-width forms of Japanese katakana characters are considered equal with this option.

NumericOrdering 32
OrdinalIgnoreCase 268435456

Indicates that the string comparison ignores case and then performs an ordinal comparison. This is equivalent to converting both strings to uppercase using the invariant culture and then performing the comparison.

StringSort 536870912

Indicates that the string comparison uses the string sort algorithm, where nonalphanumeric symbols (such as hyphens and apostrophes) are sorted before alphanumeric characters.

Ordinal 1073741824

Indicates that the string comparison uses the Unicode UTF-16 encoded values of the strings, comparing them code unit by code unit. This results in a fast, culture-insensitive comparison where strings are ordered based only on their binary values. This option can't be combined with other CompareOptions values and must be used alone.

Examples

The following code example shows how each of the CompareOptions values affect string comparisons.

using System;
using System.Globalization;

public class CompareOptionsExample
{
    public static void Run()
    {
        // Uppercase and lowercase characters are equivalent (according to the culture rules)
        // when IgnoreCase is used.
        TestStringEquality("ONE two", "one TWO", "Case sensitivity", CompareOptions.IgnoreCase);

        // Punctuation is ignored with the IgnoreSymbols option.
        TestStringEquality("hello world", "hello, world!", "Punctuation", CompareOptions.IgnoreSymbols);

        // Whitespace and mathematical symbols are also ignored with IgnoreSymbols.
        TestStringEquality("3 + 5 = 8", "358", "Whitespace and mathematical symbols", CompareOptions.IgnoreSymbols);

        // Caution: currency symbols and thousands separators are ignored with IgnoreSymbols.
        // Parse strings containing numbers/currency and compare them numerically instead.
        TestStringEquality("Total $15,000", "Total: £150.00", "Currency symbols, decimals and thousands separators", CompareOptions.IgnoreSymbols);

        // Full width characters are common in East Asian languages. Use the IgnoreWidth
        // option to treat full- and half-width characters as equal.
        TestStringEquality("abc,-", "abc,-", "Half width and full width characters", CompareOptions.IgnoreWidth);

        // The same string in Hiragana and Katakana is equal when IgnoreKanaType is used.
        TestStringEquality("ありがとう", "アリガトウ", "Hiragana and Katakana strings", CompareOptions.IgnoreKanaType);

        // When comparing with the IgnoreNonSpace option, characters like diacritical marks are ignored.
        TestStringEquality("café", "cafe", "Diacritical marks", CompareOptions.IgnoreNonSpace);

        // Ligature characters and their non-ligature forms compare equal with the IgnoreNonSpace option.
        // Note: prior to .NET 5, ligature characters were equal to their expanded forms by default.
        TestStringEquality("straße œuvre cæsar", "strasse oeuvre caesar", "Ligature characters", CompareOptions.IgnoreNonSpace);
    }

    private static void TestStringEquality(string str1, string str2, string description, CompareOptions options)
    {
        Console.WriteLine(Environment.NewLine + description + ":");
        // First test with the default CompareOptions then with the provided options
        TestStringEquality(str1, str2, CompareOptions.None);
        TestStringEquality(str1, str2, options);
    }

    private static void TestStringEquality(string str1, string str2, CompareOptions options)
    {
        Console.Write($"  When using CompareOptions.{options}, \"{str1}\" and \"{str2}\" are ");
        if (string.Compare(str1, str2, CultureInfo.InvariantCulture, options) != 0)
        {
            Console.Write("not ");
        }
        Console.WriteLine("equal.");
    }
}

/*
In .NET 5 and later, the output is the following:
 
Case sensitivity:
  When using CompareOptions.None, "ONE two" and "one TWO" are not equal.
  When using CompareOptions.IgnoreCase, "ONE two" and "one TWO" are equal.

Punctuation:
  When using CompareOptions.None, "hello world" and "hello, world!" are not equal.
  When using CompareOptions.IgnoreSymbols, "hello world" and "hello, world!" are equal.

Whitespace and mathematical symbols:
  When using CompareOptions.None, "3 + 5 = 8" and "358" are not equal.
  When using CompareOptions.IgnoreSymbols, "3 + 5 = 8" and "358" are equal.

Currency symbols, decimals and thousands separators:
  When using CompareOptions.None, "Total $15,000" and "Total: £150.00" are not equal.
  When using CompareOptions.IgnoreSymbols, "Total $15,000" and "Total: £150.00" are equal.

Half width and full width characters:
  When using CompareOptions.None, "abc,-" and "abc,-" are not equal.
  When using CompareOptions.IgnoreWidth, "abc,-" and "abc,-" are equal.

Hiragana and Katakana strings:
  When using CompareOptions.None, "ありがとう" and "アリガトウ" are not equal.
  When using CompareOptions.IgnoreKanaType, "ありがとう" and "アリガトウ" are equal.

Diacritical marks:
  When using CompareOptions.None, "café" and "cafe" are not equal.
  When using CompareOptions.IgnoreNonSpace, "café" and "cafe" are equal.

Ligature characters:
  When using CompareOptions.None, "straße œuvre cæsar" and "strasse oeuvre caesar" are not equal.
  When using CompareOptions.IgnoreNonSpace, "straße œuvre cæsar" and "strasse oeuvre caesar" are equal.

Note: When using .NET versions prior to .NET 5, ligature characters compare as equal to their
non-ligature counterparts by default, so the last test will output as follows:

Ligature characters:
  When using CompareOptions.None, "straße œuvre cæsar" and "strasse oeuvre caesar" are equal.
  When using CompareOptions.IgnoreNonSpace, "straße œuvre cæsar" and "strasse oeuvre caesar" are equal.
*/
module compareoptions_values

open System
open System.Globalization

let testStringEquality (str1: string) (str2: string) (description: string) (options: CompareOptions) =
    printfn "\n%s:" description

    let compareAndPrint opts =
        let result = String.Compare(str1, str2, CultureInfo.InvariantCulture, opts)
        let equalityStatus = if result = 0 then "equal" else "not equal"
        printfn "  When using CompareOptions.%A, \"%s\" and \"%s\" are %s." opts str1 str2 equalityStatus

    compareAndPrint CompareOptions.None
    compareAndPrint options

[<EntryPoint>]
let main argv =
    // Uppercase and lowercase characters are equivalent (according to the culture rules) when IgnoreCase is used.
    testStringEquality "ONE two" "one TWO" "Case sensitivity" CompareOptions.IgnoreCase

    // Punctuation is ignored with the IgnoreSymbols option.
    testStringEquality "hello world" "hello, world!" "Punctuation" CompareOptions.IgnoreSymbols

    // Whitespace and mathematical symbols are also ignored with IgnoreSymbols.
    testStringEquality "3 + 5 = 8" "358" "Whitespace and mathematical symbols" CompareOptions.IgnoreSymbols

    // Caution: currency symbols and thousands separators are ignored with IgnoreSymbols.
    // Parse strings containing numbers/currency and compare them numerically instead.
    testStringEquality "Total $15,000" "Total: £150.00" "Currency symbols, decimals and thousands separators" CompareOptions.IgnoreSymbols

    // Full width characters are common in East Asian languages. Use the IgnoreWidth
    // option to treat full- and half-width characters as equal.
    testStringEquality "abc,-" "abc,-" "Half width and full width characters" CompareOptions.IgnoreWidth

    // The same string in Hiragana and Katakana is equal when IgnoreKanaType is used.
    testStringEquality "ありがとう" "アリガトウ" "Hiragana and Katakana strings" CompareOptions.IgnoreKanaType

    // When comparing with the IgnoreNonSpace option, characters like diacritical marks are ignored.
    testStringEquality "café" "cafe" "Diacritical marks" CompareOptions.IgnoreNonSpace

    // Ligature characters and their non-ligature forms compare equal with the IgnoreNonSpace option.
    // Note: prior to .NET 5, ligature characters were equal to their expanded forms by default.
    testStringEquality "straße œuvre cæsar" "strasse oeuvre caesar" "Ligature characters" CompareOptions.IgnoreNonSpace

    0 // return an integer exit code

(*
In .NET 5 and later, the output will be the following:

Case sensitivity:
  When using CompareOptions.None, "ONE two" and "one TWO" are not equal.
  When using CompareOptions.IgnoreCase, "ONE two" and "one TWO" are equal.

Punctuation:
  When using CompareOptions.None, "hello world" and "hello, world!" are not equal.
  When using CompareOptions.IgnoreSymbols, "hello world" and "hello, world!" are equal.

Whitespace and mathematical symbols:
  When using CompareOptions.None, "3 + 5 = 8" and "358" are not equal.
  When using CompareOptions.IgnoreSymbols, "3 + 5 = 8" and "358" are equal.

Currency symbols, decimals and thousands separators:
  When using CompareOptions.None, "Total $15,000" and "Total: £150.00" are not equal.
  When using CompareOptions.IgnoreSymbols, "Total $15,000" and "Total: £150.00" are equal.

Half width and full width characters:
  When using CompareOptions.None, "abc,-" and "abc,-" are not equal.
  When using CompareOptions.IgnoreWidth, "abc,-" and "abc,-" are equal.

Hiragana and Katakana strings:
  When using CompareOptions.None, "ありがとう" and "アリガトウ" are not equal.
  When using CompareOptions.IgnoreKanaType, "ありがとう" and "アリガトウ" are equal.

Diacritical marks:
  When using CompareOptions.None, "café" and "cafe" are not equal.
  When using CompareOptions.IgnoreNonSpace, "café" and "cafe" are equal.

Ligature characters:
  When using CompareOptions.None, "straße œuvre cæsar" and "strasse oeuvre caesar" are not equal.
  When using CompareOptions.IgnoreNonSpace, "straße œuvre cæsar" and "strasse oeuvre caesar" are equal.


Note: when using .NET versions prior to .NET 5, ligature characters compare as equal to their
non-ligature counterparts by default, so the last test will output as follows:

Ligature characters:
  When using CompareOptions.None, "straße œuvre cæsar" and "strasse oeuvre caesar" are equal.
  When using CompareOptions.IgnoreNonSpace, "straße œuvre cæsar" and "strasse oeuvre caesar" are equal.
*)
Imports System
Imports System.Globalization

Module CompareOptionsExample
    Sub Main()
        ' Uppercase and lowercase characters are equivalent (according to the culture rules)
        ' when IgnoreCase is used.
        TestStringEquality("ONE two", "one TWO", "Case sensitivity", CompareOptions.IgnoreCase)

        ' Punctuation is ignored with the IgnoreSymbols option.
        TestStringEquality("hello world", "hello, world!", "Punctuation", CompareOptions.IgnoreSymbols)

        ' Whitespace and mathematical symbols are also ignored with IgnoreSymbols.
        TestStringEquality("3 + 5 = 8", "358", "Whitespace and mathematical symbols", CompareOptions.IgnoreSymbols)

        ' Caution: currency symbols and thousands separators are ignored with IgnoreSymbols.
        ' Parse strings containing numbers/currency and compare them numerically instead.
        TestStringEquality("Total $15,000", "Total: £150.00", "Currency symbols, decimals and thousands separators", CompareOptions.IgnoreSymbols)

        ' Full width characters are common in East Asian languages. Use the IgnoreWidth
        ' option to treat full- and half-width characters as equal.
        TestStringEquality("abc,-", "abc,-", "Half width and full width characters", CompareOptions.IgnoreWidth)

        ' The same string in Hiragana and Katakana is equal when IgnoreKanaType is used.
        TestStringEquality("ありがとう", "アリガトウ", "Hiragana and Katakana strings", CompareOptions.IgnoreKanaType)

        ' When comparing with the IgnoreNonSpace option, characters like diacritical marks are ignored.
        TestStringEquality("café", "cafe", "Diacritical marks", CompareOptions.IgnoreNonSpace)

        ' Ligature characters and their non-ligature forms compare equal with the IgnoreNonSpace option.
        ' Note: prior to .NET 5, ligature characters were equal to their expanded forms by default.
        TestStringEquality("straße œuvre cæsar", "strasse oeuvre caesar", "Ligature characters", CompareOptions.IgnoreNonSpace)
    End Sub

    Private Sub TestStringEquality(str1 As String, str2 As String, description As String, options As CompareOptions)
        Console.WriteLine(Environment.NewLine & description & ":")
        ' First test with the default CompareOptions then with the provided options
        TestStringEqualityWithOptions(str1, str2, CompareOptions.None)
        TestStringEqualityWithOptions(str1, str2, options)
    End Sub

    Private Sub TestStringEqualityWithOptions(str1 As String, str2 As String, options As CompareOptions)
        Console.Write($"  When using CompareOptions.{options}, ""{str1}"" and ""{str2}"" are ")
        If String.Compare(str1, str2, CultureInfo.InvariantCulture, options) <> 0 Then
            Console.Write("not ")
        End If
        Console.WriteLine("equal.")
    End Sub
End Module

' In .NET 5 and later, the output is the following:
'
'Case sensitivity :
'  When using CompareOptions.None, "ONE two" and "one TWO" are not equal.
'  When using CompareOptions.IgnoreCase, "ONE two" and "one TWO" are equal.
'
'Punctuation:
'  When using CompareOptions.None, "hello world" and "hello, world!" are not equal.
'  When using CompareOptions.IgnoreSymbols, "hello world" and "hello, world!" are equal.
'
'Whitespace And mathematical symbols:
'  When using CompareOptions.None, "3 + 5 = 8" and "358" are not equal.
'  When using CompareOptions.IgnoreSymbols, "3 + 5 = 8" and "358" are equal.
'
'Currency symbols, decimals And thousands separators:
'  When using CompareOptions.None, "Total $15,000" and "Total: £150.00" are not equal.
'  When using CompareOptions.IgnoreSymbols, "Total $15,000" and "Total: £150.00" are equal.
'
'Half width And full width characters:
'  When using CompareOptions.None, "abc,-" and "abc,-" are not equal.
'  When using CompareOptions.IgnoreWidth, "abc,-" and "abc,-" are equal.
'
'Hiragana And Katakana strings:
'  When using CompareOptions.None, "ありがとう" and "アリガトウ" are not equal.
'  When using CompareOptions.IgnoreKanaType, "ありがとう" and "アリガトウ" are equal.
'
'Diacritical marks :
'  When using CompareOptions.None, "café" and "cafe" are not equal.
'  When using CompareOptions.IgnoreNonSpace, "café" and "cafe" are equal.
'
'Ligature characters :
'  When using CompareOptions.None, "straße œuvre cæsar" and "strasse oeuvre caesar" are not equal.
'  When using CompareOptions.IgnoreNonSpace, "straße œuvre cæsar" and "strasse oeuvre caesar" are equal.
'
' Note: when using .NET versions prior to .NET 5, ligature characters compare as equal to their
' non-ligature counterparts by default, so the last test will output as follows:
'
'Ligature characters :
'  When using CompareOptions.None, "straße œuvre cæsar" and "strasse oeuvre caesar" are equal.
'  When using CompareOptions.IgnoreNonSpace, "straße œuvre cæsar" and "strasse oeuvre caesar" are equal.

The following code example shows how sorting with StringSort differs from sorting without StringSort.

using System;
using System.Collections.Generic;
using System.Globalization;

public class StringSort
{
    public static void Run()
    {
        var wordList = new List<string>
        {
            "cant", "bill's", "coop", "cannot", "billet", "can't", "con", "bills", "co-op"
        };

        Console.WriteLine("Before sorting:");
        foreach (string word in wordList)
        {
            Console.WriteLine(word);
        }

        Console.WriteLine(Environment.NewLine + "After sorting with CompareOptions.None:");
        SortAndDisplay(wordList, CompareOptions.None);

        Console.WriteLine(Environment.NewLine + "After sorting with CompareOptions.StringSort:");
        SortAndDisplay(wordList, CompareOptions.StringSort);
    }

    // Sort the list of words with the supplied CompareOptions.
    private static void SortAndDisplay(List<string> unsorted, CompareOptions options)
    {
        // Create a copy of the original list to sort.
        var words = new List<string>(unsorted);
        // Define the CompareInfo to use to compare strings.
        CompareInfo comparer = CultureInfo.InvariantCulture.CompareInfo;

        // Sort the copy with the supplied CompareOptions then display.
        words.Sort((str1, str2) => comparer.Compare(str1, str2, options));
        foreach (string word in words)
        {
            Console.WriteLine(word);
        }
    }
}

/*
CompareOptions.None and CompareOptions.StringSort provide identical ordering by default
in .NET 5 and later. But in prior versions, the output is the following:

Before sorting:
cant
bill's
coop
cannot
billet
can't
con
bills
co-op

After sorting with CompareOptions.None:
billet
bills
bill's
cannot
cant
can't
con
coop
co-op

After sorting with CompareOptions.StringSort:
bill's
billet
bills
can't
cannot
cant
co-op
con
coop
*/
module compareoptions_stringsort

open System
open System.Collections.Generic
open System.Globalization

let sortAndDisplay (unsorted: List<string>) (options: CompareOptions) =
    let words = new List<string>(unsorted)
    let comparer = CultureInfo.InvariantCulture.CompareInfo
    words.Sort((fun str1 str2 -> comparer.Compare(str1, str2, options)))
    for word in words do
        printfn "%s" word

[<EntryPoint>]
let main argv =
    let wordList = new List<string>(
        ["cant"; "bill's"; "coop"; "cannot"; "billet"; "can't"; "con"; "bills"; "co-op"])

    printfn "Before sorting:"
    for word in wordList do
        printfn "%s" word

    printfn "\nAfter sorting with CompareOptions.None:"
    sortAndDisplay wordList CompareOptions.None

    printfn "\nAfter sorting with CompareOptions.StringSort:"
    sortAndDisplay wordList CompareOptions.StringSort

    0 // return an integer exit code

(*
CompareOptions.None and CompareOptions.StringSort provide identical ordering by default
in .NET 5 and later, but in prior versions, the output will be the following:

Before sorting:
cant
bill's
coop
cannot
billet
can't
con
bills
co-op

After sorting with CompareOptions.None:
billet
bills
bill's
cannot
cant
can't
con
coop
co-op

After sorting with CompareOptions.StringSort:
bill's
billet
bills
can't
cannot
cant
co-op
con
coop
*)
Imports System
Imports System.Collections.Generic
Imports System.Globalization

Public Class StringSort
    Public Shared Sub Main()
        Dim wordList As New List(Of String) From {
            "cant", "bill's", "coop", "cannot", "billet", "can't", "con", "bills", "co-op"
        }

        Console.WriteLine("Before sorting:")
        For Each word In wordList
            Console.WriteLine(word)
        Next

        Console.WriteLine(Environment.NewLine & "After sorting with CompareOptions.None:")
        SortAndDisplay(wordList, CompareOptions.None)

        Console.WriteLine(Environment.NewLine & "After sorting with CompareOptions.StringSort:")
        SortAndDisplay(wordList, CompareOptions.StringSort)
    End Sub

    ' Sort the list of words with the supplied CompareOptions.
    Private Shared Sub SortAndDisplay(unsorted As List(Of String), options As CompareOptions)
        ' Create a copy of the original list to sort.
        Dim words As New List(Of String)(unsorted)

        ' Define the CompareInfo to use to compare strings.
        Dim comparer As CompareInfo = CultureInfo.InvariantCulture.CompareInfo

        ' Sort the copy with the supplied CompareOptions then display.
        words.Sort(Function(str1, str2) comparer.Compare(str1, str2, options))

        For Each word In words
            Console.WriteLine(word)
        Next
    End Sub
End Class

' CompareOptions.None and CompareOptions.StringSort provide identical ordering by default
' in .NET 5 And later, but in prior versions, the output is the following:
'
'Before sorting
'cant
'bill's
'coop
'cannot
'billet
'can't
'con
'bills
'co-op

'After sorting with CompareOptions.None
'billet
'bills
'bill's
'cannot
'cant
'can't
'con
'coop
'co-op

'After sorting with CompareOptions.StringSort
'bill's
'billet
'bills
'can't
'cannot
'cant
'co-op
'con
'coop

Remarks

In .NET 5 and later, the cross-platform ICU (International Components for Unicode) library is used for string processing. The ICU library brings the following changes to string comparison behavior:

  • The default option None is now equivalent to the StringSort option. The previous functionality of None, where equal weighting was given to alphanumeric and nonalphanumeric characters, is no longer available.
  • Ligatures (combined characters like "æ" and "œ") are now seen as distinct from their expanded forms ("ae", "oe") in string comparisons by default. To treat ligatures and their expanded forms as equivalent, use the IgnoreNonSpace option.

For more information about the change, including how to restore the previous Unicode handler, see .NET globalization and ICU.

For more information about this API, see Supplemental API remarks for CompareOptions.

Applies to

See also