ShrlAlgoToolkit/WPFluent/Controls/PersonPicture/InitialsGenerator.cs

using System;
using System.Linq;

namespace WPFluent.Controls;

/// <summary>
/// Value indicating the general character set for a given character.
/// </summary>
internal enum CharacterType
{
    /// <summary>
    /// Indicates we could not match the character set.
    /// </summary>
    Other = 0,

    /// <summary>
    /// Member of the Latin character set.
    /// </summary>
    Standard = 1,

    /// <summary>
    /// Member of a symbolic character set.
    /// </summary>
    Symbolic = 2,

    /// <summary>
    /// Member of a character set which supports glyphs.
    /// </summary>
    Glyph = 3
};

/// <summary>
/// PersonPicture Control. Displays the Profile Picture, or in its absence Initials,
/// for a given Contact.
/// </summary>
internal sealed class InitialsGenerator
{
    /// <summary>
    /// Helper function which takes a DisplayName, as generated by
    /// Windows.ApplicationModel.Contacts, and returns a initials representation.
    /// </summary>
    /// <param name="contactDisplayName>The DisplayName of the person</param>
    /// <returns>
    /// String containing the initials representation of the given DisplayName.
    /// </returns>
    public static string InitialsFromDisplayName(string contactDisplayName)
    {
        CharacterType type = GetCharacterType(contactDisplayName);

        // We'll attempt to make initials only if we recognize a name in the Standard character set.
        if (type == CharacterType.Standard)
        {
            string displayName = contactDisplayName;

            StripTrailingBrackets(ref displayName);

            string[] words = Split(displayName, ' ');

            if (words.Length == 1)
            {
                // If there's only a single long word, we'll show one initial.
                string firstWord = words.First();

                string result = GetFirstFullCharacter(firstWord);

                return result.ToUpper();
            }
            else if (words.Length > 1)
            {
                // If there's at least two words, we'll show two initials.
                //
                // NOTE: Based on current implementation, we could be showing punctuation.
                // For example, "John -Smith" would be "J-".
                string firstWord = words.First();
                string lastWord = words.Last();

                string result = GetFirstFullCharacter(firstWord);
                result += GetFirstFullCharacter(lastWord);

                return result.ToUpper();
            }
            else
            {
                // If there's only spaces in the name, we'll get a Vector size of 0.
                return string.Empty;
            }
        }
        else
        {
            // Return empty string. In our code-behind we will produce a generic glyph as a result.
            return string.Empty;
        }
    }

    /// <summary>
    /// Helper function which indicates the type of characters in a given string
    /// </summary>
    /// <param name="str">String from which to detect character-set.</param>
    /// <returns>
    /// Character set of the string: Latin, Symbolic, Glyph, or other.
    /// </returns>
    public static CharacterType GetCharacterType(string str)
    {
        // Since we're doing initials, we're really only interested in the first
        // few characters. If the first three characters aren't a glyph then
        // we don't need to count it as such because we won't be changing meaning
        // by truncating to one or two.
        CharacterType result = CharacterType.Other;

        for (int i = 0; i < 3; i++)
        {
            // Break on null character. 0xFEFF is a terminating character which appears as null.
            if ((i >= str.Length) || (str[i] == '\0') || (str[i] == 0xFEFF))
            {
                break;
            }

            char character = str[i];
            CharacterType evaluationResult = GetCharacterType(character);

            // In mix-match scenarios, we'll want to follow this order of precedence:
            // Glyph > Symbolic > Roman
            switch (evaluationResult)
            {
                case CharacterType.Glyph:
                    result = CharacterType.Glyph;
                    break;

                case CharacterType.Symbolic:
                    // Don't override a Glyph state with a Symbolic State.
                    if (result != CharacterType.Glyph)
                    {
                        result = CharacterType.Symbolic;
                    }
                    break;

                case CharacterType.Standard:
                    // Don't override a Glyph or Symbolic state with a Latin state.
                    if ((result != CharacterType.Glyph) && (result != CharacterType.Symbolic))
                    {
                        result = CharacterType.Standard;
                    }
                    break;

                default:
                    // Preserve result's current state (if we never got data other
                    // than "Other", it'll be set to other anyway).
                    break;
            }
        }

        return result;
    }

    /// <summary>
    /// Helper function which indicates the character-set of a given character.
    /// </summary>
    /// <param name="character">Character for which to detect character-set.</param>
    /// <returns>
    /// Character set of the string: Latin, Symbolic, Glyph, or other.
    /// </returns>
    public static CharacterType GetCharacterType(char character)
    {
        // To ensure predictable behavior, we're currently operating on an allowed list of character sets.
        //
        // Each block below is a HEX range in the official Unicode spec, which defines a set
        // of Unicode characters. Changes to the character sets would only be made by Unicode, and
        // are highly unlikely (as it would break virtually every modern text parser).
        // Definitions available here: http://www.unicode.org/charts/
        //
        // GLYPH
        //
        // IPA Extensions
        if ((character >= 0x0250) && (character <= 0x02AF))
        {
            return CharacterType.Glyph;
        }

        // Arabic
        if ((character >= 0x0600) && (character <= 0x06FF))
        {
            return CharacterType.Glyph;
        }

        // Arabic Supplement
        if ((character >= 0x0750) && (character <= 0x077F))
        {
            return CharacterType.Glyph;
        }

        // Arabic Extended-A
        if ((character >= 0x08A0) && (character <= 0x08FF))
        {
            return CharacterType.Glyph;
        }

        // Arabic Presentation Forms-A
        if ((character >= 0xFB50) && (character <= 0xFDFF))
        {
            return CharacterType.Glyph;
        }

        // Arabic Presentation Forms-B
        if ((character >= 0xFE70) && (character <= 0xFEFF))
        {
            return CharacterType.Glyph;
        }

        // Devanagari
        if ((character >= 0x0900) && (character <= 0x097F))
        {
            return CharacterType.Glyph;
        }

        // Devanagari Extended
        if ((character >= 0xA8E0) && (character <= 0xA8FF))
        {
            return CharacterType.Glyph;
        }

        // Bengali
        if ((character >= 0x0980) && (character <= 0x09FF))
        {
            return CharacterType.Glyph;
        }

        // Gurmukhi
        if ((character >= 0x0A00) && (character <= 0x0A7F))
        {
            return CharacterType.Glyph;
        }

        // Gujarati
        if ((character >= 0x0A80) && (character <= 0x0AFF))
        {
            return CharacterType.Glyph;
        }

        // Oriya
        if ((character >= 0x0B00) && (character <= 0x0B7F))
        {
            return CharacterType.Glyph;
        }

        // Tamil
        if ((character >= 0x0B80) && (character <= 0x0BFF))
        {
            return CharacterType.Glyph;
        }

        // Telugu
        if ((character >= 0x0C00) && (character <= 0x0C7F))
        {
            return CharacterType.Glyph;
        }

        // Kannada
        if ((character >= 0x0C80) && (character <= 0x0CFF))
        {
            return CharacterType.Glyph;
        }

        // Malayalam
        if ((character >= 0x0D00) && (character <= 0x0D7F))
        {
            return CharacterType.Glyph;
        }

        // Sinhala
        if ((character >= 0x0D80) && (character <= 0x0DFF))
        {
            return CharacterType.Glyph;
        }

        // Thai
        if ((character >= 0x0E00) && (character <= 0x0E7F))
        {
            return CharacterType.Glyph;
        }

        // Lao
        if ((character >= 0x0E80) && (character <= 0x0EFF))
        {
            return CharacterType.Glyph;
        }

        // SYMBOLIC
        //
        // CJK Unified Ideographs
        if ((character >= 0x4E00) && (character <= 0x9FFF))
        {
            return CharacterType.Symbolic;
        }

        // CJK Unified Ideographs Extension
        if ((character >= 0x3400) && (character <= 0x4DBF))
        {
            return CharacterType.Symbolic;
        }

        // CJK Unified Ideographs Extension B
        if ((character >= 0x20000) && (character <= 0x2A6DF))
        {
            return CharacterType.Symbolic;
        }

        // CJK Unified Ideographs Extension C
        if ((character >= 0x2A700) && (character <= 0x2B73F))
        {
            return CharacterType.Symbolic;
        }

        // CJK Unified Ideographs Extension D
        if ((character >= 0x2B740) && (character <= 0x2B81F))
        {
            return CharacterType.Symbolic;
        }

        // CJK Radicals Supplement
        if ((character >= 0x2E80) && (character <= 0x2EFF))
        {
            return CharacterType.Symbolic;
        }

        // CJK Symbols and Punctuation
        if ((character >= 0x3000) && (character <= 0x303F))
        {
            return CharacterType.Symbolic;
        }

        // CJK Strokes
        if ((character >= 0x31C0) && (character <= 0x31EF))
        {
            return CharacterType.Symbolic;
        }

        // Enclosed CJK Letters and Months
        if ((character >= 0x3200) && (character <= 0x32FF))
        {
            return CharacterType.Symbolic;
        }

        // CJK Compatibility
        if ((character >= 0x3300) && (character <= 0x33FF))
        {
            return CharacterType.Symbolic;
        }

        // CJK Compatibility Ideographs
        if ((character >= 0xF900) && (character <= 0xFAFF))
        {
            return CharacterType.Symbolic;
        }

        // CJK Compatibility Forms
        if ((character >= 0xFE30) && (character <= 0xFE4F))
        {
            return CharacterType.Symbolic;
        }

        // CJK Compatibility Ideographs Supplement
        if ((character >= 0x2F800) && (character <= 0x2FA1F))
        {
            return CharacterType.Symbolic;
        }

        // Greek and Coptic
        if ((character >= 0x0370) && (character <= 0x03FF))
        {
            return CharacterType.Symbolic;
        }

        // Hebrew
        if ((character >= 0x0590) && (character <= 0x05FF))
        {
            return CharacterType.Symbolic;
        }

        // Armenian
        if ((character >= 0x0530) && (character <= 0x058F))
        {
            return CharacterType.Symbolic;
        }

        // LATIN
        //
        // Basic Latin
        if ((character > 0x0000) && (character <= 0x007F))
        {
            return CharacterType.Standard;
        }

        // Latin-1 Supplement
        if ((character >= 0x0080) && (character <= 0x00FF))
        {
            return CharacterType.Standard;
        }

        // Latin Extended-A
        if ((character >= 0x0100) && (character <= 0x017F))
        {
            return CharacterType.Standard;
        }

        // Latin Extended-B
        if ((character >= 0x0180) && (character <= 0x024F))
        {
            return CharacterType.Standard;
        }

        // Latin Extended-C
        if ((character >= 0x2C60) && (character <= 0x2C7F))
        {
            return CharacterType.Standard;
        }

        // Latin Extended-D
        if ((character >= 0xA720) && (character <= 0xA7FF))
        {
            return CharacterType.Standard;
        }

        // Latin Extended-E
        if ((character >= 0xAB30) && (character <= 0xAB6F))
        {
            return CharacterType.Standard;
        }

        // Latin Extended Additional
        if ((character >= 0x1E00) && (character <= 0x1EFF))
        {
            return CharacterType.Standard;
        }

        // Cyrillic
        if ((character >= 0x0400) && (character <= 0x04FF))
        {
            return CharacterType.Standard;
        }

        // Cyrillic Supplement
        if ((character >= 0x0500) && (character <= 0x052F))
        {
            return CharacterType.Standard;
        }

        // Combining Diacritical Marks
        if ((character >= 0x0300) && (character <= 0x036F))
        {
            return CharacterType.Standard;
        }

        return CharacterType.Other;
    }

    /// <summary>
    /// Helper function which takes in a string and returns a vector of pieces, separated by delimiter.
    /// </summary>
    /// <param name="source">String on which to perform the split operation.</param>
    /// <param name="delim">String on which to perform the split operation.</param>
    /// <param name="maxIterations">Maximum number of times to perform a <code>getline</code> loop.</param>
    /// <returns>A vector of pieces from the source string, separated by delimiter</returns>
    private static string[] Split(string source, char delim, int maxIterations = 25)
    {
        return source.Split([delim], maxIterations);
    }

    /// <summary>
    /// Helper function to remove bracket qualifier from the end of a display name if present.
    /// </summary>
    /// <param name="source">String on which to perform the operation.</param>
    /// <returns>A string with the content within brackets removed.</returns>
    private static void StripTrailingBrackets(ref string source)
    {
        // Guidance from the world readiness team is that text within a final set of brackets
        // can be removed for the purposes of calculating initials. ex. John Smith (OSG)
        string[] delimiters = { "{}", "()", "[]" };

        if (source.Length == 0)
        {
            return;
        }

        foreach (var delimiter in delimiters)
        {
            if (source[source.Length - 1] != delimiter[1])
            {
                continue;
            }

            var start = source.LastIndexOf(delimiter[0]);
            if (start == -1)
            {
                continue;
            }

            source = source.Remove(start);
            return;
        }
    }

    /// <summary>
    /// Extracts the first full character from a given string, including any diacritics or combining characters.
    /// </summary>
    /// <param name="str">String from which to extract the character.</param>
    /// <returns>A wstring which represents a given character.</returns>
    private static string GetFirstFullCharacter(string str)
    {
        // Index should begin at the first desireable character.
        int start = 0;

        while (start < str.Length)
        {
            char character = str[start];

            // Omit ! " # $ % & ' ( ) * + , - . /
            if ((character >= 0x0021) && (character <= 0x002F))
            {
                start++;
                continue;
            }

            // Omit : ; < = > ? @
            if ((character >= 0x003A) && (character <= 0x0040))
            {
                start++;
                continue;
            }

            // Omit { | } ~
            if ((character >= 0x007B) && (character <= 0x007E))
            {
                start++;
                continue;
            }

            break;
        }

        // If no desireable characters exist, we'll start at index 1, as combining
        // characters begin only after the first character.
        if (start >= str.Length)
        {
            start = 0;
        }

        // Combining characters begin only after the first character, so we should start
        // looking 1 after the start character.
        int index = start + 1;

        while (index < str.Length)
        {
            char character = str[index];

            // Combining Diacritical Marks -- Official Unicode character block
            if ((character < 0x0300) || (character > 0x036F))
            {
                break;
            }

            index++;
        }

        // Determine number of diacritics by adjusting for our initial offset.
        int strLength = index - start;

        string result = str.SafeSubstring(start, strLength);
        return result;
    }
}

file static class SharedHelpers
{
    public static string SafeSubstring(this string s, int startIndex)
    {
        return s.SafeSubstring(startIndex, s.Length - startIndex);
    }

    public static string SafeSubstring(this string s, int startIndex, int length)
    {
        if (s is null)
        {
            throw new ArgumentNullException(nameof(s));
        }

        if (startIndex > s.Length)
        {
            return string.Empty;
        }

        if (length > s.Length - startIndex)
        {
            length = s.Length - startIndex;
        }

        return s.Substring(startIndex, length);
    }
}