CalculatePublisherIdHash.cs

using System;
using System.Linq;
using System.Security.Cryptography;
using System.Text;

namespace Otor
{
    public class MsixUtils
    {
        public static string GetPublisherHash(string publisherId)
        {
            using var sha = HashAlgorithm.Create("SHA256");
            var encoded = sha.ComputeHash(Encoding.Unicode.GetBytes(publisherId));
            var binaryString = string.Concat(encoded.Take(8).Select(c => Convert.ToString(c, 2).PadLeft(8, '0'))) + '0'; // representing 65-bits = 13 * 5
            var encodedPublisherId = string.Concat(Enumerable.Range(0, binaryString.Length / 5).Select(i => "0123456789abcdefghjkmnpqrstvwxyz".Substring(Convert.ToInt32(binaryString.Substring(i * 5, 5), 2), 1)));
            return encodedPublisherId;
        }
    }
}

This is a compact algorithm that given a publisher ID calculates its hash for APPX-management purposes.

The algorithm:

1. Takes an UTF-16 encoded string and calculates its SHA-256 hash
2. It takes the first 8-bytes of the checksum
3. For each byte, it converts it to a binary representation (8 characters)
4. Since this gives 8 x 8 = 64 characters, an extra zero ('0') is added to have a string of a length divisible by 5
5. Then an algorithm processes the bytes in chunk of 5 characters.
   • For each chunk, the 5-character string is converted to an integer (since 2^5 = 32, it produces a number from 0 to 31).
   • It then takes the n-th character of a string 0123456789abcdefghjkmnpqrstvwxyz (32 letters), where n is the result of the previous operation.
6. Finally, it joins all produced letters, which gives a string of 13 characters

Steps 5 and 6 actually represent the Douglas Crockford Base32 algorithm

Sample input:

E=marcin@otorowski.com, CN=Marcin Otorowski, O=Marcin Otorowski, S=zachodniopomorskie, C=PL

Output:

zxq1da1qqbeze

Verification:

Thank you, it works for me!
Migrated to Python:

import hashlib

def calculate_publisher_hash(publisher_id):
    encoded = hashlib.sha256(publisher_id.encode("utf-16-le")).digest()

    binary_string = "".join(f"{b:08b}" for b in encoded[:8]) + "0"  # 65 bits = 13 * 5

    alphabet = "0123456789abcdefghjkmnpqrstvwxyz"
    return "".join(
        alphabet[int(binary_string[i * 5:(i + 1) * 5], 2)]
        for i in range(len(binary_string) // 5)
    )

It can also be done with the use of available APIs:

#include <appmodel.h>

HRESULT PublisherIdFromPublisher(
    _In_ PCWSTR publisher,
    _Out_writes_(PACKAGE_PUBLISHERID_MAX_LENGTH + 1) PWSTR publisherId)
{
    PCWSTR name{ L"xyz" };
    const size_t nameLength{ ARRAYSIZE(L"xyz") - 1 };
    const size_t offsetToPublisherId{ name + 1 }; // xyz_...publisherid...
    PACKAGE_ID id{};
    id.name = name;
    id.publisher = publisher;
 
    WCHAR familyName[PACKAGE_PUBLISHERID_MAX_LENGTH + 1]{};
    UINT32 n{ ARRAYSIZE(familyName) };
    RETURN_IF_WIN32_ERROR(PackageFamilyNameFromId(&id, &n, familyName);
    RETURN_IF_FAILED(StringCchCopyW(publisherId, PACKAGE_PUBLISHERID_MAX_LENGTH + 1, familyName + offsetToPublisherId));
    return S_OK;
}

see: https://learn.microsoft.com/en-us/windows/apps/desktop/modernize/package-identity-overview