Fix emoji encoding in recorded .har files (#1557)

* Initial plan

* Fix emoji encoding in recorded .har files

Add UnsafeRelaxedJsonEscaping to ProxyUtils.JsonSerializerOptions to
preserve BMP non-ASCII characters as literal UTF-8 in JSON output.

Add surrogate pair unescaping in HarGeneratorPlugin to convert
\uXXXX\uYYYY escape sequences for non-BMP characters (like emoji
U+1F600+) back to their literal UTF-8 representation.

Co-authored-by: waldekmastykarz <11164679+waldekmastykarz@users.noreply.github.com>

---------

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: waldekmastykarz <11164679+waldekmastykarz@users.noreply.github.com>
Co-authored-by: Waldek Mastykarz <waldek@mastykarz.nl>
Co-authored-by: Garry Trinder <garry@trinder365.co.uk>

Author: 3 people

DevProxy.Abstractions/Utils/ProxyUtils.cs

@@ -9,6 +9,7 @@
 using Newtonsoft.Json.Schema;
 using System.Collections.ObjectModel;
 using System.Reflection;
+using System.Text.Encodings.Web;
 using System.Text.Json;
 using System.Text.Json.Serialization;
 using System.Text.RegularExpressions;
@@ -44,6 +45,7 @@ public static class ProxyUtils
     public static JsonSerializerOptions JsonSerializerOptions { get; } = new()
     {
         DefaultIgnoreCondition = JsonIgnoreCondition.WhenWritingNull,
+        Encoder = JavaScriptEncoder.UnsafeRelaxedJsonEscaping,
         PropertyNamingPolicy = JsonNamingPolicy.CamelCase,
         PropertyNameCaseInsensitive = true,
         ReadCommentHandling = JsonCommentHandling.Skip,

DevProxy.Plugins/Generation/HarGeneratorPlugin.cs

@@ -9,7 +9,9 @@
 using Microsoft.Extensions.Configuration;
 using Microsoft.Extensions.Logging;
 using System.Diagnostics;
+using System.Globalization;
 using System.Text.Json;
+using System.Text.RegularExpressions;
 using System.Web;
 
 namespace DevProxy.Plugins.Generation;
@@ -33,6 +35,8 @@ public sealed class HarGeneratorPlugin(
         proxyConfiguration,
         pluginConfigurationSection)
 {
+    private static readonly Regex surrogatePairRegex = new(@"\\u([dD][89aAbB][0-9a-fA-F]{2})\\u([dD][cCdDeEfF][0-9a-fA-F]{2})");
+
     public override string Name => nameof(HarGeneratorPlugin);
 
     public override async Task AfterRecordingStopAsync(RecordingArgs e, CancellationToken cancellationToken)
@@ -69,6 +73,7 @@ r.Context.Session is not null &&
 
         Logger.LogDebug("Serializing HAR file...");
         var harFileJson = JsonSerializer.Serialize(harFile, ProxyUtils.JsonSerializerOptions);
+        harFileJson = UnescapeSurrogatePairs(harFileJson);
         var fileName = $"devproxy-{DateTime.Now:yyyyMMddHHmmss}.har";
 
         Logger.LogDebug("Writing HAR file to {FileName}...", fileName);
@@ -159,4 +164,15 @@ private HarEntry CreateHarEntry(RequestLog log)
 
         return entry;
     }
+
+    private static string UnescapeSurrogatePairs(string json)
+    {
+        return surrogatePairRegex.Replace(json, match =>
+        {
+            var high = int.Parse(match.Groups[1].Value, NumberStyles.HexNumber, CultureInfo.InvariantCulture);
+            var low = int.Parse(match.Groups[2].Value, NumberStyles.HexNumber, CultureInfo.InvariantCulture);
+            var codePoint = 0x10000 + ((high - 0xD800) << 10) + (low - 0xDC00);
+            return char.ConvertFromUtf32(codePoint);
+        });
+    }
 }