From edefbd1606b469fa63768ad07d3b6da397738715 Mon Sep 17 00:00:00 2001
From: Jason Malinowski <jason.malinowski@microsoft.com>
Date: Fri, 29 Jul 2022 17:51:23 -0700
Subject: [PATCH] Hash the generator assembly and type names

Since the names of the generator/type may reveal sensitive things
(company names), we'll hash this; this is sharing the same code
we use for analyzers.
---
 .../DiagnosticAnalyzerTelemetry.cs            |  8 +-------
 .../Portable/Log/AnalyzerNameForTelemetry.cs  | 20 +++++++++++++++++++
 ...ratorTelemetryCollectorWorkspaceService.cs |  5 +++--
 3 files changed, 24 insertions(+), 9 deletions(-)
 create mode 100644 src/Workspaces/Core/Portable/Log/AnalyzerNameForTelemetry.cs

diff --git a/src/Features/Core/Portable/Diagnostics/DiagnosticAnalyzerTelemetry.cs b/src/Features/Core/Portable/Diagnostics/DiagnosticAnalyzerTelemetry.cs
index d18da21e37fa4..0d3e5b4dbe106 100644
--- a/src/Features/Core/Portable/Diagnostics/DiagnosticAnalyzerTelemetry.cs
+++ b/src/Features/Core/Portable/Diagnostics/DiagnosticAnalyzerTelemetry.cs
@@ -106,7 +106,7 @@ public void ReportAndClear(int correlationId)
                     else
                     {
                         // annonymize analyzer and exception names:
-                        m["Analyzer.NameHashCode"] = ComputeSha256Hash(analyzerName);
+                        m["Analyzer.NameHashCode"] = AnalyzerNameForTelemetry.ComputeSha256Hash(analyzerName);
                     }
 
                     m["Analyzer.CodeBlock"] = analyzerInfo.CodeBlockActionsCount;
@@ -130,11 +130,5 @@ public void ReportAndClear(int correlationId)
                 }));
             }
         }
-
-        private static string ComputeSha256Hash(string name)
-        {
-            using var sha256 = SHA256.Create();
-            return Convert.ToBase64String(sha256.ComputeHash(Encoding.UTF8.GetBytes(name)));
-        }
     }
 }
diff --git a/src/Workspaces/Core/Portable/Log/AnalyzerNameForTelemetry.cs b/src/Workspaces/Core/Portable/Log/AnalyzerNameForTelemetry.cs
new file mode 100644
index 0000000000000..c6e613fa4c5ac
--- /dev/null
+++ b/src/Workspaces/Core/Portable/Log/AnalyzerNameForTelemetry.cs
@@ -0,0 +1,20 @@
+// Licensed to the .NET Foundation under one or more agreements.
+// The .NET Foundation licenses this file to you under the MIT license.
+// See the LICENSE file in the project root for more information.
+
+using System;
+using System.Collections.Generic;
+using System.Security.Cryptography;
+using System.Text;
+
+namespace Microsoft.CodeAnalysis.Internal.Log
+{
+    internal static class AnalyzerNameForTelemetry
+    {
+        public static string ComputeSha256Hash(string name)
+        {
+            using var sha256 = SHA256.Create();
+            return Convert.ToBase64String(sha256.ComputeHash(Encoding.UTF8.GetBytes(name)));
+        }
+    }
+}
diff --git a/src/Workspaces/Core/Portable/SourceGeneratorTelemetry/SourceGeneratorTelemetryCollectorWorkspaceService.cs b/src/Workspaces/Core/Portable/SourceGeneratorTelemetry/SourceGeneratorTelemetryCollectorWorkspaceService.cs
index 2eca052d40aff..b17ac9e0f8dd6 100644
--- a/src/Workspaces/Core/Portable/SourceGeneratorTelemetry/SourceGeneratorTelemetryCollectorWorkspaceService.cs
+++ b/src/Workspaces/Core/Portable/SourceGeneratorTelemetry/SourceGeneratorTelemetryCollectorWorkspaceService.cs
@@ -56,9 +56,10 @@ public void ReportStatisticsAndClear(FunctionId functionId)
                 // We'll log one event per generator
                 Logger.Log(functionId, KeyValueLogMessage.Create(map =>
                 {
-                    map[nameof(telemetryKey.Identity.AssemblyName)] = telemetryKey.Identity.AssemblyName;
+                    // TODO: have a policy for when we don't have to hash them
+                    map[nameof(telemetryKey.Identity.AssemblyName) + "Hashed"] = AnalyzerNameForTelemetry.ComputeSha256Hash(telemetryKey.Identity.AssemblyName);
                     map[nameof(telemetryKey.Identity.AssemblyVersion)] = telemetryKey.Identity.AssemblyVersion.ToString();
-                    map[nameof(telemetryKey.Identity.TypeName)] = telemetryKey.Identity.TypeName;
+                    map[nameof(telemetryKey.Identity.TypeName) + "Hashed"] = AnalyzerNameForTelemetry.ComputeSha256Hash(telemetryKey.Identity.TypeName);
                     map[nameof(telemetryKey.FileVersion)] = telemetryKey.FileVersion;
 
                     var result = elapsedTimeCounter.GetStatisticResult();