feat(DbExporter): implement verify and verify-full

Add Verifier class that reads exported .pb.zstd files and: - Deserializes protobuf data through zstd decompression - Extracts schema information (column names and types) - Counts rows for verification - Optionally computes SHA256 hash and compares against sidecar file
2026-01-06 16:55:18 -05:00
parent 6f58bfd8cc
commit ba11407be4
1 changed files with 79 additions and 0 deletions
@@ -0,0 +1,79 @@
+using System.Data;
+using System.Security.Cryptography;
+using System.Text;
+using ProtoBuf.Data;
+using ZstdSharp;
+
+namespace DbExporter;
+
+public sealed class Verifier
+{
+    public record VerifyResult(int RowCount, List<ColumnInfo> Schema, string? ComputedHash, string? ExpectedHash, bool? HashMatch);
+    public record ColumnInfo(string Name, Type Type);
+
+    public VerifyResult Verify(string filePath, bool computeHash = false)
+    {
+        using var inputFile = new FileStream(filePath, FileMode.Open, FileAccess.Read, FileShare.Read, 256 * 1024);
+        using var decompressStream = new DecompressionStream(inputFile);
+
+        Stream readStream = decompressStream;
+        SHA256? sha256 = null;
+        CryptoStream? hashStream = null;
+
+        if (computeHash)
+        {
+            sha256 = SHA256.Create();
+            hashStream = new CryptoStream(decompressStream, sha256, CryptoStreamMode.Read);
+            readStream = hashStream;
+        }
+
+        using var reader = DataSerializer.Deserialize(readStream);
+
+        // Extract schema
+        var schema = new List<ColumnInfo>();
+        for (int i = 0; i < reader.FieldCount; i++)
+        {
+            schema.Add(new ColumnInfo(reader.GetName(i), reader.GetFieldType(i)));
+        }
+
+        // Count rows
+        int rowCount = 0;
+        while (reader.Read())
+        {
+            rowCount++;
+        }
+
+        string? computedHashStr = null;
+        string? expectedHash = null;
+        bool? hashMatch = null;
+
+        if (computeHash && sha256 != null)
+        {
+            hashStream?.Dispose();
+            computedHashStr = Convert.ToHexString(sha256.Hash!).ToLowerInvariant();
+
+            // Read expected hash from sidecar
+            var hashFilePath = filePath + ".sha256";
+            if (File.Exists(hashFilePath))
+            {
+                expectedHash = File.ReadAllText(hashFilePath).Trim().ToLowerInvariant();
+                hashMatch = computedHashStr == expectedHash;
+            }
+
+            sha256.Dispose();
+        }
+
+        return new VerifyResult(rowCount, schema, computedHashStr, expectedHash, hashMatch);
+    }
+
+    public string FormatSchema(List<ColumnInfo> schema)
+    {
+        var sb = new StringBuilder();
+        foreach (var col in schema)
+        {
+            if (sb.Length > 0) sb.Append(", ");
+            sb.Append($"{col.Name} ({col.Type.Name})");
+        }
+        return sb.ToString();
+    }
+}