From e7c05732cdac9a8eda8f7bd63ae0b0049896d130 Mon Sep 17 00:00:00 2001
From: Andrey Zvonov <zvonov.andrey@gmail.com>
Date: Fri, 26 Dec 2025 21:07:07 +0100
Subject: [PATCH] attempt to fix

---
 src/Processors/Formats/Impl/Parquet/Decoding.cpp    |  2 ++
 .../DataLakes/Iceberg/IcebergWrites.cpp             |  9 +++++++++
 .../DataLakes/Iceberg/ManifestFile.cpp              | 13 ++++++++++++-
 3 files changed, 23 insertions(+), 1 deletion(-)

diff --git a/src/Processors/Formats/Impl/Parquet/Decoding.cpp b/src/Processors/Formats/Impl/Parquet/Decoding.cpp
index 060d14f2cefc..e667182d0c89 100644
--- a/src/Processors/Formats/Impl/Parquet/Decoding.cpp
+++ b/src/Processors/Formats/Impl/Parquet/Decoding.cpp
@@ -1096,6 +1096,8 @@ void IntConverter::convertField(std::span<const char> data, bool /*is_max*/, Fie
     UInt64 val = 0;
     switch (input_size)
     {
+        case 1: val = unalignedLoad<UInt8>(data.data()); break;
+        case 2: val = unalignedLoad<UInt16>(data.data()); break;
         case 4: val = unalignedLoad<UInt32>(data.data()); break;
         case 8: val = unalignedLoad<UInt64>(data.data()); break;
         default: chassert(false);
diff --git a/src/Storages/ObjectStorage/DataLakes/Iceberg/IcebergWrites.cpp b/src/Storages/ObjectStorage/DataLakes/Iceberg/IcebergWrites.cpp
index 75fe44315faa..dce4b87c58a3 100644
--- a/src/Storages/ObjectStorage/DataLakes/Iceberg/IcebergWrites.cpp
+++ b/src/Storages/ObjectStorage/DataLakes/Iceberg/IcebergWrites.cpp
@@ -116,6 +116,7 @@ bool canDumpIcebergStats(const Field & field, DataTypePtr type)
         case TypeIndex::Int64:
         case TypeIndex::DateTime64:
         case TypeIndex::String:
+        case TypeIndex::UInt8: /// Boolean
             return true;
         default:
             return false;
@@ -156,6 +157,14 @@ std::vector<uint8_t> dumpFieldToBytes(const Field & field, DataTypePtr type)
             return dumpValue(field.safeGet<Float64>());
         case TypeIndex::Float32:
             return dumpValue(field.safeGet<Float32>());
+        case TypeIndex::UInt8: /// Boolean - stored as single byte in Iceberg
+        {
+            /// Field can be Bool or UInt64 type depending on source
+            UInt8 value = (field.getType() == Field::Types::Bool)
+                ? static_cast<UInt8>(field.safeGet<bool>())
+                : static_cast<UInt8>(field.safeGet<UInt64>());
+            return dumpValue(value);
+        }
         default:
         {
             throw Exception(ErrorCodes::LOGICAL_ERROR, "Can not dump such stats");
diff --git a/src/Storages/ObjectStorage/DataLakes/Iceberg/ManifestFile.cpp b/src/Storages/ObjectStorage/DataLakes/Iceberg/ManifestFile.cpp
index d5c50a6c9994..82ab834e6a48 100644
--- a/src/Storages/ObjectStorage/DataLakes/Iceberg/ManifestFile.cpp
+++ b/src/Storages/ObjectStorage/DataLakes/Iceberg/ManifestFile.cpp
@@ -116,9 +116,20 @@ namespace
                 return std::nullopt;
             }
         }
+        else if (DB::isBool(non_nullable_type))
+        {
+            /// Boolean type needs special handling:
+            /// When UInt8 is extracted from column to Field, it becomes Field(Types::UInt64).
+            /// But query conditions like WHERE bool_col = true create Field(Types::Bool).
+            /// While accurateLess/accurateEquals handle type differences correctly,
+            /// we create a Bool-typed Field for consistency.
+            if (str.empty())
+                return std::nullopt;
+            return static_cast<bool>(static_cast<unsigned char>(str[0]) != 0);
+        }
         else
         {
-            /// For all other types except decimal binary representation
+            /// For all other types except decimal and boolean, binary representation
             /// matches our internal representation
             column->insertData(str.data(), str.length());
             DB::Field result;