Fix parquet endian issue for s390x

ClickHouse · Sep 19, 2023 · e40416e · e40416e
1 parent 9d9c464
commit e40416e
Show file tree

Hide file tree

Showing 3 changed files with 20 additions and 9 deletions.
diff --git a/cpp/src/parquet/column_reader.cc b/cpp/src/parquet/column_reader.cc
@@ -118,7 +118,7 @@ int LevelDecoder::SetData(Encoding::type encoding, int16_t max_level,
       if (data_size < 4) {
         throw ParquetException("Received invalid levels (corrupt data page?)");
       }
-      num_bytes = ::arrow::util::SafeLoadAs<int32_t>(data);
+      num_bytes = ::arrow::bit_util::ToLittleEndian(::arrow::util::SafeLoadAs<int32_t>(data));
       if (num_bytes < 0 || num_bytes > data_size - 4) {
         throw ParquetException("Received invalid number of bytes (corrupt data page?)");
       }

diff --git a/cpp/src/parquet/encoding.cc b/cpp/src/parquet/encoding.cc
@@ -1028,7 +1028,7 @@ int PlainDecoder<DType>::DecodeArrow(
   VisitNullBitmapInline(
       valid_bits, valid_bits_offset, num_values, null_count,
       [&]() {
-        builder->UnsafeAppend(SafeLoadAs<value_type>(data_));
+        builder->UnsafeAppend(::arrow::bit_util::ToLittleEndian(SafeLoadAs<value_type>(data_)));
         data_ += sizeof(value_type);
       },
       [&]() { builder->UnsafeAppendNull(); });
@@ -1055,7 +1055,8 @@ int PlainDecoder<DType>::DecodeArrow(
   VisitNullBitmapInline(
       valid_bits, valid_bits_offset, num_values, null_count,
       [&]() {
-        PARQUET_THROW_NOT_OK(builder->Append(SafeLoadAs<value_type>(data_)));
+        PARQUET_THROW_NOT_OK(
+            builder->Append(::arrow::bit_util::ToLittleEndian(SafeLoadAs<value_type>(data_))));
         data_ += sizeof(value_type);
       },
       [&]() { PARQUET_THROW_NOT_OK(builder->AppendNull()); });
@@ -1075,7 +1076,17 @@ inline int DecodePlain(const uint8_t* data, int64_t data_size, int num_values,
   }
   // If bytes_to_decode == 0, data could be null
   if (bytes_to_decode > 0) {
+#if defined(__BYTE_ORDER__) && defined(__ORDER_BIG_ENDIAN__)
+    for (size_t i = 0; i < num_values; ++i)
+    {
+      memcpy(out + i, data + sizeof(T) * i, sizeof(T));
+      auto begin = reinterpret_cast<uint8_t*>(out + i);
+      auto end = begin + sizeof(T);
+      std::reverse(begin, end);
+    }
+#else
     memcpy(out, data, bytes_to_decode);
+#endif
   }
   return static_cast<int>(bytes_to_decode);
 }
@@ -1098,7 +1109,7 @@ static inline int64_t ReadByteArray(const uint8_t* data, int64_t data_size,
   if (ARROW_PREDICT_FALSE(data_size < 4)) {
     ParquetException::EofException();
   }
-  const int32_t len = SafeLoadAs<int32_t>(data);
+  const int32_t len = ::arrow::bit_util::ToLittleEndian(SafeLoadAs<int32_t>(data));
   if (len < 0) {
     throw ParquetException("Invalid BYTE_ARRAY value");
   }
@@ -1387,7 +1398,7 @@ class PlainByteArrayDecoder : public PlainDecoder<ByteArrayType>,
           if (ARROW_PREDICT_FALSE(len_ < 4)) {
             ParquetException::EofException();
           }
-          auto value_len = SafeLoadAs<int32_t>(data_);
+          auto value_len = ::arrow::bit_util::ToLittleEndian(SafeLoadAs<int32_t>(data_));
           if (ARROW_PREDICT_FALSE(value_len < 0 || value_len > INT32_MAX - 4)) {
             return Status::Invalid("Invalid or corrupted value_len '", value_len, "'");
           }
@@ -1433,7 +1444,7 @@ class PlainByteArrayDecoder : public PlainDecoder<ByteArrayType>,
           if (ARROW_PREDICT_FALSE(len_ < 4)) {
             ParquetException::EofException();
           }
-          auto value_len = SafeLoadAs<int32_t>(data_);
+          auto value_len = ::arrow::bit_util::ToLittleEndian(SafeLoadAs<int32_t>(data_));
           if (ARROW_PREDICT_FALSE(value_len < 0 || value_len > INT32_MAX - 4)) {
             return Status::Invalid("Invalid or corrupted value_len '", value_len, "'");
           }
@@ -3312,7 +3323,7 @@ int ByteStreamSplitDecoder<DType>::DecodeArrow(
           const size_t byte_index = b * num_values_in_buffer_ + offset;
           gathered_byte_data[b] = data[byte_index];
         }
-        builder->UnsafeAppend(SafeLoadAs<T>(&gathered_byte_data[0]));
+        builder->UnsafeAppend(::arrow::bit_util::ToLittleEndian(SafeLoadAs<T>(&gathered_byte_data[0])));
         ++offset;
       },
       [&]() { builder->UnsafeAppendNull(); });

diff --git a/cpp/src/parquet/file_reader.cc b/cpp/src/parquet/file_reader.cc
@@ -473,9 +473,9 @@ class SerializedFile : public ParquetFileReader::Contents {
           "is not a parquet file.");
     }
     // Both encrypted/unencrypted footers have the same footer length check.
-    uint32_t metadata_len = ::arrow::util::SafeLoadAs<uint32_t>(
+    uint32_t metadata_len = ::arrow::bit_util::FromLittleEndian(::arrow::util::SafeLoadAs<uint32_t>(
         reinterpret_cast<const uint8_t*>(footer_buffer->data()) + footer_read_size -
-        kFooterSize);
+        kFooterSize));
     if (metadata_len > source_size_ - kFooterSize) {
       throw ParquetInvalidOrCorruptedFileException(
           "Parquet file size is ", source_size_,