be/src/exprs/anyval-util.h

// Licensed to the Apache Software Foundation (ASF) under one
// or more contributor license agreements.  See the NOTICE file
// distributed with this work for additional information
// regarding copyright ownership.  The ASF licenses this file
// to you under the Apache License, Version 2.0 (the
// "License"); you may not use this file except in compliance
// with the License.  You may obtain a copy of the License at
//
//   http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing,
// software distributed under the License is distributed on an
// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
// KIND, either express or implied.  See the License for the
// specific language governing permissions and limitations
// under the License.

#ifndef IMPALA_EXPRS_ANYVAL_UTIL_H
#define IMPALA_EXPRS_ANYVAL_UTIL_H

#include <algorithm>

#include "runtime/date-value.h"
#include "runtime/runtime-state.h"
#include "runtime/string-value.inline.h"
#include "runtime/timestamp-value.h"
#include "udf/udf-internal.h"
#include "util/decimal-util.h"
#include "util/hash-util.h"

namespace impala {

using impala_udf::FunctionContext;
using impala_udf::AnyVal;
using impala_udf::BooleanVal;
using impala_udf::TinyIntVal;
using impala_udf::SmallIntVal;
using impala_udf::IntVal;
using impala_udf::BigIntVal;
using impala_udf::FloatVal;
using impala_udf::DoubleVal;
using impala_udf::TimestampVal;
using impala_udf::StringVal;
using impala_udf::DecimalVal;
using impala_udf::DateVal;

class ObjectPool;

/// Utilities for AnyVals
class AnyValUtil {
 public:
  static uint32_t Hash(const BooleanVal& v, const FunctionContext::TypeDesc&, int seed) {
    return HashUtil::Hash(&v.val, 1, seed);
  }

  static uint32_t Hash(const TinyIntVal& v, const FunctionContext::TypeDesc&, int seed) {
    return HashUtil::Hash(&v.val, 1, seed);
  }

  static uint32_t Hash(const SmallIntVal& v, const FunctionContext::TypeDesc&, int seed) {
    return HashUtil::Hash(&v.val, 2, seed);
  }

  static uint32_t Hash(const IntVal& v, const FunctionContext::TypeDesc&, int seed) {
    return HashUtil::Hash(&v.val, 4, seed);
  }

  static uint32_t Hash(const DateVal& v, const FunctionContext::TypeDesc&, int seed) {
    return HashUtil::Hash(&v.val, 4, seed);
  }

  static uint32_t Hash(const BigIntVal& v, const FunctionContext::TypeDesc&, int seed) {
    return HashUtil::Hash(&v.val, 8, seed);
  }

  static uint32_t Hash(const FloatVal& v, const FunctionContext::TypeDesc&, int seed) {
    return HashUtil::Hash(&v.val, 4, seed);
  }

  static uint32_t Hash(const DoubleVal& v, const FunctionContext::TypeDesc&, int seed) {
    return HashUtil::Hash(&v.val, 8, seed);
  }

  static uint32_t Hash(const StringVal& v, const FunctionContext::TypeDesc&, int seed) {
    return HashUtil::Hash(v.ptr, v.len, seed);
  }

  static uint32_t Hash(const TimestampVal& v, const FunctionContext::TypeDesc&,
      int seed) {
    TimestampValue tv = TimestampValue::FromTimestampVal(v);
    return tv.Hash(seed);
  }

  static uint64_t Hash(const DecimalVal& v, const FunctionContext::TypeDesc& t,
      int64_t seed) {
    DCHECK_GT(t.precision, 0);
    switch (ColumnType::GetDecimalByteSize(t.precision)) {
      case 4: return HashUtil::Hash(&v.val4, 4, seed);
      case 8: return HashUtil::Hash(&v.val8, 8, seed);
      case 16: return HashUtil::Hash(&v.val16, 16, seed);
      default:
        DCHECK(false);
        return 0;
    }
  }

  static uint64_t Hash64(const BooleanVal& v, const FunctionContext::TypeDesc&,
      int64_t seed) {
    return HashUtil::MurmurHash2_64(&v.val, 1, seed);
  }

  static uint64_t Hash64(const TinyIntVal& v, const FunctionContext::TypeDesc&,
      int64_t seed) {
    return HashUtil::MurmurHash2_64(&v.val, 1, seed);
  }

  static uint64_t Hash64(const SmallIntVal& v, const FunctionContext::TypeDesc&,
      int64_t seed) {
    return HashUtil::MurmurHash2_64(&v.val, 2, seed);
  }

  static uint64_t Hash64(const IntVal& v, const FunctionContext::TypeDesc&,
      int64_t seed) {
    return HashUtil::MurmurHash2_64(&v.val, 4, seed);
  }

  static uint64_t Hash64(const DateVal& v, const FunctionContext::TypeDesc&,
      int64_t seed) {
    return HashUtil::MurmurHash2_64(&v.val, 4, seed);
  }

  static uint64_t Hash64(const BigIntVal& v, const FunctionContext::TypeDesc&,
      int64_t seed) {
    return HashUtil::MurmurHash2_64(&v.val, 8, seed);
  }

  static uint64_t Hash64(const FloatVal& v, const FunctionContext::TypeDesc&,
      int64_t seed) {
    return HashUtil::MurmurHash2_64(&v.val, 4, seed);
  }

  static uint64_t Hash64(const DoubleVal& v, const FunctionContext::TypeDesc&,
      int64_t seed) {
    return HashUtil::MurmurHash2_64(&v.val, 8, seed);
  }

  static uint64_t Hash64(const StringVal& v, const FunctionContext::TypeDesc&,
      int64_t seed) {
    return HashUtil::MurmurHash2_64(v.ptr, v.len, seed);
  }

  static uint64_t Hash64(const TimestampVal& v, const FunctionContext::TypeDesc&,
      int64_t seed) {
    TimestampValue tv = TimestampValue::FromTimestampVal(v);
    return HashUtil::MurmurHash2_64(&tv, 12, seed);
  }

  static uint64_t Hash64(
      const DecimalVal& v, const FunctionContext::TypeDesc& t, int64_t seed) {
    return HashDecimal64(v, ColumnType::GetDecimalByteSize(t.precision), seed);
  }

  static uint64_t HashDecimal64(const DecimalVal& v, int byte_size, int64_t seed) {
    switch (byte_size) {
      case 4:
        return HashUtil::MurmurHash2_64(&v.val4, 4, seed);
      case 8:
        return HashUtil::MurmurHash2_64(&v.val8, 8, seed);
      case 16:
        return HashUtil::MurmurHash2_64(&v.val16, 16, seed);
      default:
        DCHECK(false);
        return 0;
    }
  }

  /// Templated equality functions. These assume the input values are not NULL.
  template<typename T>
  static inline bool Equals(const ColumnType& type, const T& x, const T& y) {
    return EqualsInternal(x, y);
  }

  template<typename T>
  static inline bool Equals(const FunctionContext::TypeDesc& type, const T& x,
      const T& y) {
    return EqualsInternal(x, y);
  }

  /// Returns the byte size of *Val for type t.
  static int AnyValSize(const ColumnType& t) {
    switch (t.type) {
      case TYPE_BOOLEAN: return sizeof(BooleanVal);
      case TYPE_TINYINT: return sizeof(TinyIntVal);
      case TYPE_SMALLINT: return sizeof(SmallIntVal);
      case TYPE_INT: return sizeof(IntVal);
      case TYPE_BIGINT: return sizeof(BigIntVal);
      case TYPE_FLOAT: return sizeof(FloatVal);
      case TYPE_DOUBLE: return sizeof(DoubleVal);
      case TYPE_STRING:
      case TYPE_VARCHAR:
      case TYPE_CHAR:
      case TYPE_FIXED_UDA_INTERMEDIATE:
        return sizeof(StringVal);
      case TYPE_TIMESTAMP: return sizeof(TimestampVal);
      case TYPE_DECIMAL: return sizeof(DecimalVal);
      case TYPE_DATE: return sizeof(DateVal);
      default:
        DCHECK(false) << t;
        return 0;
    }
  }

  /// Returns the byte alignment of *Val for type t.
  static int AnyValAlignment(const ColumnType& t) {
    switch (t.type) {
      case TYPE_BOOLEAN: return alignof(BooleanVal);
      case TYPE_TINYINT: return alignof(TinyIntVal);
      case TYPE_SMALLINT: return alignof(SmallIntVal);
      case TYPE_INT: return alignof(IntVal);
      case TYPE_BIGINT: return alignof(BigIntVal);
      case TYPE_FLOAT: return alignof(FloatVal);
      case TYPE_DOUBLE: return alignof(DoubleVal);
      case TYPE_STRING:
      case TYPE_VARCHAR:
      case TYPE_CHAR:
      case TYPE_FIXED_UDA_INTERMEDIATE:
        return alignof(StringVal);
      case TYPE_TIMESTAMP: return alignof(TimestampVal);
      case TYPE_DECIMAL: return alignof(DecimalVal);
      case TYPE_DATE: return alignof(DateVal);
      default:
        DCHECK(false) << t;
        return 0;
    }
  }

  static std::string ToString(const StringVal& v) {
    return std::string(reinterpret_cast<char*>(v.ptr), v.len);
  }

  static StringVal FromString(FunctionContext* ctx, const std::string& s) {
    StringVal val = FromBuffer(ctx, s.c_str(), s.size());
    return val;
  }

  static void TruncateIfNecessary(const FunctionContext::TypeDesc& type, StringVal *val) {
    if (type.type == FunctionContext::TYPE_VARCHAR) {
      DCHECK(type.len >= 0);
      val->len = std::min(val->len, type.len);
    }
  }

  static StringVal FromBuffer(FunctionContext* ctx, const char* ptr, int len) {
    return StringVal::CopyFrom(ctx, reinterpret_cast<const uint8_t*>(ptr), len);
  }

  static FunctionContext::TypeDesc ColumnTypeToTypeDesc(const ColumnType& type);
  static std::vector<FunctionContext::TypeDesc> ColumnTypesToTypeDescs(
      const std::vector<ColumnType>& types);
  // Note: constructing a ColumnType is expensive and should be avoided in query execution
  // paths (i.e. non-setup paths).
  static ColumnType TypeDescToColumnType(const FunctionContext::TypeDesc& type);

  /// Utility to put val into an AnyVal struct
  static void SetAnyVal(const void* slot, const ColumnType& type, AnyVal* dst) {
    if (slot == NULL) {
      dst->is_null = true;
      return;
    }

    dst->is_null = false;
    switch (type.type) {
      case TYPE_NULL: return;
      case TYPE_BOOLEAN:
        reinterpret_cast<BooleanVal*>(dst)->val = *reinterpret_cast<const bool*>(slot);
        return;
      case TYPE_TINYINT:
        reinterpret_cast<TinyIntVal*>(dst)->val = *reinterpret_cast<const int8_t*>(slot);
        return;
      case TYPE_SMALLINT:
        reinterpret_cast<SmallIntVal*>(dst)->val = *reinterpret_cast<const int16_t*>(slot);
        return;
      case TYPE_INT:
        reinterpret_cast<IntVal*>(dst)->val = *reinterpret_cast<const int32_t*>(slot);
        return;
      case TYPE_BIGINT:
        reinterpret_cast<BigIntVal*>(dst)->val = *reinterpret_cast<const int64_t*>(slot);
        return;
      case TYPE_FLOAT:
        reinterpret_cast<FloatVal*>(dst)->val = *reinterpret_cast<const float*>(slot);
        return;
      case TYPE_DOUBLE:
        reinterpret_cast<DoubleVal*>(dst)->val = *reinterpret_cast<const double*>(slot);
        return;
      case TYPE_STRING:
      case TYPE_VARCHAR:
        reinterpret_cast<const StringValue*>(slot)->ToStringVal(
            reinterpret_cast<StringVal*>(dst));
        if (type.type == TYPE_VARCHAR) {
          StringVal* sv = reinterpret_cast<StringVal*>(dst);
          DCHECK_GE(type.len, 0);
          DCHECK_LE(sv->len, type.len);
        }
        return;
      case TYPE_CHAR:
      case TYPE_FIXED_UDA_INTERMEDIATE: {
        StringVal* sv = reinterpret_cast<StringVal*>(dst);
        sv->ptr = const_cast<uint8_t*>(reinterpret_cast<const uint8_t*>(slot));
        sv->len = type.len;
        return;
      }
      case TYPE_TIMESTAMP:
        reinterpret_cast<const TimestampValue*>(slot)->ToTimestampVal(
            reinterpret_cast<TimestampVal*>(dst));
        return;
      case TYPE_DECIMAL:
        switch (type.GetByteSize()) {
          case 4:
            reinterpret_cast<DecimalVal*>(dst)->val4 =
                *reinterpret_cast<const int32_t*>(slot);
            return;
          case 8:
            reinterpret_cast<DecimalVal*>(dst)->val8 =
                *reinterpret_cast<const int64_t*>(slot);
            return;
#if __BYTE_ORDER == __LITTLE_ENDIAN
          case 16:
            memcpy(&reinterpret_cast<DecimalVal*>(dst)->val16, slot, 16);
#else
            DCHECK(false) << "Not implemented.";
#endif
            return;
          default:
            break;
        }
      case TYPE_DATE:
        *reinterpret_cast<DateVal*>(dst) =
            reinterpret_cast<const DateValue*>(slot)->ToDateVal();
        return;
      default:
        DCHECK(false) << "NYI: " << type;
    }
  }

 private:
  /// Implementations of Equals().
  template <typename T>
  static inline bool EqualsInternal(const T& x, const T& y);
  static inline bool DecimalEquals(
      int precision, const DecimalVal& x, const DecimalVal& y);
};

/// Allocates an AnyVal subclass of 'type' from 'pool'. The AnyVal's memory is
/// initialized to all 0's. Returns a MemLimitExceeded() error with message
/// 'mem_limit_exceeded_msg' if the allocation cannot be made because of a memory
/// limit.
Status AllocateAnyVal(RuntimeState* state, MemPool* pool, const ColumnType& type,
    const std::string& mem_limit_exceeded_msg, AnyVal** result);

template <typename T>
inline bool AnyValUtil::EqualsInternal(const T& x, const T& y) {
  DCHECK(!x.is_null);
  DCHECK(!y.is_null);
  return x.val == y.val;
}

template<> inline bool AnyValUtil::EqualsInternal(const StringVal& x,
    const StringVal& y) {
  DCHECK(!x.is_null);
  DCHECK(!y.is_null);
  StringValue x_sv = StringValue::FromStringVal(x);
  StringValue y_sv = StringValue::FromStringVal(y);
  return x_sv == y_sv;
}

template<> inline bool AnyValUtil::EqualsInternal(const TimestampVal& x,
    const TimestampVal& y) {
  DCHECK(!x.is_null);
  DCHECK(!y.is_null);
  TimestampValue x_tv = TimestampValue::FromTimestampVal(x);
  TimestampValue y_tv = TimestampValue::FromTimestampVal(y);
  return x_tv == y_tv;
}

template<> inline bool AnyValUtil::Equals(const ColumnType& type, const DecimalVal& x,
    const DecimalVal& y) {
  return DecimalEquals(type.precision, x, y);
}

template<> inline bool AnyValUtil::Equals(const FunctionContext::TypeDesc& type,
    const DecimalVal& x, const DecimalVal& y) {
  return DecimalEquals(type.precision, x, y);
}

inline bool AnyValUtil::DecimalEquals(int precision, const DecimalVal& x,
    const DecimalVal& y) {
  DCHECK(!x.is_null);
  DCHECK(!y.is_null);
  if (precision <= ColumnType::MAX_DECIMAL4_PRECISION) {
    return x.val4 == y.val4;
  } else if (precision <= ColumnType::MAX_DECIMAL8_PRECISION) {
    return x.val8 == y.val8;
  } else {
    return x.val16 == y.val16;
  }
}

}

#endif