P0 Starter Code (#729)

* P0 Starter Code

* Minor Change to fix in GCC

* More changes to fix

* Fixing format erros

* Fixing Clang-tidy errors

* Tidy fix 2

* Improving Grammar in comments

Author: prashanthduvvada

CMakeLists.txt

@@ -250,10 +250,10 @@ add_custom_target(fix-clang-tidy-diff
 # hardcode some files to check here for each project.
 # ##########################################################
 set(P0_FILES
-        "src/planner/plan_func_call.cpp"
-        "src/include/execution/expressions/string_expression.h"
-        "src/include/primer/orset.h"
-        "src/primer/orset.cpp"
+        "src/include/primer/hyperloglog.h"
+        "src/include/primer/hyperloglog_presto.h"
+        "src/primer/hyperloglog.cpp"
+        "src/primer/hyperloglog_presto.cpp"
 )
 
 add_custom_target(check-clang-tidy-p0

src/include/primer/hyperloglog.h

@@ -0,0 +1,79 @@
+#pragma once
+
+#include <bitset>
+#include <functional>
+#include <memory>
+#include <mutex>  // NOLINT
+#include <string>
+#include <vector>
+
+#define MAX_BITS 64
+#define DEFAULT_CARDINALITY 0
+
+namespace bustub {
+
+/** @brief Hash value type. */
+using hash_t = uint64_t;
+
+/** @brief Constant for HLL. */
+static const double CONSTANT = 0.79402;
+
+template <typename T>
+class HyperLogLog {
+ public:
+  HyperLogLog() = delete;
+
+  explicit HyperLogLog(int16_t n_bits) : cardinality_(0) {}
+
+  /**
+   * @brief Getter value for cardinality.
+   *
+   * @returns cardinality value
+   */
+  auto GetCardinality() { return cardinality_; }
+
+  /**
+   * @brief Ddds a value into the HyperLogLog.
+   *
+   * @param[in] val - value that's added into hyperloglog
+   */
+  auto AddElem(T val) -> void;
+
+  /**
+   * @brief Function that computes cardinality.
+   */
+  auto ComputeCardinality() -> void;
+
+ private:
+  /**
+   * @brief Calculates Hash of a given value.
+   *
+   * @param[in] val - value
+   * @returns hash integer of given input value
+   */
+  inline auto CalculateHash(T val) -> hash_t;
+
+  /**
+   * @brief Function that computes binary.
+   *
+   *
+   * @param[in] hash
+   * @returns binary of a given hash
+   */
+  auto ComputeBinary(const hash_t &hash) const -> std::bitset<MAX_BITS>;
+
+  /**
+   * @brief Function that computes leading zeros.
+   *
+   * @param[in] bset - binary values of a given bitset
+   * @returns leading zeros of given binary set
+   */
+  auto PositionOfLeftmostOne(const std::bitset<MAX_BITS> &bset) const -> uint64_t;
+
+  /** @brief Cardinality value. */
+  size_t cardinality_;
+
+  /** @todo (student) can add their data structures that support HyperLogLog */
+};
+
+}  // namespace bustub

src/include/primer/hyperloglog_presto.h

@@ -0,0 +1,71 @@
+#pragma once
+
+#include <bitset>
+#include <functional>
+#include <memory>
+#include <mutex>  // NOLINT
+#include <string>
+#include <unordered_map>
+#include <vector>
+
+#define BUCKET_SIZE 4UL
+#define CONSTANT 0.79402
+#define MAX_SIZE 7UL
+
+namespace bustub {
+
+template <typename T>
+class HyperLogLogPresto {
+  /**
+   * INSTRUCTIONS: Testing framework will use the GetDenseBucket and GetOverflow function,
+   * hence SHOULD NOT be deleted. It's essential to use the dense_bucket_
+   * data structure.
+   */
+
+  /** @brief Hash type. */
+  using hash_t = uint64_t;
+
+ public:
+  /** @brief Disabling default constructor. */
+  HyperLogLogPresto() = delete;
+
+  /** @brief Parameterized constructor. */
+  explicit HyperLogLogPresto(int16_t n_leading_bits) : cardinality_(0) {}
+
+  /** @brief Returns the dense_bucket_ data structure. */
+  auto GetDenseBucket() const -> std::vector<std::bitset<BUCKET_SIZE>> { return dense_bucket_; }
+
+  /** @brief Returns overflow bucket of a specific given index. */
+  auto GetOverflowBucketofIndex(uint16_t idx) { return overflow_bucket_[idx]; }
+
+  /** @brief Retusn the cardinality of the set. */
+  auto GetCardinality() const -> uint64_t { return cardinality_; }
+
+  /** @brief Element is added for HLL calculation. */
+  auto AddElem(T val) -> void;
+
+  /** @brief Function to compute cardinality. */
+  auto ComputeCardinality() -> void;
+
+ private:
+  /** @brief Calculate Hash.
+   *
+   * @param[in] val
+   *
+   * @returns hash value
+   */
+  inline auto CalculateHash(T val) -> hash_t { return std::hash<T>{}(val); }
+
+  /** @brief Structure holding dense buckets (or also known as registers). */
+  std::vector<std::bitset<BUCKET_SIZE>> dense_bucket_;
+
+  /** @brief Structure holding overflow buckets. */
+  std::unordered_map<uint16_t, std::bitset<MAX_SIZE - BUCKET_SIZE>> overflow_bucket_;
+
+  /** @brief Storing cardinality value */
+  uint64_t cardinality_;
+
+  // TODO(student) - can add more data structures as required
+};
+
+}  // namespace bustub

src/primer/CMakeLists.txt

@@ -1,6 +1,8 @@
 add_library(
   bustub_primer
   OBJECT
+  hyperloglog.cpp
+  hyperloglog_presto.cpp
   orset.cpp
   orset_driver.cpp
   trie.cpp

src/primer/hyperloglog.cpp

@@ -0,0 +1,36 @@
+#include "primer/hyperloglog.h"
+
+namespace bustub {
+
+template <typename T>
+auto inline HyperLogLog<T>::CalculateHash(T val) -> hash_t {
+  /** @todo student -  implement the function */
+  return 0;
+}
+
+template <typename T>
+auto HyperLogLog<T>::ComputeBinary(const hash_t &hash) const -> std::bitset<MAX_BITS> {
+  /** @todo student - implement the function */
+  return {0};
+}
+
+template <typename T>
+auto HyperLogLog<T>::PositionOfLeftmostOne(const std::bitset<MAX_BITS> &bset) const -> uint64_t {
+  /** @todo student - implement the function*/
+  return 0;
+}
+
+template <typename T>
+auto HyperLogLog<T>::AddElem(T val) -> void {
+  /** @todo implement the function */
+}
+
+template <typename T>
+auto HyperLogLog<T>::ComputeCardinality() -> void {
+  /** @todo - student implement the function */
+}
+
+template class HyperLogLog<int64_t>;
+template class HyperLogLog<std::string>;
+
+}  // namespace bustub

src/primer/hyperloglog_presto.cpp

@@ -0,0 +1,17 @@
+#include "primer/hyperloglog_presto.h"
+
+namespace bustub {
+
+template <typename T>
+auto HyperLogLogPresto<T>::AddElem(T val) -> void {
+  /** @todo (student) has to fill the function */
+}
+
+template <typename T>
+auto HyperLogLogPresto<T>::ComputeCardinality() -> void {
+  // TODO(student) - implement the function
+}
+
+template class HyperLogLogPresto<int64_t>;
+template class HyperLogLogPresto<std::string>;
+}  // namespace bustub