Added support for Ubuntu 24.04. Dropped Ubuntu 18.04

.github/workflows/cmake.yml

@@ -19,27 +19,33 @@ jobs:
       matrix:
         config:
           - {
-              name: "Ubuntu 22.04 GCC",
-              os: "ubuntu-22.04",
+              name: "Ubuntu 24.04 GCC",
+              os: "ubuntu-24.04",
               cc: "/usr/bin/gcc",
               cxx: "/usr/bin/g++",
-              format: "/usr/bin/clang-format-14",
-              tidy: "/usr/bin/clang-tidy-14",
+              format: "/usr/bin/clang-format-15",
+              tidy: "/usr/bin/clang-tidy-15",
             }
           - {
-              name: "Ubuntu 22.04 Clang",
-              os: "ubuntu-22.04",
-              cc: "/usr/bin/clang-14",
-              cxx: "/usr/bin/clang++-14",
-              format: "/usr/bin/clang-format-14",
-              tidy: "/usr/bin/clang-tidy-14",
+              name: "Ubuntu 24.04 Clang",
+              os: "ubuntu-24.04",
+              cc: "/usr/bin/clang-15",
+              cxx: "/usr/bin/clang++-15",
+              format: "/usr/bin/clang-format-15",
+              tidy: "/usr/bin/clang-tidy-15",
             }
           - {
-              name: "macOS 13 (Ventura) AppleClang",
-              os: macos-13,
-              format: "/usr/local/opt/llvm@14/bin/clang-format",
-              tidy: "/usr/local/opt/llvm@14/bin/clang-tidy",
+              name: "macOS 14 (Sonoma) AppleClang",
+              os: macos-14,
+              format: "/opt/homebrew/opt/llvm@15/bin/clang-format",
+              tidy: "/opt/homebrew/opt/llvm@15/bin/clang-tidy",
             }
+          - {
+              name: "macOS 15 (Sequoia) AppleClang",
+              os: macos-15,
+              format: "/opt/homebrew/opt/llvm@15/bin/clang-format",
+              tidy: "/opt/homebrew/opt/llvm@15/bin/clang-tidy",
+          }
     steps:
       - uses: actions/checkout@v2
 
@@ -75,13 +81,6 @@ jobs:
         working-directory: ${{github.workspace}}/build
         run: make check-clang-tidy
 
-      - name: Check Tests (Ubuntu)
-        if: runner.os == 'Linux'
-        working-directory: ${{github.workspace}}/build
-        run: make build-tests && make check-public-ci-tests
-
-      - name: Check Tests (OSX)
-        if: runner.os == 'macOS'
+      - name: Check Tests
         working-directory: ${{github.workspace}}/build
-        # Disable container overflow checks on OSX
         run: make build-tests && make check-public-ci-tests

.gitignore

@@ -244,3 +244,6 @@ docs/dev
 
 # Submission zip files
 *.zip
+
+# LLM Stuff
+.aider*

CMakeLists.txt

@@ -3,9 +3,10 @@ set(CMAKE_EXPORT_COMPILE_COMMANDS ON) # For clang-tidy.
 set(BUILD_SHARED_LIBS OFF) # We expect external libraries to be linked statically.
 set(CMAKE_CXX_STANDARD 17) # Compile as C++17.
 set(CMAKE_CXX_STANDARD_REQUIRED ON) # Require C++17 support.
+set(CLANG_VERSION 15) # If you change this, make sure you change `packages.sh`
 
 project(BusTub
-        VERSION 2023.1
+        VERSION 2025.8
         DESCRIPTION "The BusTub Relational Database Management System (Educational) @ https://github.com/cmu-db/bustub"
         LANGUAGES C CXX
 )
@@ -38,19 +39,19 @@ endif()
 
 # Expected directory structure.
 set(BUSTUB_BUILD_SUPPORT_DIR "${CMAKE_SOURCE_DIR}/build_support")
-set(BUSTUB_CLANG_SEARCH_PATH "/usr/local/bin" "/usr/bin" "/usr/local/opt/llvm/bin" "/usr/local/opt/llvm@14/bin"
-        "/opt/homebrew/opt/llvm@14/bin/")
+set(BUSTUB_CLANG_SEARCH_PATH "/usr/local/bin" "/usr/bin" "/usr/local/opt/llvm/bin" "/usr/local/opt/llvm@${CLANG_VERSION}/bin"
+        "/opt/homebrew/opt/llvm@${CLANG_VERSION}/bin/")
 
 if(CMAKE_CXX_COMPILER_ID STREQUAL "Clang")
-        if(CMAKE_CXX_COMPILER_VERSION MATCHES "^14.")
+        if(CMAKE_CXX_COMPILER_VERSION MATCHES "^${CLANG_VERSION}.")
                 message(STATUS "You're using ${CMAKE_CXX_COMPILER_ID} ${CMAKE_CXX_COMPILER_VERSION}")
         else()
-                message(WARNING "!! We recommend that you use clang-14 for developing BusTub. You're using ${CMAKE_CXX_COMPILER_ID} ${CMAKE_CXX_COMPILER_VERSION}, a different version.")
+                message(WARNING "!! We recommend that you use clang-${CLANG_VERSION} for developing BusTub. You're using ${CMAKE_CXX_COMPILER_ID} ${CMAKE_CXX_COMPILER_VERSION}, a different version.")
         endif()
 elseif(CMAKE_CXX_COMPILER_ID STREQUAL "AppleClang")
         message(STATUS "You're using ${CMAKE_CXX_COMPILER_ID} ${CMAKE_CXX_COMPILER_VERSION}")
 else()
-        message(WARNING "!! We recommend that you use clang-14 for developing BusTub. You're using ${CMAKE_CXX_COMPILER_ID} ${CMAKE_CXX_COMPILER_VERSION}, which is not clang.")
+        message(WARNING "!! We recommend that you use clang-${CLANG_VERSION} for developing BusTub. You're using ${CMAKE_CXX_COMPILER_ID} ${CMAKE_CXX_COMPILER_VERSION}, which is not clang.")
 endif()
 
 # #####################################################################################################################
@@ -64,7 +65,7 @@ enable_testing()
 
 # attempt to find the binary if user did not specify
 find_program(CLANG_FORMAT_BIN
-        NAMES clang-format clang-format-14
+        NAMES clang-format clang-format-${CLANG_VERSION}
         HINTS ${BUSTUB_CLANG_SEARCH_PATH})
 
 if("${CLANG_FORMAT_BIN}" STREQUAL "CLANG_FORMAT_BIN-NOTFOUND")
@@ -75,7 +76,7 @@ endif()
 
 # attempt to find the binary if user did not specify
 find_program(CLANG_TIDY_BIN
-        NAMES clang-tidy clang-tidy-14
+        NAMES clang-tidy clang-tidy-${CLANG_VERSION}
         HINTS ${BUSTUB_CLANG_SEARCH_PATH})
 
 if("${CLANG_TIDY_BIN}" STREQUAL "CLANG_TIDY_BIN-NOTFOUND")
@@ -87,7 +88,7 @@ else()
 endif()
 
 find_program(CLANG_APPLY_REPLACEMENTS_BIN
-        NAMES clang-apply-replacements clang-apply-replacements-14
+        NAMES clang-apply-replacements clang-apply-replacements-${CLANG_VERSION}
         HINTS ${BUSTUB_CLANG_SEARCH_PATH})
 
 if("${CLANG_APPLY_REPLACEMENTS_BIN}" STREQUAL "CLANG_APPLY_REPLACEMENTS_BIN-NOTFOUND")

build_support/packages.sh

@@ -7,10 +7,12 @@
 ## build and run the DBMS.
 ##
 ## Supported environments:
-##  * Ubuntu 22.04 (x86-64)
+##  * Ubuntu 24.04 (x86-64)
 ##  * macOS 13 Ventura (x86-64 or ARM)
 ## =================================================================
 
+CLANG_VERSION=15
+
 main() {
   set -o errexit
 
@@ -41,9 +43,8 @@ install() {
     LINUX)
       version=$(cat /etc/os-release | grep VERSION_ID | cut -d '"' -f 2)
       case $version in
-        18.04) install_linux ;;
-        20.04) install_linux ;;
         22.04) install_linux ;;
+        24.04) install_linux ;;
         *) give_up ;;
       esac
       ;;
@@ -77,7 +78,7 @@ install_mac() {
   brew ls --versions coreutils || brew install coreutils
   brew ls --versions doxygen || brew install doxygen
   brew ls --versions git || brew install git
-  (brew ls --versions llvm | grep 14) || brew install llvm@14
+  (brew ls --versions llvm | grep ${CLANG_VERSION}) || brew install llvm@${CLANG_VERSION}
   brew ls --versions libelf || brew install libelf
 }
 
@@ -87,9 +88,9 @@ install_linux() {
   # Install packages.
   apt-get -y install \
       build-essential \
-      clang-14 \
-      clang-format-14 \
-      clang-tidy-14 \
+      clang-${CLANG_VERSION} \
+      clang-format-${CLANG_VERSION} \
+      clang-tidy-${CLANG_VERSION} \
       cmake \
       doxygen \
       git \

src/binder/bind_insert.cpp

@@ -90,7 +90,7 @@ auto Binder::BindUpdate(duckdb_libpgquery::PGUpdateStmt *stmt) -> std::unique_pt
   for (auto cell = root->head; cell != nullptr; cell = cell->next) {
     auto target = reinterpret_cast<duckdb_libpgquery::PGResTarget *>(cell->data.ptr_value);
     auto column = ResolveColumnRefFromBaseTableRef(*table, std::vector{std::string{target->name}});
-    target_expr.emplace_back(std::make_pair(std::move(column), BindExpression(target->val)));
+    target_expr.emplace_back(std::move(column), BindExpression(target->val));
   }
 
   return std::make_unique<UpdateStatement>(std::move(table), std::move(filter_expr), std::move(target_expr));

src/binder/fmt_impl.cpp

@@ -45,7 +45,7 @@ auto BoundCTERef::ToString() const -> std::string {
 }
 
 auto BoundSubqueryRef::ToString() const -> std::string {
-  std::vector<std::string> columns;
+  std::vector<std::string> columns(select_list_name_.size());
   for (const auto &name : select_list_name_) {
     columns.push_back(fmt::format("{}", fmt::join(name, ".")));
   }
@@ -54,12 +54,12 @@ auto BoundSubqueryRef::ToString() const -> std::string {
 }
 
 auto BoundWindow::ToString() const -> std::string {
-  std::vector<std::string> partition_by;
+  std::vector<std::string> partition_by(partition_by_.size());
   for (const auto &expr : partition_by_) {
     partition_by.push_back(expr->ToString());
   }
 
-  std::vector<std::string> order_bys;
+  std::vector<std::string> order_bys(order_bys_.size());
   for (const auto &expr : order_bys_) {
     order_bys.push_back(expr->ToString());
   }

src/catalog/table_generator.cpp

@@ -78,7 +78,7 @@ void TableGenerator::FillTable(const std::shared_ptr<TableInfo> &info, TableInse
   uint32_t num_inserted = 0;
   uint32_t batch_size = 128;
   while (num_inserted < table_meta->num_rows_) {
-    std::vector<std::vector<Value>> values;
+    std::vector<std::vector<Value>> values(table_meta->col_meta_.size());
     uint32_t num_values = std::min(batch_size, table_meta->num_rows_ - num_inserted);
     for (auto &col_meta : table_meta->col_meta_) {
       values.emplace_back(MakeValues(&col_meta, num_values));

src/execution/hash_join_executor.cpp

@@ -26,7 +26,7 @@ HashJoinExecutor::HashJoinExecutor(ExecutorContext *exec_ctx, const HashJoinPlan
                                    std::unique_ptr<AbstractExecutor> &&left_child,
                                    std::unique_ptr<AbstractExecutor> &&right_child)
     : AbstractExecutor(exec_ctx) {
-  if (!(plan->GetJoinType() == JoinType::LEFT || plan->GetJoinType() == JoinType::INNER)) {
+  if (plan->GetJoinType() != JoinType::LEFT && plan->GetJoinType() != JoinType::INNER) {
     // Note for Spring 2025: You ONLY need to implement left join and inner join.
     throw bustub::NotImplementedException(fmt::format("join type {} not supported", plan->GetJoinType()));
   }

src/execution/nested_index_join_executor.cpp

@@ -24,7 +24,7 @@ namespace bustub {
 NestedIndexJoinExecutor::NestedIndexJoinExecutor(ExecutorContext *exec_ctx, const NestedIndexJoinPlanNode *plan,
                                                  std::unique_ptr<AbstractExecutor> &&child_executor)
     : AbstractExecutor(exec_ctx) {
-  if (!(plan->GetJoinType() == JoinType::LEFT || plan->GetJoinType() == JoinType::INNER)) {
+  if (plan->GetJoinType() != JoinType::LEFT && plan->GetJoinType() != JoinType::INNER) {
     // Note for Spring 2025: You ONLY need to implement left join and inner join.
     throw bustub::NotImplementedException(fmt::format("join type {} not supported", plan->GetJoinType()));
   }

src/execution/nested_loop_join_executor.cpp

@@ -28,7 +28,7 @@ NestedLoopJoinExecutor::NestedLoopJoinExecutor(ExecutorContext *exec_ctx, const
                                                std::unique_ptr<AbstractExecutor> &&left_executor,
                                                std::unique_ptr<AbstractExecutor> &&right_executor)
     : AbstractExecutor(exec_ctx) {
-  if (!(plan->GetJoinType() == JoinType::LEFT || plan->GetJoinType() == JoinType::INNER)) {
+  if (plan->GetJoinType() != JoinType::LEFT && plan->GetJoinType() != JoinType::INNER) {
     // Note for Spring 2025: You ONLY need to implement left join and inner join.
     throw bustub::NotImplementedException(fmt::format("join type {} not supported", plan->GetJoinType()));
   }

src/include/binder/table_ref/bound_base_table_ref.h

@@ -35,14 +35,14 @@ class BoundBaseTableRef : public BoundTableRef {
         schema_(std::move(schema)) {}
 
   auto ToString() const -> std::string override {
-    if (alias_ == std::nullopt) {
+    if (alias_ == std::nullopt || !alias_.has_value()) {
       return fmt::format("BoundBaseTableRef {{ table={}, oid={} }}", table_, oid_);
     }
     return fmt::format("BoundBaseTableRef {{ table={}, oid={}, alias={} }}", table_, oid_, *alias_);
   }
 
   auto GetBoundTableName() const -> std::string {
-    if (alias_ != std::nullopt) {
+    if (alias_ != std::nullopt && alias_.has_value()) {
       return *alias_;
     }
     return table_;

src/include/catalog/catalog.h

@@ -360,7 +360,7 @@ class Catalog {
   }
 
   auto GetTableNames() -> std::vector<std::string> {
-    std::vector<std::string> result;
+    std::vector<std::string> result(table_names_.size());
     for (const auto &x : table_names_) {
       result.push_back(x.first);
     }

src/include/execution/expressions/arithmetic_expression.h

@@ -51,7 +51,7 @@ class ArithmeticExpression : public AbstractExpression {
     Value lhs = GetChildAt(0)->Evaluate(tuple, schema);
     Value rhs = GetChildAt(1)->Evaluate(tuple, schema);
     auto res = PerformComputation(lhs, rhs);
-    if (res == std::nullopt) {
+    if (res == std::nullopt || !res.has_value()) {
       return ValueFactory::GetNullValueByType(TypeId::INTEGER);
     }
     return ValueFactory::GetIntegerValue(*res);
@@ -62,7 +62,7 @@ class ArithmeticExpression : public AbstractExpression {
     Value lhs = GetChildAt(0)->EvaluateJoin(left_tuple, left_schema, right_tuple, right_schema);
     Value rhs = GetChildAt(1)->EvaluateJoin(left_tuple, left_schema, right_tuple, right_schema);
     auto res = PerformComputation(lhs, rhs);
-    if (res == std::nullopt) {
+    if (res == std::nullopt || !res.has_value()) {
       return ValueFactory::GetNullValueByType(TypeId::INTEGER);
     }
     return ValueFactory::GetIntegerValue(*res);

src/include/primer/hyperloglog.h

@@ -22,7 +22,7 @@
 #include "common/util/hash_util.h"
 
 /** @brief Capacity of the bitset stream. */
-#define BITSET_CAPACITY 64
+static constexpr int BITSET_CAPACITY = 64;
 
 namespace bustub {
 

src/include/primer/hyperloglog_presto.h

@@ -24,12 +24,12 @@
 #include "common/util/hash_util.h"
 
 /** @brief Dense bucket size. */
-#define DENSE_BUCKET_SIZE 4
+static constexpr int DENSE_BUCKET_SIZE = 4;
 /** @brief Overflow bucket size. */
-#define OVERFLOW_BUCKET_SIZE 3
+static constexpr int OVERFLOW_BUCKET_SIZE = 3;
 
 /** @brief Total bucket size. */
-#define TOTAL_BUCKET_SIZE (DENSE_BUCKET_SIZE + OVERFLOW_BUCKET_SIZE)
+static constexpr int TOTAL_BUCKET_SIZE = (DENSE_BUCKET_SIZE + OVERFLOW_BUCKET_SIZE);
 
 namespace bustub {
 

src/include/storage/page/hash_table_page_defs.h

@@ -49,4 +49,4 @@
  * Extending the directory implementation to span multiple pages would be a meaningful improvement to the
  * implementation.
  */
-#define DIRECTORY_ARRAY_SIZE 512
+static constexpr int DIRECTORY_ARRAY_SIZE = 512;

src/optimizer/nlj_as_index_join.cpp

@@ -78,7 +78,7 @@ auto Optimizer::OptimizeNLJAsIndexJoin(const AbstractPlanNodeRef &plan) -> Abstr
               const auto &right_seq_scan = dynamic_cast<const SeqScanPlanNode &>(*nlj_plan.GetRightPlan());
               if (left_expr->GetTupleIdx() == 0 && right_expr->GetTupleIdx() == 1) {
                 if (auto index = MatchIndex(right_seq_scan.table_name_, right_expr->GetColIdx());
-                    index != std::nullopt) {
+                    index != std::nullopt && index.has_value()) {
                   auto [index_oid, index_name] = *index;
                   return std::make_shared<NestedIndexJoinPlanNode>(
                       nlj_plan.output_schema_, nlj_plan.GetLeftPlan(), std::move(left_expr_tuple_0),
@@ -88,7 +88,7 @@ auto Optimizer::OptimizeNLJAsIndexJoin(const AbstractPlanNodeRef &plan) -> Abstr
               }
               if (left_expr->GetTupleIdx() == 1 && right_expr->GetTupleIdx() == 0) {
                 if (auto index = MatchIndex(right_seq_scan.table_name_, left_expr->GetColIdx());
-                    index != std::nullopt) {
+                    index != std::nullopt && index.has_value()) {
                   auto [index_oid, index_name] = *index;
                   return std::make_shared<NestedIndexJoinPlanNode>(
                       nlj_plan.output_schema_, nlj_plan.GetLeftPlan(), std::move(right_expr_tuple_0),

src/optimizer/order_by_index_scan.cpp

@@ -50,7 +50,7 @@ auto Optimizer::OptimizeOrderByAsIndexScan(const AbstractPlanNodeRef &plan) -> A
     std::vector<uint32_t> order_by_column_ids;
     for (const auto &[order_type, expr] : order_bys) {
       // Order type is asc or default
-      if (!(order_type == OrderByType::ASC || order_type == OrderByType::DEFAULT)) {
+      if (order_type != OrderByType::ASC && order_type != OrderByType::DEFAULT) {
         return optimized_plan;
       }
 

src/planner/plan_select.cpp

@@ -133,7 +133,7 @@ auto Planner::PlanSelect(const SelectStatement &statement) -> AbstractPlanNodeRe
     for (const auto &order_by : statement.sort_) {
       auto [_, expr] = PlanExpression(*order_by->expr_, {plan});
       auto abstract_expr = std::move(expr);
-      order_bys.emplace_back(std::make_pair(order_by->type_, abstract_expr));
+      order_bys.emplace_back(order_by->type_, abstract_expr);
     }
     plan = std::make_shared<SortPlanNode>(std::make_shared<Schema>(plan->OutputSchema()), plan, std::move(order_bys));
   }
@@ -174,8 +174,9 @@ auto Planner::PlanSelect(const SelectStatement &statement) -> AbstractPlanNodeRe
     if (offset != std::nullopt) {
       throw NotImplementedException("OFFSET clause is not supported yet.");
     }
-
-    plan = std::make_shared<LimitPlanNode>(std::make_shared<Schema>(plan->OutputSchema()), plan, *limit);
+    if (limit.has_value()) {
+      plan = std::make_shared<LimitPlanNode>(std::make_shared<Schema>(plan->OutputSchema()), plan, *limit);
+    }
   }
 
   return plan;

src/storage/page/table_page.cpp

@@ -55,7 +55,7 @@ auto TablePage::GetNextTupleOffset(const TupleMeta &meta, const Tuple &tuple) co
  */
 auto TablePage::InsertTuple(const TupleMeta &meta, const Tuple &tuple) -> std::optional<uint16_t> {
   auto tuple_offset = GetNextTupleOffset(meta, tuple);
-  if (tuple_offset == std::nullopt) {
+  if (tuple_offset == std::nullopt || !tuple_offset.has_value()) {
     return std::nullopt;
   }
   auto tuple_id = num_tuples_;

src/storage/table/table_heap.cpp

@@ -81,7 +81,7 @@ auto TableHeap::InsertTuple(const TupleMeta &meta, const Tuple &tuple, LockManag
   auto last_page_id = last_page_id_;
 
   auto page = page_guard.AsMut<TablePage>();
-  auto slot_id = *page->InsertTuple(meta, tuple);
+  auto slot_id = *page->InsertTuple(meta, tuple);  // NOLINT
 
   // only allow one insertion at a time, otherwise it will deadlock.
   guard.unlock();