Merge branch 'remove-regex-prefix' into remove-uneeded-this

y-scope · Jan 7, 2025 · ed631b3 · ed631b3
2 parents 9a82620 + 5dc5953
commit ed631b3
Show file tree

Hide file tree

Showing 8 changed files with 37 additions and 51 deletions.
diff --git a/CMakeLists.txt b/CMakeLists.txt
@@ -99,11 +99,10 @@ set(SOURCE_FILES
     src/log_surgeon/finite_automata/Dfa.hpp
     src/log_surgeon/finite_automata/DfaState.hpp
     src/log_surgeon/finite_automata/DfaStatePair.hpp
-    src/log_surgeon/finite_automata/DfaStateType.hpp
     src/log_surgeon/finite_automata/Nfa.hpp
     src/log_surgeon/finite_automata/NfaState.hpp
-    src/log_surgeon/finite_automata/NfaStateType.hpp
     src/log_surgeon/finite_automata/RegisterHandler.hpp
+    src/log_surgeon/finite_automata/StateType.hpp
     src/log_surgeon/finite_automata/Tag.hpp
     src/log_surgeon/finite_automata/TaggedTransition.hpp
     src/log_surgeon/finite_automata/UnicodeIntervalTree.hpp

diff --git a/src/log_surgeon/finite_automata/DfaState.hpp b/src/log_surgeon/finite_automata/DfaState.hpp
@@ -9,17 +9,17 @@
 #include <vector>
 
 #include <log_surgeon/Constants.hpp>
-#include <log_surgeon/finite_automata/DfaStateType.hpp>
+#include <log_surgeon/finite_automata/StateType.hpp>
 #include <log_surgeon/finite_automata/UnicodeIntervalTree.hpp>
 
 namespace log_surgeon::finite_automata {
-template <DfaStateType state_type>
+template <StateType state_type>
 class DfaState;
 
-using ByteDfaState = DfaState<DfaStateType::Byte>;
-using Utf8DfaState = DfaState<DfaStateType::Utf8>;
+using ByteDfaState = DfaState<StateType::Byte>;
+using Utf8DfaState = DfaState<StateType::Utf8>;
 
-template <DfaStateType state_type>
+template <StateType state_type>
 class DfaState {
 public:
     using Tree = UnicodeIntervalTree<DfaState*>;
@@ -51,14 +51,14 @@ class DfaState {
 private:
     std::vector<uint32_t> m_matching_variable_ids;
     DfaState* m_bytes_transition[cSizeOfByte];
-    // NOTE: We don't need m_tree_transitions for the `state_type == DfaStateType::Byte` case, so we
+    // NOTE: We don't need m_tree_transitions for the `state_type == StateType::Byte` case, so we
     // use an empty class (`std::tuple<>`) in that case.
-    std::conditional_t<state_type == DfaStateType::Utf8, Tree, std::tuple<>> m_tree_transitions;
+    std::conditional_t<state_type == StateType::Utf8, Tree, std::tuple<>> m_tree_transitions;
 };
 
-template <DfaStateType state_type>
+template <StateType state_type>
 auto DfaState<state_type>::next(uint32_t character) const -> DfaState* {
-    if constexpr (DfaStateType::Byte == state_type) {
+    if constexpr (StateType::Byte == state_type) {
         return m_bytes_transition[character];
     } else {
         if (character < cSizeOfByte) {

diff --git a/src/log_surgeon/finite_automata/DfaStateType.hpp b/src/log_surgeon/finite_automata/DfaStateType.hpp
diff --git a/src/log_surgeon/finite_automata/NfaState.hpp b/src/log_surgeon/finite_automata/NfaState.hpp
@@ -12,18 +12,18 @@
 
 #include <fmt/format.h>
 
-#include <log_surgeon/finite_automata/NfaStateType.hpp>
+#include <log_surgeon/finite_automata/StateType.hpp>
 #include <log_surgeon/finite_automata/TaggedTransition.hpp>
 #include <log_surgeon/finite_automata/UnicodeIntervalTree.hpp>
 
 namespace log_surgeon::finite_automata {
-template <NfaStateType state_type>
+template <StateType state_type>
 class NfaState;
 
-using ByteNfaState = NfaState<NfaStateType::Byte>;
-using Utf8NfaState = NfaState<NfaStateType::Utf8>;
+using ByteNfaState = NfaState<StateType::Byte>;
+using Utf8NfaState = NfaState<StateType::Utf8>;
 
-template <NfaStateType state_type>
+template <StateType state_type>
 class NfaState {
 public:
     using Tree = UnicodeIntervalTree<NfaState*>;
@@ -113,12 +113,12 @@ class NfaState {
     std::vector<NfaState*> m_epsilon_transitions;
     std::array<std::vector<NfaState*>, cSizeOfByte> m_bytes_transitions;
     // NOTE: We don't need m_tree_transitions for the `stateType ==
-    // NfaStateType::Byte` case, so we use an empty class (`std::tuple<>`)
+    // StateType::Byte` case, so we use an empty class (`std::tuple<>`)
     // in that case.
-    std::conditional_t<state_type == NfaStateType::Utf8, Tree, std::tuple<>> m_tree_transitions;
+    std::conditional_t<state_type == StateType::Utf8, Tree, std::tuple<>> m_tree_transitions;
 };
 
-template <NfaStateType state_type>
+template <StateType state_type>
 auto NfaState<state_type>::add_interval(Interval interval, NfaState* dest_state) -> void {
     if (interval.first < cSizeOfByte) {
         uint32_t const bound = std::min(interval.second, cSizeOfByte - 1);
@@ -127,7 +127,7 @@ auto NfaState<state_type>::add_interval(Interval interval, NfaState* dest_state)
         }
         interval.first = bound + 1;
     }
-    if constexpr (NfaStateType::Utf8 == state_type) {
+    if constexpr (StateType::Utf8 == state_type) {
         if (interval.second < cSizeOfByte) {
             return;
         }
@@ -165,7 +165,7 @@ auto NfaState<state_type>::add_interval(Interval interval, NfaState* dest_state)
     }
 }
 
-template <NfaStateType state_type>
+template <StateType state_type>
 auto NfaState<state_type>::serialize(std::unordered_map<NfaState const*, uint32_t> const& state_ids
 ) const -> std::optional<std::string> {
     std::vector<std::string> byte_transitions;

diff --git a/src/log_surgeon/finite_automata/NfaStateType.hpp b/src/log_surgeon/finite_automata/NfaStateType.hpp
diff --git a/src/log_surgeon/finite_automata/StateType.hpp b/src/log_surgeon/finite_automata/StateType.hpp
@@ -0,0 +1,13 @@
+#ifndef LOG_SURGEON_FINITE_AUTOMATA_STATE_TYPE
+#define LOG_SURGEON_FINITE_AUTOMATA_STATE_TYPE
+
+#include <cstdint>
+
+namespace log_surgeon::finite_automata {
+enum class StateType : uint8_t {
+    Byte,
+    Utf8
+};
+}  // namespace log_surgeon::finite_automata
+
+#endif  // LOG_SURGEON_FINITE_AUTOMATA_STATE_TYPE
diff --git a/tests/CMakeLists.txt b/tests/CMakeLists.txt
@@ -7,8 +7,8 @@ set(
         ../src/log_surgeon/finite_automata/RegexAST.hpp
         ../src/log_surgeon/finite_automata/Nfa.hpp
         ../src/log_surgeon/finite_automata/NfaState.hpp
-        ../src/log_surgeon/finite_automata/NfaStateType.hpp
         ../src/log_surgeon/finite_automata/RegisterHandler.hpp
+        ../src/log_surgeon/finite_automata/StateType.hpp
         ../src/log_surgeon/finite_automata/Tag.hpp
         ../src/log_surgeon/finite_automata/TaggedTransition.hpp
         ../src/log_surgeon/Lalr1Parser.cpp
@@ -24,7 +24,7 @@ set(
         ../src/log_surgeon/Token.hpp
 )
 
-set(SOURCES_TESTS test-lexer.cpp test-NFA.cpp test-prefix-tree.cpp test-register-handler.cpp test-tag.cpp)
+set(SOURCES_TESTS test-lexer.cpp test-nfa.cpp test-prefix-tree.cpp test-register-handler.cpp test-tag.cpp)
 
 add_executable(unit-test ${SOURCES_LOG_SURGEON} ${SOURCES_TESTS})
 target_link_libraries(unit-test PRIVATE Catch2::Catch2WithMain log_surgeon::log_surgeon)

diff --git a/tests/test-NFA.cpp → tests/test-nfa.cpp b/tests/test-NFA.cpp → tests/test-nfa.cpp
@@ -21,7 +21,7 @@ using std::stringstream;
 using std::vector;
 
 using ByteLexicalRule = log_surgeon::LexicalRule<ByteNfaState>;
-using ByteNFA = log_surgeon::finite_automata::Nfa<ByteNfaState>;
+using ByteNfa = log_surgeon::finite_automata::Nfa<ByteNfaState>;
 using RegexASTCatByte = log_surgeon::finite_automata::RegexASTCat<ByteNfaState>;
 using RegexASTCaptureByte = log_surgeon::finite_automata::RegexASTCapture<ByteNfaState>;
 using RegexASTGroupByte = log_surgeon::finite_automata::RegexASTGroup<ByteNfaState>;
@@ -44,7 +44,7 @@ TEST_CASE("Test NFA", "[NFA]") {
     auto& capture_rule_ast = dynamic_cast<SchemaVarAST&>(*schema_ast->m_schema_vars[0]);
     vector<ByteLexicalRule> rules;
     rules.emplace_back(0, std::move(capture_rule_ast.m_regex_ptr));
-    ByteNFA const nfa{std::move(rules)};
+    ByteNfa const nfa{std::move(rules)};
 
     // Compare against expected output
     string expected_serialized_nfa = "0:byte_transitions={A-->1,Z-->2},"