Add backslash identifier support to v2scriptadd/backslash-identifier

Add buffer::char_type size multiplier to max file size of string intern buffer

Fix list grammar segfaults
Fix diagnostic logger intern segfaults from buffer reallocation
Fix non-string-supported CSV parser not supporting Victoria 2 CSV escaping behavior

6 files changed, 61 insertions, 56 deletions

diff --git a/src/openvic-dataloader/AbstractSyntaxTree.hpp b/src/openvic-dataloader/AbstractSyntaxTree.hpp
index ade1c82..f9f5796 100644
--- a/src/openvic-dataloader/AbstractSyntaxTree.hpp
+++ b/src/openvic-dataloader/AbstractSyntaxTree.hpp
@@ -3,6 +3,7 @@
 #include <concepts>
 #include <cstdio>
 #include <string_view>
+#include <type_traits>
 #include <utility>
 
 #include <openvic-dataloader/NodeLocation.hpp>
@@ -51,12 +52,12 @@ namespace ovdl {
 		using node_type = typename file_type::node_type;
 
 		explicit BasicAbstractSyntaxTree(file_type&& file)
-			: AbstractSyntaxTree(file.size()),
+			: AbstractSyntaxTree(file.size() * file.visit_buffer([](auto&& buffer) -> size_t { return sizeof(typename std::decay_t<decltype(buffer)>::char_type); })),
 			  _file { std::move(file) } {}
 
 		template<typename Encoding, typename MemoryResource = void>
 		explicit BasicAbstractSyntaxTree(lexy::buffer<Encoding, MemoryResource>&& buffer)
-			: AbstractSyntaxTree(buffer.size()),
+			: AbstractSyntaxTree(buffer.size() * sizeof(Encoding::char_type)),
 			  _file { std::move(buffer) } {}
 
 		void set_location(const node_type* n, NodeLocation loc) {
diff --git a/src/openvic-dataloader/DiagnosticLogger.hpp b/src/openvic-dataloader/DiagnosticLogger.hpp
index 9810e1e..8c491ca 100644
--- a/src/openvic-dataloader/DiagnosticLogger.hpp
+++ b/src/openvic-dataloader/DiagnosticLogger.hpp
@@ -2,6 +2,7 @@
 
 #include <concepts> // IWYU pragma: keep
 #include <cstdio>
+#include <iostream>
 #include <ostream>
 #include <string>
 #include <type_traits>
@@ -37,12 +38,7 @@ namespace ovdl {
 	template<typename ParseState>
 	struct BasicDiagnosticLogger;
 
-	struct DiagnosticLogger {
-		struct SymbolId;
-		using index_type = std::uint32_t;
-		using symbol_type = dryad::symbol<SymbolId, index_type>;
-		using symbol_interner_type = dryad::symbol_interner<SymbolId, char, index_type>;
-
+	struct DiagnosticLogger : error::ErrorSymbolInterner {
 		using AnnotationKind = lexy_ext::annotation_kind;
 		using DiagnosticKind = lexy_ext::diagnostic_kind;
 
@@ -116,23 +112,23 @@ namespace ovdl {
 						}
 						result = writer.error();
 					} else {
-						auto production = _logger.intern_cstr(production_name);
+						auto production = production_name;
 						if constexpr (std::is_same_v<Tag, lexy::expected_literal>) {
 							auto string = lexy::_detail::make_literal_lexeme<typename Reader::encoding>(error.string(), error.length());
 							NodeLocation loc = NodeLocation::make_from(context.position(), error.position() - 1);
-							auto message = _logger.intern_cstr(fmt::format("expected '{}'", string.data()));
+							auto message = _logger.intern(fmt::format("expected '{}'", string.data()));
 							result = _logger.template create<error::ExpectedLiteral>(loc, message, production);
 						} else if constexpr (std::is_same_v<Tag, lexy::expected_keyword>) {
 							auto string = lexy::_detail::make_literal_lexeme<typename Reader::encoding>(error.string(), error.length());
 							NodeLocation loc = NodeLocation::make_from(context.position(), error.position() - 1);
-							auto message = _logger.intern_cstr(fmt::format("expected keyword '{}'", string.data()));
+							auto message = _logger.intern(fmt::format("expected keyword '{}'", string.data()));
 							result = _logger.template create<error::ExpectedKeyword>(loc, message, production);
 						} else if constexpr (std::is_same_v<Tag, lexy::expected_char_class>) {
-							auto message = _logger.intern_cstr(fmt::format("expected {}", error.name()));
+							auto message = _logger.intern(fmt::format("expected {}", error.name()));
 							result = _logger.template create<error::ExpectedCharClass>(error.position(), message, production);
 						} else {
 							NodeLocation loc = NodeLocation::make_from(error.begin(), error.end());
-							auto message = _logger.intern_cstr(error.message());
+							auto message = _logger.intern(error.message());
 							result = _logger.template create<error::GenericParseError>(loc, message, production);
 						}
 					}
@@ -361,7 +357,7 @@ namespace ovdl {
 				});
 
 				error::Annotation* annotation;
-				auto message = _logger.intern_cstr(output);
+				auto message = _logger.intern(output);
 				switch (kind) {
 					case AnnotationKind::primary:
 						annotation = _logger.create<error::PrimaryAnnotation>(loc, message);
@@ -404,7 +400,7 @@ namespace ovdl {
 				});
 			impl.write_path(iter, file().path());
 
-			auto message = intern_cstr(output);
+			auto message = intern(output);
 			error->_set_message(message);
 			if (!error->is_linked_in_tree())
 				insert(error);
@@ -422,8 +418,8 @@ namespace ovdl {
 				});
 			impl.write_path(iter, file().path());
 
-			auto production = intern_cstr(production_name);
-			auto message = intern_cstr(output);
+			auto production = production_name;
+			auto message = intern(output);
 			auto* error = [&] {
 				if constexpr (std::is_same_v<Tag, lexy::expected_literal>) {
 					return create<error::ExpectedLiteral>(loc, message, production);
diff --git a/src/openvic-dataloader/csv/CsvGrammar.hpp b/src/openvic-dataloader/csv/CsvGrammar.hpp
index 19aee54..91226c7 100644
--- a/src/openvic-dataloader/csv/CsvGrammar.hpp
+++ b/src/openvic-dataloader/csv/CsvGrammar.hpp
@@ -11,10 +11,7 @@
 
 #include <lexy/_detail/config.hpp>
 #include <lexy/callback.hpp>
-#include <lexy/callback/string.hpp>
 #include <lexy/dsl.hpp>
-#include <lexy/dsl/ascii.hpp>
-#include <lexy/dsl/option.hpp>
 #include <lexy/encoding.hpp>
 
 #include "detail/Convert.hpp"
@@ -73,6 +70,9 @@ namespace ovdl::csv::grammar {
 	constexpr auto escaped_quote = lexy::symbol_table<char> //
 									   .map<'"'>('"');
 
+	constexpr auto escaped_newline = lexy::symbol_table<char> //
+										 .map<'n'>('\n');
+
 	template<ParseOptions Options>
 	struct CsvGrammar {
 		struct StringValue : lexy::scan_production<std::string>,
@@ -118,6 +118,11 @@ namespace ovdl::csv::grammar {
 			template<auto character>
 			static constexpr auto _escape_check = character - (lexy::dsl::lit_b<Options.SepChar> / lexy::dsl::ascii::newline);
 
+			struct Backslash {
+				static constexpr auto rule = LEXY_LIT("\\n");
+				static constexpr auto value = lexy::constant('\n');
+			};
+
 			template<typename Context, typename Reader>
 			static constexpr scan_result scan(lexy::rule_scanner<Context, Reader>& scanner, detail::IsFileParseState auto& state) {
 				using encoding = typename Reader::encoding;
@@ -134,13 +139,16 @@ namespace ovdl::csv::grammar {
 					if constexpr (Options.SupportStrings) {
 						return lexy::dsl::identifier(character - (lexy::dsl::lit_b<Options.SepChar> / lexy::dsl::ascii::newline));
 					} else {
-						auto escape_check_char = _escape_check<character>;
-						auto id_check_char = escape_check_char - lexy::dsl::lit_b<'\\'>;
-						auto id_segment = lexy::dsl::identifier(id_check_char);
-						auto escape_segement = lexy::dsl::token(escape_check_char);
-						auto escape_sym = lexy::dsl::symbol<escaped_symbols>(escape_segement);
-						auto escape_rule = lexy::dsl::lit_b<'\\'> >> escape_sym;
-						return lexy::dsl::list(id_segment | escape_rule);
+						constexpr auto backslash = lexy::dsl::lit_b<'\\'>;
+
+						constexpr auto escape_check_char = _escape_check<character>;
+						constexpr auto escape_rule = lexy::dsl::p<Backslash>;
+
+						return lexy::dsl::list(
+							lexy::dsl::identifier(escape_check_char - backslash) |
+							escape_rule |
+							lexy::dsl::capture(escape_check_char) //
+						);
 					}
 				}();
 
diff --git a/src/openvic-dataloader/csv/Parser.cpp b/src/openvic-dataloader/csv/Parser.cpp
index 8fe8b17..bbd1be4 100644
--- a/src/openvic-dataloader/csv/Parser.cpp
+++ b/src/openvic-dataloader/csv/Parser.cpp
@@ -195,6 +195,10 @@ typename Parser::error_range Parser::get_errors() const {
 	return _parse_handler->get_errors();
 }
 
+std::string_view Parser::error(const ovdl::error::Error* error) const {
+	return error->message(_parse_handler->parse_state().logger().symbol_interner());
+}
+
 const FilePosition Parser::get_error_position(const error::Error* error) const {
 	if (!error || !error->is_linked_in_tree()) {
 		return {};
@@ -231,20 +235,20 @@ void Parser::print_errors_to(std::basic_ostream<char>& stream) const {
 		dryad::visit_tree(
 			error,
 			[&](const error::BufferError* buffer_error) {
-				stream << "buffer error: " << buffer_error->message() << '\n';
+				stream << "buffer error: " << this->error(buffer_error) << '\n';
 			},
 			[&](dryad::child_visitor<error::ErrorKind> visitor, const error::AnnotatedError* annotated_error) {
-				stream << annotated_error->message() << '\n';
+				stream << this->error(annotated_error) << '\n';
 				auto annotations = annotated_error->annotations();
 				for (auto annotation : annotations) {
 					visitor(annotation);
 				}
 			},
 			[&](const error::PrimaryAnnotation* primary) {
-				stream << primary->message() << '\n';
+				stream << this->error(primary) << '\n';
 			},
 			[&](const error::SecondaryAnnotation* secondary) {
-				stream << secondary->message() << '\n';
+				stream << this->error(secondary) << '\n';
 			});
 	}
 }
\ No newline at end of file
diff --git a/src/openvic-dataloader/v2script/Parser.cpp b/src/openvic-dataloader/v2script/Parser.cpp
index 2375a1a..e5234d6 100644
--- a/src/openvic-dataloader/v2script/Parser.cpp
+++ b/src/openvic-dataloader/v2script/Parser.cpp
@@ -320,6 +320,10 @@ Parser::error_range Parser::get_errors() const {
 	return _parse_handler->get_errors();
 }
 
+std::string_view Parser::error(const ovdl::error::Error* error) const {
+	return error->message(_parse_handler->parse_state().logger().symbol_interner());
+}
+
 const FilePosition Parser::get_error_position(const error::Error* error) const {
 	if (!error || !error->is_linked_in_tree()) {
 		return {};
@@ -352,20 +356,20 @@ void Parser::print_errors_to(std::basic_ostream<char>& stream) const {
 		dryad::visit_tree(
 			error,
 			[&](const error::BufferError* buffer_error) {
-				stream << "buffer error: " << buffer_error->message() << '\n';
+				stream << "buffer error: " << this->error(buffer_error) << '\n';
 			},
 			[&](dryad::child_visitor<error::ErrorKind> visitor, const error::AnnotatedError* annotated_error) {
-				stream << annotated_error->message() << '\n';
+				stream << this->error(annotated_error) << '\n';
 				auto annotations = annotated_error->annotations();
 				for (auto annotation : annotations) {
 					visitor(annotation);
 				}
 			},
 			[&](const error::PrimaryAnnotation* primary) {
-				stream << primary->message() << '\n';
+				stream << this->error(primary) << '\n';
 			},
 			[&](const error::SecondaryAnnotation* secondary) {
-				stream << secondary->message() << '\n';
+				stream << this->error(secondary) << '\n';
 			});
 	}
 }
\ No newline at end of file
diff --git a/src/openvic-dataloader/v2script/SimpleGrammar.hpp b/src/openvic-dataloader/v2script/SimpleGrammar.hpp
index c47b243..5474c79 100644
--- a/src/openvic-dataloader/v2script/SimpleGrammar.hpp
+++ b/src/openvic-dataloader/v2script/SimpleGrammar.hpp
@@ -5,15 +5,6 @@
 
 #include <lexy/callback.hpp>
 #include <lexy/dsl.hpp>
-#include <lexy/dsl/any.hpp>
-#include <lexy/dsl/identifier.hpp>
-#include <lexy/dsl/option.hpp>
-#include <lexy/dsl/peek.hpp>
-#include <lexy/dsl/punctuator.hpp>
-#include <lexy/dsl/recover.hpp>
-#include <lexy/dsl/scan.hpp>
-#include <lexy/dsl/symbol.hpp>
-#include <lexy/dsl/unicode.hpp>
 #include <lexy/encoding.hpp>
 #include <lexy/input/base.hpp>
 #include <lexy/input/buffer.hpp>
@@ -63,7 +54,7 @@ namespace ovdl::v2script::grammar {
 	/* REQUIREMENTS: DAT-631 */
 	static constexpr auto comment_specifier = LEXY_LIT("#") >> lexy::dsl::until(lexy::dsl::newline).or_eof();
 
-	static constexpr auto ascii = lexy::dsl::ascii::alpha_digit_underscore / LEXY_ASCII_ONE_OF("+:@%&'-.");
+	static constexpr auto ascii = lexy::dsl::ascii::alpha_digit_underscore / LEXY_ASCII_ONE_OF("+:@%&'-.\\");
 
 	/* REQUIREMENTS:
 	 * DAT-632
@@ -91,7 +82,7 @@ namespace ovdl::v2script::grammar {
 
 	static constexpr auto data_char_class = LEXY_CHAR_CLASS("DataSpecifier", data_specifier);
 
-	static constexpr auto utf_data_specifier = lexy::dsl::unicode::xid_continue / LEXY_ASCII_ONE_OF("+:@%&'-.");
+	static constexpr auto utf_data_specifier = lexy::dsl::unicode::xid_continue / LEXY_ASCII_ONE_OF("+:@%&'-.\\");
 
 	static constexpr auto utf_char_class = LEXY_CHAR_CLASS("DataSpecifier", utf_data_specifier);
 
@@ -196,9 +187,9 @@ namespace ovdl::v2script::grammar {
 					if constexpr (Options.NoStringEscape) {
 						auto c = [] {
 							if constexpr (std::same_as<encoding, lexy::default_encoding> || std::same_as<encoding, lexy::byte_encoding>) {
-								return dsl::lit_b_range<0x20, 0xFF> / lexy::dsl::lit_b<0x07> / lexy::dsl::lit_b<0x09> / lexy::dsl::lit_b<0x0A> / lexy::dsl::lit_b<0x0D>;
+								return dsl::lit_b_range<0x01, 0xFF>;
 							} else {
-								return -lexy::dsl::unicode::control;
+								return lexy::dsl::unicode::character;
 							}
 						}();
 						return lexy::dsl::quoted(c);
@@ -287,7 +278,7 @@ namespace ovdl::v2script::grammar {
 			}();
 
 			static constexpr auto value = dsl::callback<ast::Statement*>(
-				[](detail::IsParseState auto& state, const char* pos, ast::IdentifierValue* name, ast::Value* initializer) {
+				[](detail::IsParseState auto& state, const char* pos, ast::IdentifierValue* name, ast::Value* initializer) -> ast::AssignStatement* {
 					return state.ast().template create<ast::AssignStatement>(pos, name, initializer);
 				},
 				[](detail::IsParseState auto& state, bool&, const char* pos, ast::IdentifierValue* name, ast::Value* initializer) {
@@ -306,11 +297,15 @@ namespace ovdl::v2script::grammar {
 					return state.ast().template create<ast::ValueStatement>(pos, left);
 				},
 				[](detail::IsParseState auto& state, ast::Value* left) -> ast::ValueStatement* {
-					if (left == nullptr) return nullptr;
+					if (left == nullptr) { // May no longer be neccessary
+						return nullptr;
+					}
 					return state.ast().template create<ast::ValueStatement>(state.ast().location_of(left), left);
 				},
 				[](detail::IsParseState auto& state, bool&, ast::Value* left) -> ast::ValueStatement* {
-					if (left == nullptr) return nullptr;
+					if (left == nullptr) { // May no longer be neccessary
+						return nullptr;
+					}
 					return state.ast().template create<ast::ValueStatement>(state.ast().location_of(left), left);
 				});
 		};
@@ -322,12 +317,12 @@ namespace ovdl::v2script::grammar {
 
 				auto assign_statement = lexy::dsl::recurse_branch<AssignmentStatement>;
 
-				auto assign_try = lexy::dsl::try_(assign_statement);
+				auto assign_try = lexy::dsl::try_(assign_statement, lexy::dsl::nullopt);
 				auto assign_opt = lexy::dsl::opt(lexy::dsl::list(assign_try));
 
 				auto curly_bracket = dsl::curly_bracketed(assign_opt + lexy::dsl::opt(lexy::dsl::semicolon));
 
-				return lexy::dsl::try_(curly_bracket, lexy::dsl::find(right_brace));
+				return curly_bracket;
 			}();
 
 			static constexpr auto value =
@@ -346,9 +341,6 @@ namespace ovdl::v2script::grammar {
 						} else {
 							return state.ast().template create<ast::ListValue>(ovdl::NodeLocation::make_from(begin, end), LEXY_MOV(list));
 						}
-					},
-					[](detail::IsParseState auto& state, lexy::nullopt fail = {}) {
-						return fail;
 					});
 		};
 	};