From 04206efb102990c2ce0fa26be30db803cdc816be Mon Sep 17 00:00:00 2001
From: mattkae <mattkae@protonmail.com>
Date: Sun, 30 Apr 2023 11:15:18 -0400
Subject: Moving over to CMakeList for qcreator support

---
 .gitignore        |  3 ++-
 CMakeLists.txt    | 13 +++++++++++++
 Makefile          | 45 -------------------------------------------
 src/tokenizer.cpp | 57 ++++++++++++++++++++++++++++++++++++++++++++-----------
 4 files changed, 61 insertions(+), 57 deletions(-)
 create mode 100644 CMakeLists.txt
 delete mode 100644 Makefile

diff --git a/.gitignore b/.gitignore
index 704b56a..e926f51 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,4 +1,5 @@
 .cproject
 .project
 all
-build
\ No newline at end of file
+build
+CMakeLists.txt.user
\ No newline at end of file
diff --git a/CMakeLists.txt b/CMakeLists.txt
new file mode 100644
index 0000000..a65ed8b
--- /dev/null
+++ b/CMakeLists.txt
@@ -0,0 +1,13 @@
+cmake_minimum_required(VERSION 3.5)
+
+project(html_parser LANGUAGES CXX)
+
+set(CMAKE_CXX_STANDARD 17)
+set(CMAKE_CXX_STANDARD_REQUIRED ON)
+
+add_executable(html_parser src/main.cpp src/html_token.cpp src/tokenizer.cpp)
+
+target_link_libraries(html_parser -lmatte)
+
+install(TARGETS html_parser
+    LIBRARY DESTINATION ${CMAKE_INSTALL_LIBDIR})
diff --git a/Makefile b/Makefile
deleted file mode 100644
index e453aa6..0000000
--- a/Makefile
+++ /dev/null
@@ -1,45 +0,0 @@
-TARGET_EXEC ?= html_parser
-
-BUILD_DIR ?= ./build
-SRC_DIRS ?= ./src
-
-CC   := g++ -g
-SRCS := $(shell find $(SRC_DIRS) -name *.cpp -or -name *.c -or -name *.s)
-OBJS := $(SRCS:%=$(BUILD_DIR)/%.o)
-DEPS := $(OBJS:.o=.d)
-
-INC_DIRS := $(shell find $(SRC_DIRS) -type d)
-INC_FLAGS := $(addprefix -I,$(INC_DIRS))
-
-CPPFLAGS ?= $(INC_FLAGS) -MMD -MP
-
-$(BUILD_DIR)/$(TARGET_EXEC): $(OBJS)
-	$(CC) $(OBJS) -o $@ -lmatte
-
-# assembly
-$(BUILD_DIR)/%.s.o: %.s
-	$(MKDIR_P) $(dir $@)
-	$(AS) $(ASFLAGS) -c $< -o $@
-
-# c source
-$(BUILD_DIR)/%.c.o: %.c
-	$(MKDIR_P) $(dir $@)
-	$(CC) $(CPPFLAGS) $(CFLAGS) -c $< -o $@
-
-# c++ source
-$(BUILD_DIR)/%.cpp.o: %.cpp
-	$(MKDIR_P) $(dir $@)
-	$(CXX) $(CPPFLAGS) $(CXXFLAGS) -c $< -o $@
-
-
-.PHONY: clean
-
-clean:
-	$(RM) -r $(BUILD_DIR)
-
-all:
-	$(BUILD_DIR)/$(TARGET_EXEC)
-
--include $(DEPS)
-
-MKDIR_P ?= mkdir -p
diff --git a/src/tokenizer.cpp b/src/tokenizer.cpp
index 30ca6bd..8cf9b31 100644
--- a/src/tokenizer.cpp
+++ b/src/tokenizer.cpp
@@ -1,3 +1,4 @@
+// https://dev.w3.org/html5/spec-LC/tokenization.html#tokenization
 #include "tokenizer.hpp"
 #include "code_point.h"
 #include "html_token.hpp"
@@ -37,14 +38,30 @@ namespace CodePoints {
     const code_point_t UPPERCASE_X = 0x0058;
     const code_point_t DIGIT_ZERO = 0x0030;
     const code_point_t DIGIT_NINE = 0x0039;
+
+    inline bool is_decimal(code_point_t c) {
+        return c >= CodePoints::DIGIT_ZERO && c <= CodePoints::DIGIT_NINE;
+    }
+
+    inline bool is_hex(code_point_t c) {
+        return (c >= CodePoints::DIGIT_ZERO && c <= CodePoints::DIGIT_NINE)
+            || (c >= CodePoints::LOWERCASE_A && c <= CodePoints::LOWERCASE_F)
+            || (c >= CodePoints::UPPERCASE_A && c <= CodePoints::UPPERCASE_F);
+    }
 };
 
-// https://dev.w3.org/html5/spec-LC/tokenization.html#tokenization
+
+// Helpers
+
 
 inline void consume_next(Tokenizer* tokenizer) {
     tokenizer->ptr++;
 }
 
+inline void unconsume_previous(Tokenizer* tokenizer) {
+    tokenizer->ptr--;
+}
+
 
 /// https://dev.w3.org/html5/spec-LC/tokenization.html#data-state
 inline void data_state(Tokenizer* tokenizer) {
@@ -202,19 +219,37 @@ inline void try_consume_character_reference(Tokenizer* tokenizer) {
         consume_next(tokenizer);
         c = *tokenizer->ptr;
 
-        bool is_hex = false; // If set to true, we should interpret within the range 0 to F as hex, otherwise 0 to 9 as decimal
-        switch (c) {
-        case CodePoints::UPPERCASE_X:
-        case CodePoints::LOWERCASE_X: {
+        bool none_match_range = false;
+        code_point_t value = 0x0000;
+        if (c == CodePoints::UPPERCASE_X || c == CodePoints::LOWERCASE_X) {
             consume_next(tokenizer);
-            is_hex = true;
-            break;
-        }
-        default: {
-            is_hex = false;
-            break;
+            c = *tokenizer->ptr;
+
+            if (!CodePoints::is_hex(c)) {
+                unconsume_previous(tokenizer); // X
+                unconsume_previous(tokenizer); // Number sign
+                // TODO: @Error parse error
+                return;
+            }
+
+            while (CodePoints::is_hex(c)) {
+                consume_next(tokenizer);
+                c = *tokenizer->ptr;
+            }
         }
+        else {
+            if (!CodePoints::is_decimal(c)) {
+                unconsume_previous(tokenizer); // Number sign
+                // TODO: @Error parse error
+                return;
+            }
+            
+            while (CodePoints::is_decimal(c)) {
+                consume_next(tokenizer);
+                c = *tokenizer->ptr;
+            }
         }
+
         break;
     }
     }
-- 
cgit v1.2.1