6 files changed, 190 insertions, 0 deletions
diff --git a/include/lang_analysis/ast.h b/include/lang_analysis/ast.h
new file mode 100644
index 0000000..9e5a717
--- /dev/null
+++ b/include/lang_analysis/ast.h
@@ -0,0 +1,30 @@
+#pragma once
+
+#include <string>
+
+#include "lang_analysis/lexer_class.h"
+#include "lang_analysis/enum_type.h"
+#include "lang_analysis/token.h"
+
+using namespace std; // remove this
+
+namespace ov4
+{
+
+class T_lexer::T_ast
+{
+public:
+    T_ast() = default;
+
+    enum_token_type token_type;
+    string command_text;
+    size_t left = string::npos;
+    size_t right = string::npos;
+
+    // if this is marked as `subshell`
+    // it will fork a subshell
+    // then the subshell will handler the rest
+    bool subshell;
+};
+
+}
diff --git a/include/lang_analysis/enum_type.h b/include/lang_analysis/enum_type.h
new file mode 100644
index 0000000..bfc7cdc
--- /dev/null
+++ b/include/lang_analysis/enum_type.h
@@ -0,0 +1,35 @@
+#pragma once
+
+namespace ov4
+{
+
+enum enum_token_type
+{
+    TEXT,                   //    foo
+    SINGLE_QUOTATION,       //    '
+    DOUBLE_QUOTATION,       //    "
+    LOGIC_AND,              //    &&
+    LOGIC_OR,               //    ||
+    ASYNC,                  //    &
+    PIPE,                   //    |    
+    SEMICOLON,              //    ;
+    EXPAND,                 //    $
+    LEFT_BRACKET,           //    (
+    RIGHT_BRACKET,          //    )
+};
+
+enum guard_type
+{
+    GUARD_OFF,
+    GUARD_WEAK,
+    GUARD_STRONG,
+};
+
+enum class child_existence_type
+{
+    NOT_EXIST,
+    OPTIONAL,
+    EXIST,
+};
+
+}
diff --git a/include/lang_analysis/lexer_class.h b/include/lang_analysis/lexer_class.h
new file mode 100644
index 0000000..4100b97
--- /dev/null
+++ b/include/lang_analysis/lexer_class.h
@@ -0,0 +1,42 @@
+#pragma once
+
+#include <string>
+#include <unordered_map>
+#include <vector>
+
+#include "lang_analysis/enum_type.h"
+
+using namespace std;
+
+namespace ov4
+{
+
+class T_lexer
+{
+public:
+
+    // token.h
+    class T_token;
+    string trim(const string &s, const string &trim_target = " \t");
+    bool is_single_symbol(char c);
+    void bracket_depth_changer(char c, int &bracket_depth);
+    bool token_continue(const string &s, char next);
+    void token_push(const string &push_s, int bracket_depth, bool force_text = false);
+    int tokenizer(const string &s);
+
+    // ast.h
+    class T_ast;
+    string final_trim(size_t _begin, size_t _end);
+    size_t alloc_ast();
+    int get_pivot_order(enum_token_type x);
+    bool is_operator(enum_token_type x);
+    void parse(size_t cmd_begin, size_t cmd_end, size_t ast_vec);
+
+    // shared
+    std::vector<T_ast> ast;
+    std::vector<T_token> token;
+    size_t ast_index = 0;
+
+};
+
+}
diff --git a/include/lang_analysis/parseline.h b/include/lang_analysis/parseline.h
new file mode 100644
index 0000000..e52b50e
--- /dev/null
+++ b/include/lang_analysis/parseline.h
@@ -0,0 +1,10 @@
+#pragma once
+
+#include <string>
+
+namespace ov4
+{
+
+char **parseline(const std::string &s);
+
+}
diff --git a/include/lang_analysis/symbol_property.h b/include/lang_analysis/symbol_property.h
new file mode 100644
index 0000000..f21b145
--- /dev/null
+++ b/include/lang_analysis/symbol_property.h
@@ -0,0 +1,48 @@
+#pragma once
+
+#include <unordered_map>
+#include <string>
+
+#include "lang_analysis/enum_type.h"
+
+using namespace std;
+
+namespace ov4
+{
+
+class T_property
+{
+public:
+    bool continuable = true;
+
+    // -1 means it's NOT operator
+    // any positive values mean it's operator
+    int pivot_priority = -1;
+
+    // require: MUST
+    // accept: optional, but if it doesn't accept, but there is indeed a child, then error
+    child_existence_type left;
+    child_existence_type right;
+
+    enum_token_type enum_type = TEXT;
+};
+
+inline const unordered_map<string, T_property> symbol_property = {
+
+    // bracket
+    {"(",    {.continuable = false, .enum_type = LEFT_BRACKET}},
+    {")",    {.continuable = false, .enum_type = RIGHT_BRACKET}},
+
+    // operator
+    {"&",    {.continuable = true, .pivot_priority = 2000, .left = child_existence_type::EXIST,     .right = child_existence_type::NOT_EXIST, .enum_type = ASYNC}},
+    {"|",    {.continuable = true, .pivot_priority = 100,  .left = child_existence_type::EXIST,     .right = child_existence_type::EXIST,     .enum_type = PIPE}},
+    {"&&",   {.continuable = true, .pivot_priority = 1000, .left = child_existence_type::EXIST,     .right = child_existence_type::EXIST,     .enum_type = LOGIC_AND}},
+    {"||",   {.continuable = true, .pivot_priority = 1000, .left = child_existence_type::EXIST,     .right = child_existence_type::EXIST,     .enum_type = LOGIC_OR}},
+
+};
+
+string symbol_enum2string(enum_token_type e);
+
+bool symbol_contains_enum_type(enum_token_type e);
+    
+}
diff --git a/include/lang_analysis/token.h b/include/lang_analysis/token.h
new file mode 100644
index 0000000..508d142
--- /dev/null
+++ b/include/lang_analysis/token.h
@@ -0,0 +1,25 @@
+#pragma once
+
+#include <string>
+
+#include "lang_analysis/lexer_class.h"
+#include "lang_analysis/enum_type.h"
+#include "lang_analysis/symbol_property.h"
+
+using namespace std; // remove this
+
+namespace ov4
+{
+
+class T_lexer::T_token
+{
+public:
+    T_token() = default;
+
+    enum_token_type token_type;
+    string text;
+    guard_type guard = GUARD_OFF;
+    int bracket_depth = 0;
+};
+
+}