update
This commit is contained in:
Valk Richard Li
GitHub
parent
54c855e17e
commit
c1803a455d
|
|
@ -4,9 +4,10 @@
|
|||
#include <iostream>
|
||||
#include <fstream>
|
||||
#include <cstring>
|
||||
#include <cstdlib>
|
||||
#include <ctime>
|
||||
#include <cmath>
|
||||
// if thread is used, don't forget to add -std=c++11 or higher standard before executing
|
||||
/* if thread is used, don't forget to add -std=c++11 or higher standard before executing */
|
||||
// #include <thread>
|
||||
#include <list>
|
||||
#include <stack>
|
||||
|
|
@ -20,6 +21,7 @@
|
|||
including a function that print the hex format number of an integer
|
||||
*/
|
||||
#include "nasal_misc.h"
|
||||
|
||||
/*
|
||||
nasal_enum.h
|
||||
including enums of: lexer token type,parse generated type,scalar type
|
||||
|
|
@ -29,12 +31,14 @@
|
|||
scalar type is used in nasal_runtime and nasal_gc
|
||||
*/
|
||||
#include "nasal_enum.h"
|
||||
|
||||
/*
|
||||
nasal_ast.h
|
||||
including a class named abstract_syntax_tree
|
||||
this class is frequently used in nasal_parse nasal_runtime
|
||||
*/
|
||||
#include "nasal_ast.h"
|
||||
|
||||
/*
|
||||
nasal_lexer.h
|
||||
including a class named resource_file
|
||||
|
|
@ -45,6 +49,7 @@
|
|||
including a function named is_reserve_word, checking if an identifier is a reserve word
|
||||
*/
|
||||
#include "nasal_lexer.h"
|
||||
|
||||
/*
|
||||
nasal_parse.h
|
||||
including a class named nasal_parse
|
||||
|
|
@ -53,6 +58,7 @@
|
|||
if parse errors occur,this enum will be into use
|
||||
*/
|
||||
#include "nasal_parse.h"
|
||||
|
||||
/*
|
||||
nasal_gc.h(garbage collector and memory manager of nasal_runtime)
|
||||
including basic classed named: nasal_number, nasal_string, nasal_vector, nasal_hash, nasal_function
|
||||
|
|
@ -61,6 +67,14 @@
|
|||
nasal_gc is an object of class gc_manager,and nasal_runtime uses this object as it's memory manager
|
||||
*/
|
||||
#include "nasal_gc.h"
|
||||
|
||||
/*
|
||||
nasal_builtinfunc.h
|
||||
including built-in functions of nasal lib
|
||||
all functions in this .cpp will be used in nasal_runtime::builtin_function()
|
||||
*/
|
||||
#include "nasal_builtinfunc.h"
|
||||
|
||||
/*
|
||||
nasal_runtime.h
|
||||
including a class named nasal_runtime
|
||||
|
|
|
|||
|
|
@ -2,7 +2,38 @@
|
|||
#define __NASAL_ENUM_H__
|
||||
|
||||
// lexer token type is only used in nasal_lexer
|
||||
// each scanned token will be recognized as one of these below
|
||||
/*
|
||||
__token_reserve_word:
|
||||
for,foreach,forindex,while : loop head
|
||||
var,func : definition
|
||||
break,continue : in loop
|
||||
return : in function
|
||||
if,else,elsif : conditional expr
|
||||
and,or : calculation
|
||||
nil : special type
|
||||
__token_identifier:
|
||||
must begin with '_' or 'a'~'z' or 'A'~'Z'
|
||||
can include '_' or 'a'~'z' or 'A'~'Z' or '0'~'9'
|
||||
__token_string:
|
||||
example:
|
||||
"string"
|
||||
'string'
|
||||
if a string does not end with " or ' then lexer will throw an error
|
||||
__token_number:
|
||||
example:
|
||||
2147483647 (integer)
|
||||
2.71828 (float)
|
||||
0xdeadbeef (hex) or 0xDEADBEEF (hex)
|
||||
0o170001 (oct)
|
||||
1e-1234 (dec) or 10E2 (dec)
|
||||
__token_operator:
|
||||
! + - * / ~
|
||||
= += -= *= /= ~=
|
||||
== != > >= < <=
|
||||
('and' 'or' are operators too but they are recognized as operator in generate_detail_token())
|
||||
() [] {} ; , . : ?
|
||||
others: __unknown_operator
|
||||
*/
|
||||
enum lexer_token_type
|
||||
{
|
||||
__token_reserve_word=1,
|
||||
|
|
@ -29,40 +60,39 @@ enum parse_gen_type
|
|||
{
|
||||
/*
|
||||
stack end is an important flag for parse token stack to
|
||||
check if it's stack is at end
|
||||
check if it's stack is at end/empty
|
||||
if stack is empty,the parser will get a wrong memory space and cause SIGSEGV
|
||||
*/
|
||||
__stack_end=1,
|
||||
// operators == != < <= > >=
|
||||
|
||||
__cmp_equal,
|
||||
__cmp_not_equal,
|
||||
__cmp_less,__cmp_less_or_equal,
|
||||
__cmp_more,__cmp_more_or_equal,
|
||||
// operators and or ! + - * / ~
|
||||
__cmp_more,__cmp_more_or_equal, // operators == != < <= > >=
|
||||
|
||||
__and_operator, __or_operator,__nor_operator,
|
||||
__add_operator,__sub_operator,
|
||||
__mul_operator,__div_operator,__link_operator,
|
||||
// operators = += -= *= /= ~=
|
||||
__mul_operator,__div_operator,__link_operator, // operators and or ! + - * / ~
|
||||
|
||||
__equal,
|
||||
__add_equal,__sub_equal,
|
||||
__mul_equal,__div_equal,__link_equal,
|
||||
// operators {} [] () ; , : . ?
|
||||
__mul_equal,__div_equal,__link_equal, // operators = += -= *= /= ~=
|
||||
|
||||
__left_brace,__right_brace, // {}
|
||||
__left_bracket,__right_bracket, // []
|
||||
__left_curve,__right_curve, // ()
|
||||
__semi,__comma,__colon,__dot,__ques_mark, // ; , : . ?
|
||||
__unknown_operator,
|
||||
// reserve words
|
||||
|
||||
__var,
|
||||
__func,__return,__nil,
|
||||
__if,__elsif,__else,
|
||||
__continue,__break,
|
||||
__for,__forindex,__foreach,__while,
|
||||
__for,__forindex,__foreach,__while, // reserve words
|
||||
|
||||
// basic scalar type: number string
|
||||
__number,__string,
|
||||
// basic identifier type: identifier dynamic_identifier
|
||||
__id,__dynamic_id,
|
||||
__number,__string, // basic scalar type: number string
|
||||
|
||||
__id,__dynamic_id, // basic identifier type: identifier dynamic_identifier
|
||||
|
||||
// abstract_syntax_tree type below
|
||||
// abstract_syntax_tree also uses the types above, such as operators
|
||||
|
|
|
|||
|
|
@ -1,38 +1,17 @@
|
|||
#ifndef __NASAL_LEXER_H__
|
||||
#define __NASAL_LEXER_H__
|
||||
|
||||
/*
|
||||
__token_reserve_word:
|
||||
for,foreach,forindex,while : loop
|
||||
var,func : definition
|
||||
break,continue : in loop
|
||||
return : in function
|
||||
if,else,elsif : conditional expr
|
||||
and,or : calculation
|
||||
nil : special type
|
||||
__token_identifier:
|
||||
must begin with '_' or 'a'~'z' or 'A'~'Z'
|
||||
can include '_' or 'a'~'z' or 'A'~'Z' or '0'~'9'
|
||||
__token_string:
|
||||
example:
|
||||
"string"
|
||||
'string'
|
||||
if a string does not end with " or ' then lexer will throw an error
|
||||
__token_number:
|
||||
example:
|
||||
2147483647 (integer)
|
||||
2.71828 (float)
|
||||
0xdeadbeef (hex) or 0xDEADBEEF (hex)
|
||||
0o170001 (oct)
|
||||
1e-1234 (dec) or 10E2 (dec)
|
||||
__token_operator:
|
||||
! + - * / ~
|
||||
= += -= *= /= ~=
|
||||
== != > >= < <=
|
||||
('and' 'or' are operators too but they are recognized as operator in generate_detail_token())
|
||||
() [] {} ; , . : ?
|
||||
others: __unknown_operator
|
||||
*/
|
||||
#define IS_IDENTIFIER_HEAD(c) (c=='_')||('a'<=c && c<='z')||('A'<=c&&c<='Z')
|
||||
#define IS_IDENTIFIER_BODY(c) (c=='_')||('a'<=c && c<='z')||('A'<=c&&c<='Z')||('0'<=c&&c<='9')
|
||||
#define IS_NUMBER_HEAD(c) ('0'<=c&&c<='9')
|
||||
#define IS_NUMBER_BODY(c) ('0'<=c&&c<='9')||('a'<=c&&c<='f')||('A'<=c&&c<='F')||(c=='e'||c=='E'||c=='.'||c=='x'||c=='o')
|
||||
#define IS_STRING_HEAD(c) (c=='\''||c=='\"')
|
||||
// single operators have only one character
|
||||
#define IS_SINGLE_OPRATOR(c) (c=='('||c==')'||c=='['||c==']'||c=='{'||c=='}'||c==','||c==';'||c=='|'||c==':'||\
|
||||
c=='?'||c=='.'||c=='`'||c=='&'||c=='@'||c=='%'||c=='$'||c=='^'||c=='\\')
|
||||
// calculation operators may have two chars, for example: += -= *= /= ~= != == >= <=
|
||||
#define IS_CALC_OPERATOR(c) (c=='='||c=='+'||c=='-'||c=='*'||c=='!'||c=='/'||c=='<'||c=='>'||c=='~')
|
||||
#define IS_NOTE_HEAD(c) (c=='#')
|
||||
|
||||
/* filenames of lib files */
|
||||
#ifndef LIB_FILE_NUM
|
||||
|
|
@ -52,6 +31,7 @@ const std::string lib_filename[LIB_FILE_NUM]=
|
|||
"lib/utf8.nas"
|
||||
};
|
||||
#endif
|
||||
|
||||
/* reserve words */
|
||||
#ifndef RESERVE_WORD_NUM
|
||||
#define RESERVE_WORD_NUM 15
|
||||
|
|
@ -141,6 +121,7 @@ void resource_file::delete_all_source()
|
|||
std::vector<char> tmp;
|
||||
source_code.clear();
|
||||
source_code.swap(tmp);
|
||||
// use tmp's destructor to delete the memory space that source_code used before
|
||||
return;
|
||||
}
|
||||
void resource_file::input_file(std::string filename)
|
||||
|
|
@ -156,9 +137,7 @@ void resource_file::input_file(std::string filename)
|
|||
while(!fin.eof())
|
||||
{
|
||||
c=fin.get();
|
||||
if(fin.eof())
|
||||
break;
|
||||
//source_code.push_back(c<0? '?':c);
|
||||
if(fin.eof()) break;
|
||||
source_code.push_back(c);
|
||||
}
|
||||
fin.close();
|
||||
|
|
@ -178,9 +157,8 @@ void resource_file::load_lib_file()
|
|||
while(!fin.eof())
|
||||
{
|
||||
c=fin.get();
|
||||
if(fin.eof())
|
||||
break;
|
||||
source_code.push_back(c<0? '?':c);
|
||||
if(fin.eof()) break;
|
||||
source_code.push_back(c);
|
||||
}
|
||||
}
|
||||
fin.close();
|
||||
|
|
@ -193,26 +171,25 @@ std::vector<char>& resource_file::get_source()
|
|||
}
|
||||
void resource_file::print_resource()
|
||||
{
|
||||
int size=source_code.size();
|
||||
int line=1;
|
||||
std::cout<<line<<"\t";
|
||||
for(int i=0;i<source_code.size();++i)
|
||||
for(int i=0;i<size;++i)
|
||||
{
|
||||
if(32<=source_code[i])
|
||||
std::cout<<source_code[i];
|
||||
if(32<=source_code[i]) std::cout<<source_code[i];
|
||||
else if(source_code[i]<0)
|
||||
{
|
||||
// print unicode
|
||||
std::string tmp="";
|
||||
for(;i<source_code.size();++i)
|
||||
for(;i<size;++i)
|
||||
{
|
||||
if(source_code[i]>=0)
|
||||
break;
|
||||
if(source_code[i]>=0) break;
|
||||
tmp.push_back(source_code[i]);
|
||||
}
|
||||
std::cout<<tmp;--i;
|
||||
}
|
||||
else
|
||||
std::cout<<" ";
|
||||
if(source_code[i]=='\n')
|
||||
else std::cout<<" ";
|
||||
if(i<size && source_code[i]=='\n')
|
||||
{
|
||||
++line;
|
||||
std::cout<<std::endl<<line<<"\t";
|
||||
|
|
@ -225,42 +202,27 @@ void resource_file::print_resource()
|
|||
std::string nasal_lexer::identifier_gen(std::vector<char>& res,int& ptr,int& line)
|
||||
{
|
||||
std::string token_str="";
|
||||
while(res[ptr]=='_' || ('a'<=res[ptr] && res[ptr]<='z') || ('A'<=res[ptr] && res[ptr]<='Z') || ('0'<=res[ptr] && res[ptr]<='9'))
|
||||
while(IS_IDENTIFIER_BODY(res[ptr]))
|
||||
{
|
||||
token_str+=res[ptr];
|
||||
++ptr;
|
||||
if(ptr>=res.size())
|
||||
break;
|
||||
if(ptr>=res.size()) break;
|
||||
}
|
||||
// check dynamic identifier "..."
|
||||
if(res[ptr]=='.')
|
||||
{
|
||||
++ptr;
|
||||
if(ptr<res.size() && res[ptr]=='.')
|
||||
{
|
||||
++ptr;
|
||||
if(ptr<res.size() && res[ptr]=='.')
|
||||
if(ptr+2<res.size() && res[ptr]=='.' && res[ptr+1]=='.' && res[ptr+2]=='.')
|
||||
{
|
||||
token_str+="...";
|
||||
++ptr;
|
||||
}
|
||||
else
|
||||
ptr-=2;
|
||||
}
|
||||
else
|
||||
--ptr;
|
||||
ptr+=3;
|
||||
}
|
||||
return token_str;
|
||||
// after running this process, ptr will point to the next token's beginning character
|
||||
}
|
||||
|
||||
std::string nasal_lexer::number_gen(std::vector<char>& res,int& ptr,int& line)
|
||||
{
|
||||
bool scientific_notation=false;
|
||||
bool scientific_notation=false;// numbers like 1e8 are scientific_notation
|
||||
std::string token_str="";
|
||||
while(('0'<=res[ptr] && res[ptr]<='9') ||
|
||||
('a'<=res[ptr] && res[ptr]<='f') ||
|
||||
('A'<=res[ptr] && res[ptr]<='F') ||
|
||||
res[ptr]=='.' || res[ptr]=='x' || res[ptr]=='o' ||
|
||||
res[ptr]=='e' || res[ptr]=='E')
|
||||
while(IS_NUMBER_BODY(res[ptr]))
|
||||
{
|
||||
token_str+=res[ptr];
|
||||
if(res[ptr]=='e' || res[ptr]=='E')
|
||||
|
|
@ -299,14 +261,12 @@ std::string nasal_lexer::string_gen(std::vector<char>& res,int& ptr,int& line)
|
|||
std::string token_str="";
|
||||
char str_begin=res[ptr];
|
||||
++ptr;
|
||||
if(ptr>=res.size())
|
||||
return token_str;
|
||||
if(ptr>=res.size()) return token_str;
|
||||
while(ptr<res.size() && res[ptr]!=str_begin)
|
||||
{
|
||||
token_str+=res[ptr];
|
||||
if(res[ptr]=='\n')
|
||||
++line;
|
||||
if(res[ptr]=='\\')
|
||||
if(res[ptr]=='\n') ++line;
|
||||
if(res[ptr]=='\\' && ptr+1<res.size())
|
||||
{
|
||||
++ptr;
|
||||
switch(res[ptr])
|
||||
|
|
@ -321,8 +281,7 @@ std::string nasal_lexer::string_gen(std::vector<char>& res,int& ptr,int& line)
|
|||
}
|
||||
}
|
||||
++ptr;
|
||||
if(ptr>=res.size())
|
||||
break;
|
||||
if(ptr>=res.size()) break;
|
||||
}
|
||||
// check if this string ends with a " or '
|
||||
if(ptr>=res.size())
|
||||
|
|
@ -377,13 +336,12 @@ void nasal_lexer::scanner(std::vector<char>& res)
|
|||
{
|
||||
while(ptr<res.size() && (res[ptr]==' ' || res[ptr]=='\n' || res[ptr]=='\t' || res[ptr]=='\r' || res[ptr]<0))
|
||||
{
|
||||
if(res[ptr]=='\n')
|
||||
++line;
|
||||
// these characters will be ignored, and '\n' will cause ++line
|
||||
if(res[ptr]=='\n') ++line;
|
||||
++ptr;
|
||||
}
|
||||
if(ptr>=res.size())
|
||||
break;
|
||||
if(res[ptr]=='_' || ('a'<=res[ptr] && res[ptr]<='z') || ('A'<=res[ptr] && res[ptr]<='Z'))
|
||||
if(ptr>=res.size()) break;
|
||||
if(IS_IDENTIFIER_HEAD(res[ptr]))
|
||||
{
|
||||
token_str=identifier_gen(res,ptr,line);
|
||||
token new_token;
|
||||
|
|
@ -392,7 +350,7 @@ void nasal_lexer::scanner(std::vector<char>& res)
|
|||
new_token.str=token_str;
|
||||
token_list.push_back(new_token);
|
||||
}
|
||||
else if('0'<=res[ptr] && res[ptr]<='9')
|
||||
else if(IS_NUMBER_HEAD(res[ptr]))
|
||||
{
|
||||
token_str=number_gen(res,ptr,line);
|
||||
token new_token;
|
||||
|
|
@ -401,7 +359,7 @@ void nasal_lexer::scanner(std::vector<char>& res)
|
|||
new_token.str=token_str;
|
||||
token_list.push_back(new_token);
|
||||
}
|
||||
else if(res[ptr]=='\'' || res[ptr]=='\"')
|
||||
else if(IS_STRING_HEAD(res[ptr]))
|
||||
{
|
||||
token_str=string_gen(res,ptr,line);
|
||||
token new_token;
|
||||
|
|
@ -410,10 +368,7 @@ void nasal_lexer::scanner(std::vector<char>& res)
|
|||
new_token.str=token_str;
|
||||
token_list.push_back(new_token);
|
||||
}
|
||||
else if(res[ptr]=='(' || res[ptr]==')' || res[ptr]=='[' || res[ptr]==']' || res[ptr]=='{' ||
|
||||
res[ptr]=='}' || res[ptr]==',' || res[ptr]==';' || res[ptr]=='|' || res[ptr]==':' ||
|
||||
res[ptr]=='?' || res[ptr]=='.' || res[ptr]=='`' || res[ptr]=='&' || res[ptr]=='@' ||
|
||||
res[ptr]=='%' || res[ptr]=='$' || res[ptr]=='^' || res[ptr]=='\\')
|
||||
else if(IS_SINGLE_OPRATOR(res[ptr]))
|
||||
{
|
||||
token_str="";
|
||||
token_str+=res[ptr];
|
||||
|
|
@ -424,8 +379,7 @@ void nasal_lexer::scanner(std::vector<char>& res)
|
|||
token_list.push_back(new_token);
|
||||
++ptr;
|
||||
}
|
||||
else if(res[ptr]=='=' || res[ptr]=='+' || res[ptr]=='-' || res[ptr]=='*' || res[ptr]=='!' ||
|
||||
res[ptr]=='/' || res[ptr]=='<' || res[ptr]=='>' || res[ptr]=='~')
|
||||
else if(IS_CALC_OPERATOR(res[ptr]))
|
||||
{
|
||||
// get calculation operator
|
||||
token_str="";
|
||||
|
|
@ -442,7 +396,7 @@ void nasal_lexer::scanner(std::vector<char>& res)
|
|||
new_token.str=token_str;
|
||||
token_list.push_back(new_token);
|
||||
}
|
||||
else if(res[ptr]=='#')
|
||||
else if(IS_NOTE_HEAD(res[ptr]))
|
||||
{
|
||||
// avoid note
|
||||
while(ptr<res.size() && res[ptr]!='\n')
|
||||
|
|
|
|||
|
|
@ -773,7 +773,8 @@ int nasal_runtime::call_function(std::list<std::map<std::string,int> >& local_s
|
|||
nasal_gc.get_scalar(addr).get_function().get_parameter_list(),
|
||||
nasal_gc.get_scalar(addr).get_function().get_statement_block(),
|
||||
*call_node,
|
||||
last_hash_addr);
|
||||
last_hash_addr
|
||||
);
|
||||
if(addr<0)
|
||||
return -1;
|
||||
nasal_gc.reference_delete(tmp_addr);
|
||||
|
|
@ -968,6 +969,8 @@ int nasal_runtime::function_generation(std::list<std::map<std::string,int> >& lo
|
|||
void nasal_runtime::update_closure(std::list<std::map<std::string,int> >& local_scope,int local_scope_addr)
|
||||
{
|
||||
// update_closure
|
||||
// each new function will be updated only once, after updating closure,functions' closure_updated flag will be set true
|
||||
// but this has a bug, if this new function is a member of vector or hash, it will not be updated
|
||||
if(!local_scope.size())
|
||||
return;
|
||||
for(std::map<std::string,int>::iterator i=local_scope.back().begin();i!=local_scope.back().end();++i)
|
||||
|
|
|
|||
Loading…
Reference in New Issue