mr4th-public
/
4coder
1
Fork 0
Code Issues Releases Activity
4coder/4coder_metadata_generator.cpp

793 lines
20 KiB
C++
Raw Blame History

/*
4coder_metadata_generator.cpp - A preprocessor program for generating a list of commands and their descriptions.
TYPE: 'code-preprocessor'
*/
// TOP
#include "4coder_lib/4coder_mem.h"
#define FSTRING_IMPLEMENTATION
#include "4coder_lib/4coder_string.h"
#include "4cpp/4cpp_lexer.h"
#include <stdlib.h>
#include <stdio.h>
#include <stdint.h>
typedef int32_t bool32;
//// WINDOWS BEGIN ////
#define UNICODE
#include <Windows.h>
typedef TCHAR Filename_Character;
//// WINDOWS END ////
struct File_Info{
Filename_Character *name;
int32_t len;
bool32 is_folder;
};
struct File_List{
File_Info *info;
int32_t count;
int32_t final_length;
Filename_Character final_name[4096];
};
static File_List
get_file_list(Partition *part, Filename_Character *dir);
static Filename_Character*
encode(Partition *part, char *str);
static char*
unencode(Partition *part, Filename_Character *str, int32_t len);
//// WINDOWS BEGIN ////
static bool32
is_code_file(Filename_Character *name, int32_t len){
bool32 is_code = false;
if (len >= 5){
Filename_Character *ext = &name[len - 4];
if (ext[0] == '.' && ext[1] == 'c' && ext[2] == 'p' && ext[3] == 'p'){
is_code = true;
}
else if (ext[0] == '.' && ext[1] == 'h' && ext[2] == 'p' && ext[3] == 'p'){
is_code = true;
}
}
if (len >= 4){
Filename_Character *ext = &name[len - 3];
if (ext[0] == '.' && ext[1] == 'c' && ext[2] == 'c'){
is_code = true;
}
}
if (len >= 3){
Filename_Character *ext = &name[len - 2];
if (ext[0] == '.' && ext[1] == 'h'){
is_code = true;
}
else if (ext[0] == '.' && ext[1] == 'c'){
is_code = true;
}
}
return(is_code);
}
static File_List
get_file_list(Partition *part, Filename_Character *dir){
if (part == 0){
fprintf(stdout, "fatal error: NULL part passed to %s\n", __FUNCTION__);
exit(1);
}
if (dir == 0){
fprintf(stdout, "fatal error: NULL dir passed to %s\n", __FUNCTION__);
exit(1);
}
File_List list = {0};
Temp_Memory part_reset = begin_temp_memory(part);
HANDLE dir_handle =
CreateFile(dir,
FILE_LIST_DIRECTORY,
FILE_SHARE_READ | FILE_SHARE_WRITE | FILE_SHARE_DELETE,
0,
OPEN_EXISTING,
FILE_FLAG_BACKUP_SEMANTICS | FILE_FLAG_OVERLAPPED,
0);
if (dir_handle == INVALID_HANDLE_VALUE){
fprintf(stdout, "fatal error: could not open directory handle\n");
exit(1);
}
Filename_Character final_name[4096];
DWORD final_length = GetFinalPathNameByHandle(dir_handle, final_name, sizeof(final_name), 0);
if (final_length > sizeof(final_name)){
fprintf(stdout, "fatal error: path name too long for local buffer\n");
exit(1);
}
CloseHandle(dir_handle);
final_length -= 4;
memmove(final_name, final_name + 4, final_length*sizeof(*final_name));
final_name[final_length] = '\\';
final_name[final_length + 1] = '*';
final_name[final_length + 2] = 0;
WIN32_FIND_DATA find_data = {0};
HANDLE search = FindFirstFile(final_name, &find_data);
if (search == INVALID_HANDLE_VALUE){
fprintf(stdout, "fatal error: could not begin a file search\n");
exit(1);
}
int32_t character_count = 0;
int32_t file_count = 0;
BOOL more_files = true;
do{
Filename_Character *name = &find_data.cFileName[0];
int32_t size = 0;
for(;name[size];++size);
uint32_t attribs = find_data.dwFileAttributes;
bool32 is_folder = ((attribs & FILE_ATTRIBUTE_DIRECTORY) != 0);
if (name[0] != '.' && (is_folder || is_code_file(name, size))){
++file_count;
character_count += size + 1;
}
more_files = FindNextFile(search, &find_data);
}while(more_files);
FindClose(search);
int32_t rounded_char_size = (character_count*sizeof(Filename_Character) + 7)&(~7);
int32_t memsize = rounded_char_size + file_count*sizeof(File_Info);
void *mem = push_array(part, uint8_t, memsize);
if (mem == 0){
fprintf(stdout, "fatal error: not enough memory on the partition for a file list.\n");
exit(1);
}
Filename_Character *char_ptr = (Filename_Character*)mem;
File_Info *info_ptr = (File_Info*)((uint8_t*)mem + rounded_char_size);
Filename_Character *char_ptr_end = (Filename_Character*)info_ptr;
File_Info *info_ptr_end = info_ptr + file_count;
File_Info *info_ptr_base = info_ptr;
search = FindFirstFile(final_name, &find_data);
if (search == INVALID_HANDLE_VALUE){
fprintf(stdout, "fatal error: could not restart a file search\n");
exit(1);
}
int32_t adjusted_file_count = 0;
more_files = true;
do{
Filename_Character *name = &find_data.cFileName[0];
int32_t size = 0;
for(;name[size]!=0;++size);
uint32_t attribs = find_data.dwFileAttributes;
bool32 is_folder = ((attribs & FILE_ATTRIBUTE_DIRECTORY) != 0);
if (name[0] != '.' && (is_folder || is_code_file(name, size))){
if (info_ptr + 1 > info_ptr_end || char_ptr + size + 1 > char_ptr_end){
memset(&list, 0, sizeof(list));
end_temp_memory(part_reset);
FindClose(search);
return(list);
}
info_ptr->name = char_ptr;
info_ptr->len = size;
info_ptr->is_folder = is_folder;
memmove(char_ptr, name, size*sizeof(*name));
char_ptr[size] = 0;
char_ptr += size + 1;
++info_ptr;
++adjusted_file_count;
}
more_files = FindNextFile(search, &find_data);
}while(more_files);
FindClose(search);
list.info = info_ptr_base;
list.count = adjusted_file_count;
list.final_length = final_length;
memcpy(list.final_name, final_name, list.final_length*sizeof(*final_name));
list.final_name[list.final_length] = 0;
return(list);
}
static Filename_Character*
encode(Partition *part, char *str){
int32_t size = 0;
for (;str[size]!=0;++size);
Filename_Character *out = push_array(part, Filename_Character, size + 1);
push_align(part, 8);
if (out == 0){
fprintf(stdout, "fatal error: ran out of memory encoding string to filename\n");
exit(1);
}
for (int32_t i = 0, j = 0; i <= size; ++i){
if (str[i] != '"'){
out[j++] = str[i];
}
}
return(out);
}
static char*
unencode(Partition *part, Filename_Character *str, int32_t len){
Temp_Memory temp = begin_temp_memory(part);
char *out = push_array(part, char, len + 1);
push_align(part, 8);
if (out == 0){
fprintf(stdout, "fatal error: ran out of memory unencoding string to filename\n");
exit(1);
}
for (int32_t i = 0; i <= len; ++i){
if (str[i] <= 127){
out[i] = (char)str[i];
}
else{
out = 0;
end_temp_memory(temp);
break;
}
}
return(out);
}
//// WINDOWS END ////
static String
file_dump(Partition *part, char *name){
String text = {0};
FILE *file = fopen(name, "rb");
if (file != 0){
fseek(file, 0, SEEK_END);
text.size = ftell(file);
fseek(file, 0, SEEK_SET);
text.memory_size = text.size + 1;
text.str = push_array(part, char, text.memory_size);
fread(text.str, 1, text.size, file);
terminate_with_null(&text);
fclose(file);
}
return(text);
}
static void
error(char *source_name, String text, int32_t pos, char *msg){
if (pos < 0){
pos = 0;
}
if (pos > text.size){
pos = text.size;
}
int32_t line_number = 1;
int32_t character_pos = 1;
char *end = text.str + pos;
for (char *p = text.str; p < end; ++p){
if (*p == '\n'){
++line_number;
character_pos = 1;
}
else{
++character_pos;
}
}
fprintf(stdout, "%s:%d:%d: %s\n", source_name, line_number, character_pos, msg);
fflush(stdout);
}
struct Reader{
char *source_name;
String text;
Cpp_Token_Array tokens;
Cpp_Token *ptr;
};
static Reader
make_reader(Cpp_Token_Array array, char *source_name, String text){
Reader reader = {0};
reader.tokens = array;
reader.ptr = array.tokens;
reader.source_name = source_name;
reader.text = text;
return(reader);
}
static Cpp_Token
prev_token(Reader *reader){
Cpp_Token result = {0};
for (;;){
if (reader->ptr > reader->tokens.tokens + reader->tokens.count){
reader->ptr = reader->tokens.tokens + reader->tokens.count;
}
if (reader->ptr > reader->tokens.tokens){
--reader->ptr;
result = *reader->ptr;
}
else{
reader->ptr = reader->tokens.tokens;
memset(&result, 0, sizeof(result));
break;
}
if (result.type != CPP_TOKEN_COMMENT && result.type != CPP_TOKEN_JUNK){
break;
}
}
return(result);
}
static Cpp_Token
get_token(Reader *reader){
Cpp_Token result = {0};
for (;;){
if (reader->ptr < reader->tokens.tokens){
reader->ptr = reader->tokens.tokens;
}
if (reader->ptr < reader->tokens.tokens + reader->tokens.count){
result = *reader->ptr;
++reader->ptr;
}
else{
reader->ptr = reader->tokens.tokens + reader->tokens.count;
memset(&result, 0, sizeof(result));
result.start = reader->text.size;
break;
}
if (result.type != CPP_TOKEN_COMMENT && result.type != CPP_TOKEN_JUNK){
break;
}
}
return(result);
}
static Cpp_Token
peek_token(Reader *reader){
Cpp_Token result = {0};
if (reader->ptr < reader->tokens.tokens){
reader->ptr = reader->tokens.tokens;
}
if (reader->ptr >= reader->tokens.tokens + reader->tokens.count){
result.start = reader->text.size;
}
else{
result = *reader->ptr;
}
return(result);
}
static int32_t
peek_pos(Reader *reader){
Cpp_Token token = peek_token(reader);
return(token.start);
}
static void
error(Reader *reader, int32_t pos, char *msg){
error(reader->source_name, reader->text, pos, msg);
}
struct Temp_Read{
Reader *reader;
Cpp_Token *pos;
};
static Temp_Read
begin_temp_read(Reader *reader){
Temp_Read temp = {0};
temp.reader = reader;
temp.pos = reader->ptr;
return(temp);
}
static void
end_temp_read(Temp_Read temp){
temp.reader->ptr = temp.pos;
}
static String
token_str(String text, Cpp_Token token){
String str = substr(text, token.start, token.size);
return(str);
}
static bool32
require_key_identifier(Reader *reader, char *str){
bool32 success = false;
Cpp_Token token = get_token(reader);
if (token.type == CPP_TOKEN_IDENTIFIER){
String lexeme = token_str(reader->text, token);
if (match(lexeme, str)){
success = true;
}
}
if (!success){
char space[1024];
String s = make_fixed_width_string(space);
copy(&s, "expected to find '");
append(&s, str);
append(&s, "'");
terminate_with_null(&s);
error(reader, token.start, s.str);
}
return(success);
}
static bool32
require_open_parenthese(Reader *reader){
bool32 success = false;
Cpp_Token token = get_token(reader);
if (token.type == CPP_TOKEN_PARENTHESE_OPEN){
success = true;
}
if (!success){
error(reader, token.start, "expected to find '('");
}
return(success);
}
static bool32
require_close_parenthese(Reader *reader){
bool32 success = false;
Cpp_Token token = get_token(reader);
if (token.type == CPP_TOKEN_PARENTHESE_CLOSE){
success = true;
}
if (!success){
error(reader, token.start, "expected to find ')'");
}
return(success);
}
static bool32
require_define(Reader *reader){
bool32 success = false;
Cpp_Token token = get_token(reader);
if (token.type == CPP_PP_DEFINE){
success = true;
}
if (!success){
error(reader, token.start, "expected to find '#define'");
}
return(success);
}
static bool32
extract_identifier(Reader *reader, String *str_out){
bool32 success = false;
Cpp_Token token = get_token(reader);
if (token.type == CPP_TOKEN_IDENTIFIER){
String lexeme = token_str(reader->text, token);
*str_out = lexeme;
success = true;
}
if (!success){
error(reader, token.start, "expected to find an identifier");
}
return(success);
}
static bool32
extract_string(Reader *reader, String *str_out){
bool32 success = false;
Cpp_Token token = get_token(reader);
if (token.type == CPP_TOKEN_STRING_CONSTANT){
String lexeme = token_str(reader->text, token);
*str_out = lexeme;
success = true;
}
if (!success){
error(reader, token.start, "expected to find a string literal");
}
return(success);
}
static bool32
parse_documented_command(Partition *part, Reader *reader){
String name = {0};
String doc = {0};
// Getting the command's name
if (!require_key_identifier(reader, "CUSTOM_COMMAND_SIG")){
return(false);
}
if (!require_open_parenthese(reader)){
return(false);
}
if (!extract_identifier(reader, &name)){
return(false);
}
if (!require_close_parenthese(reader)){
return(false);
}
// Getting the command's doc string
if (!require_key_identifier(reader, "CUSTOM_DOC")){
return(false);
}
if (!require_open_parenthese(reader)){
return(false);
}
if (!extract_string(reader, &doc)){
return(false);
}
if (!require_close_parenthese(reader)){
return(false);
}
// TODO(allen): Store into data structure for codegen.
//error(reader, name_pos, "name of a command");
//error(reader, str_pos, "doc string of a command");
return(true);
}
static bool32
parse_alias(Partition *part, Reader *reader){
String name = {0};
String potential = {0};
// Getting the alias's name
if (!require_define(reader)){
return(false);
}
int32_t name_pos = peek_pos(reader);
if (!extract_identifier(reader, &name)){
return(false);
}
// Getting the alias's target
if (!require_key_identifier(reader, "CUSTOM_ALIAS")){
return(false);
}
if (!require_open_parenthese(reader)){
return(false);
}
int32_t potential_pos = peek_pos(reader);
if (!extract_identifier(reader, &potential)){
return(false);
}
if (!require_close_parenthese(reader)){
return(false);
}
error(reader, name_pos, "name of an alias");
error(reader, potential_pos, "name of a potential");
return(true);
}
static void
parse_text(Partition *part, char *source_name, String text){
Cpp_Token_Array array = cpp_make_token_array(1024);
cpp_lex_file(text.str, text.size, &array);
Reader reader_ = make_reader(array, source_name, text);
Reader *reader = &reader_;
for (;;){
Cpp_Token token = get_token(reader);
if (token.type == CPP_TOKEN_IDENTIFIER){
String lexeme = token_str(text, token);
bool32 in_preproc_body = ((token.flags & CPP_TFLAG_PP_BODY) != 0);
if (!in_preproc_body && match(lexeme, "CUSTOM_DOC")){
Temp_Read temp_read = begin_temp_read(reader);
bool32 found_start_pos = false;
for (int32_t R = 0; R < 5; ++R){
Cpp_Token p_token = prev_token(reader);
if (p_token.type == CPP_TOKEN_IDENTIFIER){
String p_lexeme = token_str(text, p_token);
if (match(p_lexeme, "CUSTOM_COMMAND_SIG")){
found_start_pos = true;
break;
}
}
if (p_token.type == 0){
break;
}
}
if (!found_start_pos){
end_temp_read(temp_read);
}
else{
if (!parse_documented_command(part, reader)){
end_temp_read(temp_read);
}
}
}
else if (match(lexeme, "CUSTOM_ALIAS")){
Temp_Read temp_read = begin_temp_read(reader);
bool32 found_start_pos = false;
for (int32_t R = 0; R < 3; ++R){
Cpp_Token p_token = prev_token(reader);
if (p_token.type == CPP_PP_DEFINE){
if (R == 2){
found_start_pos = true;
}
break;
}
if (p_token.type == 0){
break;
}
}
if (!found_start_pos){
end_temp_read(temp_read);
}
else{
if (!parse_alias(part, reader)){
end_temp_read(temp_read);
}
}
}
}
if (token.type == 0){
break;
}
}
cpp_free_token_array(array);
}
static void
parse_file(Partition *part, Filename_Character *name_, int32_t len){
char *name = unencode(part, name_, len);
if (name == 0){
if (sizeof(*name_) == 2){
fprintf(stdout, "warning: could not unencode file name %ls - file skipped\n", name_);
}
else{
fprintf(stdout, "warning: could not unencode file name %s - file skipped\n", name_);
}
return;
}
String text = file_dump(part, name);
parse_text(part, name, text);
}
static void
parse_files_in_directory(Partition *part, Filename_Character *root, bool32 recursive){
File_List list = get_file_list(part, root);
for (int32_t i = 0; i < list.count; ++i){
File_Info *info = &list.info[i];
int32_t full_name_len = list.final_length + 1 + info->len;
Filename_Character *full_name = push_array(part, Filename_Character, full_name_len + 1);
push_align(part, 8);
if (full_name == 0){
fprintf(stdout, "fatal error: not enough memory to recurse to sub directory\n");
exit(1);
}
memmove(full_name, list.final_name, list.final_length*sizeof(*full_name));
full_name[list.final_length] = '\\';
memmove(full_name + list.final_length + 1, info->name, info->len*sizeof(*full_name));
full_name[full_name_len] = 0;
if (!info->is_folder){
parse_file(part, full_name, full_name_len);
}
else{
parse_files_in_directory(part, full_name, recursive);
}
}
}
static void
show_usage(int argc, char **argv){
char *name = "metadata_generator";
if (argc >= 1){
name = argv[0];
}
fprintf(stdout, "usage:\n%s [-R] <root-directory> [<root-directory2> ...]\n", name);
exit(0);
}
int
main(int argc, char **argv){
if (argc < 2){
show_usage(argc, argv);
}
bool32 recursive = match(argv[1], "-R");
if (recursive && argc < 3){
show_usage(argc, argv);
}
int32_t size = (256 << 20);
void *mem = malloc(size);
Partition part_ = make_part(mem, size);
Partition *part = &part_;
int32_t start_i = 1;
if (recursive){
start_i = 2;
}
for (int32_t i = start_i; i < argc; ++i){
Filename_Character *root_name = encode(part, argv[i]);
parse_files_in_directory(part, root_name, recursive);
}
return(0);
}
// BOTTOM
Reference in New Issue View Git Blame Copy Permalink