ps
/
4coder
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

wrote lexer function docs, expanded doc generator

This commit is contained in:
Allen Webster 2016-09-06 17:22:35 -04:00
parent 161c3cc871
commit 8cf49b566e
12 changed files with 1000 additions and 630 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

411
4coder_API.html
View File

File diff suppressed because one or more lines are too long

488
4cpp_lexer.h
View File

@ -12,8 +12,9 @@
# define FCPP_LINK static
#endif
#define FCPP_INTERNAL FCPP_LINK
#include "4cpp_lexer_types.h"
#include "4cpp_lexer_fsms.h"
#include "4cpp_lexer_tables.c"
// TODO(allen): revisit this keyword data declaration system
@ -145,12 +146,26 @@ static String_And_Flag keywords[] = {
FCPP_LINK Cpp_Get_Token_Result
cpp_get_token(Cpp_Token_Stack *token_stack, int32_t pos){
cpp_get_token(Cpp_Token_Array *token_array_in, int32_t pos)/*
DOC_PARAM(token_array, The array of tokens from which to get a token.)
DOC_PARAM(pos, The position, measured in bytes, to get the token for.)
DOC_RETURN(A Cpp_Get_Token_Result struct is returned containing the index
of a token and a flag indicating whether the pos is contained in the token
or in whitespace after the token.)
DOC(This call performs a binary search over all of the tokens looking
for the token that contains the specified position. If the position
is in whitespace between the tokens, the returned token index is the
index of the token immediately before the provided position. The returned
index can be -1 if the position is before the first token.)
DOC_SEE(Cpp_Get_Token_Result)
*/{
Cpp_Get_Token_Result result = {};
Cpp_Token *token_array = token_stack->tokens;
Cpp_Token *token_array = token_array_in->tokens;
Cpp_Token *token = 0;
int32_t first = 0;
int32_t count = token_stack->count;
int32_t count = token_array_in->count;
int32_t last = count;
int32_t this_start = 0, next_start = 0;
@ -200,19 +215,23 @@ cpp_get_token(Cpp_Token_Stack *token_stack, int32_t pos){
return(result);
}
FCPP_LINK void
cpp_shift_token_starts(Cpp_Token_Stack *stack, int32_t from_token_i, int32_t shift_amount){
Cpp_Token *token = stack->tokens + from_token_i;
int32_t count = stack->count, i;
for (i = from_token_i; i < count; ++i, ++token){
token->start += shift_amount;
}
FCPP_LINK Cpp_Lex_Data
cpp_lex_data_init(char *mem_buffer)/*
DOC_PARAM(tb, The memory to use for initializing the lex state's temp memory buffer.)
DOC_RETURN(A brand new lex state ready to begin lexing a file from the beginning.)
DOC(Creates a new lex state in the form of a Cpp_Lex_Data struct and returns the struct.
The system needs a temporary buffer that is as long as the longest token. 4096 is usually
enough but the buffer is not checked, so to be 100% bullet proof it has to be the same length
as the file being lexed.)
*/{
Cpp_Lex_Data data = {0};
data.tb = mem_buffer;
return(data);
}
FCPP_LINK Lex_PP_State
FCPP_INTERNAL Cpp_Lex_PP_State
cpp_pp_directive_to_state(Cpp_Token_Type type){
Lex_PP_State result = LSPP_default;
Cpp_Lex_PP_State result = LSPP_default;
switch (type){
case CPP_PP_INCLUDE:
case CPP_PP_IMPORT:
@ -256,85 +275,27 @@ cpp_pp_directive_to_state(Cpp_Token_Type type){
return(result);
}
FCPP_LINK int32_t
cpp_place_token_nonalloc(Cpp_Token *out_tokens, int32_t token_i, Cpp_Token token){
Cpp_Token prev_token = {(Cpp_Token_Type)0};
if (token_i > 0){
prev_token = out_tokens[token_i - 1];
}
out_tokens[token_i++] = token;
return(token_i);
}
FCPP_LINK bool
cpp_push_token_nonalloc(Cpp_Token_Stack *out_tokens, Cpp_Token token){
bool result = 0;
if (out_tokens->count == out_tokens->max_count){
out_tokens->count =
cpp_place_token_nonalloc(out_tokens->tokens, out_tokens->count, token);
result = 1;
}
return(result);
}
struct Lex_Data{
char *tb;
int32_t tb_pos;
int32_t token_start;
int32_t pos;
int32_t pos_overide;
int32_t chunk_pos;
Lex_FSM fsm;
unsigned char white_done;
unsigned char pp_state;
unsigned char completed;
Cpp_Token token;
int32_t __pc__;
};
FCPP_LINK Lex_Data
lex_data_init(char *tb){
Lex_Data data = {0};
data.tb = tb;
return(data);
}
#define DrCase(PC) case PC: goto resumespot_##PC
#define DrYield(PC, n) { \
token_stack_out->count = token_i; \
token_array_out->count = token_i; \
*S_ptr = S; S_ptr->__pc__ = PC; return(n); resumespot_##PC:; }
#define DrReturn(n) { \
token_stack_out->count = token_i; \
token_array_out->count = token_i; \
*S_ptr = S; S_ptr->__pc__ = -1; return(n); }
enum Lex_Result{
LexFinished,
LexNeedChunk,
LexNeedTokenMemory,
LexHitTokenLimit
};
FCPP_LINK int32_t
cpp_lex_nonalloc(Lex_Data *S_ptr,
char *chunk, int32_t size,
Cpp_Token_Stack *token_stack_out){
Lex_Data S = *S_ptr;
FCPP_INTERNAL Cpp_Lex_Result
cpp_lex_nonalloc_null_end_no_limit(Cpp_Lex_Data *S_ptr, char *chunk, int32_t size,
Cpp_Token_Array *token_array_out){
Cpp_Lex_Data S = *S_ptr;
Cpp_Token *out_tokens = token_stack_out->tokens;
int32_t token_i = token_stack_out->count;
int32_t max_token_i = token_stack_out->max_count;
//Pos_Update_Rule pos_update_rule = PUR_none;
Cpp_Token *out_tokens = token_array_out->tokens;
int32_t token_i = token_array_out->count;
int32_t max_token_i = token_array_out->max_count;
char c = 0;
@ -362,7 +323,7 @@ cpp_lex_nonalloc(Lex_Data *S_ptr,
if (S.white_done == 0){
S.chunk_pos += size;
DrYield(4, LexNeedChunk);
DrYield(4, LexResult_NeedChunk);
}
else break;
}
@ -394,7 +355,7 @@ cpp_lex_nonalloc(Lex_Data *S_ptr,
if (S.fsm.emit_token == 0){
S.chunk_pos += size;
DrYield(3, LexNeedChunk);
DrYield(3, LexResult_NeedChunk);
}
else break;
}
@ -449,7 +410,7 @@ cpp_lex_nonalloc(Lex_Data *S_ptr,
if (S.white_done == 0){
S.chunk_pos += size;
DrYield(1, LexNeedChunk);
DrYield(1, LexResult_NeedChunk);
}
else break;
}
@ -560,7 +521,7 @@ cpp_lex_nonalloc(Lex_Data *S_ptr,
if (S.fsm.emit_token == 0){
S.chunk_pos += size;
DrYield(5, LexNeedChunk);
DrYield(5, LexResult_NeedChunk);
}
else break;
}
@ -905,13 +866,13 @@ cpp_lex_nonalloc(Lex_Data *S_ptr,
S.token.flags |= (S.pp_state != LSPP_default)?(CPP_TFLAG_PP_BODY):(0);
}
token_i = cpp_place_token_nonalloc(out_tokens, token_i, S.token);
out_tokens[token_i++] = S.token;
if (token_i == max_token_i){
if (S.pos == end_pos){
S.chunk_pos += size;
DrYield(7, LexNeedChunk);
DrYield(7, LexResult_NeedChunk);
}
DrYield(2, LexNeedTokenMemory);
DrYield(2, LexResult_NeedTokenMemory);
}
}
@ -920,109 +881,203 @@ cpp_lex_nonalloc(Lex_Data *S_ptr,
}
}
DrReturn(LexFinished);
DrReturn(LexResult_Finished);
}
#undef DrYield
#undef DrReturn
#undef DrCase
FCPP_LINK int32_t
cpp_lex_nonalloc(Lex_Data *S_ptr,
char *chunk, int32_t size,
Cpp_Token_Stack *token_stack_out, int32_t max_tokens){
Cpp_Token_Stack temp_stack = *token_stack_out;
if (temp_stack.max_count > temp_stack.count + max_tokens){
temp_stack.max_count = temp_stack.count + max_tokens;
FCPP_INTERNAL Cpp_Lex_Result
cpp_lex_nonalloc_null_end_out_limit(Cpp_Lex_Data *S_ptr, char *chunk, int32_t size,
Cpp_Token_Array *token_array_out, int32_t max_tokens_out){
Cpp_Token_Array temp_array = *token_array_out;
if (temp_array.max_count > temp_array.count + max_tokens_out){
temp_array.max_count = temp_array.count + max_tokens_out;
}
int32_t result = cpp_lex_nonalloc(S_ptr, chunk, size, &temp_stack);
Cpp_Lex_Result result = cpp_lex_nonalloc_null_end_no_limit(S_ptr, chunk, size, &temp_array);
token_stack_out->count = temp_stack.count;
if (result == LexNeedTokenMemory){
if (token_stack_out->count < token_stack_out->max_count){
result = LexHitTokenLimit;
token_array_out->count = temp_array.count;
if (result == LexResult_NeedTokenMemory){
if (token_array_out->count < token_array_out->max_count){
result = LexResult_HitTokenLimit;
}
}
return(result);
}
FCPP_LINK int32_t
cpp_lex_size_nonalloc(Lex_Data *S_ptr,
char *chunk, int32_t size, int32_t full_size,
Cpp_Token_Stack *token_stack_out){
int32_t result = 0;
FCPP_INTERNAL Cpp_Lex_Result
cpp_lex_nonalloc_no_null_no_limit(Cpp_Lex_Data *S_ptr, char *chunk, int32_t size, int32_t full_size,
Cpp_Token_Array *token_array_out){
Cpp_Lex_Result result = 0;
if (S_ptr->pos >= full_size){
char end_null = 0;
result = cpp_lex_nonalloc(S_ptr, &end_null, 1, token_stack_out);
result = cpp_lex_nonalloc_null_end_no_limit(S_ptr, &end_null, 1, token_array_out);
}
else{
result = cpp_lex_nonalloc(S_ptr, chunk, size, token_stack_out);
if (result == LexNeedChunk){
result = cpp_lex_nonalloc_null_end_no_limit(S_ptr, chunk, size, token_array_out);
if (result == LexResult_NeedChunk){
if (S_ptr->pos >= full_size){
char end_null = 0;
result = cpp_lex_nonalloc(S_ptr, &end_null, 1, token_stack_out);
result = cpp_lex_nonalloc_null_end_no_limit(S_ptr, &end_null, 1, token_array_out);
}
}
}
return(result);
}
FCPP_LINK int32_t
cpp_lex_size_nonalloc(Lex_Data *S_ptr,
char *chunk, int32_t size, int32_t full_size,
Cpp_Token_Stack *token_stack_out, int32_t max_tokens){
Cpp_Token_Stack temp_stack = *token_stack_out;
if (temp_stack.max_count > temp_stack.count + max_tokens){
temp_stack.max_count = temp_stack.count + max_tokens;
FCPP_INTERNAL Cpp_Lex_Result
cpp_lex_nonalloc_no_null_out_limit(Cpp_Lex_Data *S_ptr, char *chunk, int32_t size, int32_t full_size,
Cpp_Token_Array *token_array_out, int32_t max_tokens_out){
Cpp_Token_Array temp_stack = *token_array_out;
if (temp_stack.max_count > temp_stack.count + max_tokens_out){
temp_stack.max_count = temp_stack.count + max_tokens_out;
}
int32_t result = cpp_lex_size_nonalloc(S_ptr, chunk, size, full_size,
&temp_stack);
Cpp_Lex_Result result = cpp_lex_nonalloc_no_null_no_limit(S_ptr, chunk, size, full_size,
&temp_stack);
token_stack_out->count = temp_stack.count;
token_array_out->count = temp_stack.count;
if (result == LexNeedTokenMemory){
if (token_stack_out->count < token_stack_out->max_count){
result = LexHitTokenLimit;
if (result == LexResult_NeedTokenMemory){
if (token_array_out->count < token_array_out->max_count){
result = LexResult_HitTokenLimit;
}
}
return(result);
}
FCPP_LINK Cpp_Relex_State
cpp_relex_nonalloc_start(char *data, int32_t size, Cpp_Token_Stack *stack,
#define HAS_NULL_TERM ((int32_t)(-1))
#define NO_OUT_LIMIT ((int32_t)(-1))
FCPP_LINK Cpp_Lex_Result
cpp_lex_nonalloc(Cpp_Lex_Data *S_ptr, char *chunk, int32_t size, int32_t full_size,
Cpp_Token_Array *token_array_out, int32_t max_tokens_out)/*
DOC_PARAM(S_ptr, The lexer state. Go to the Cpp_Lex_Data section to see how to initialize the state.)
DOC_PARAM(chunk, The first or next chunk of the file being lexed.)
DOC_PARAM(size, The number of bytes in the chunk including the null terminator if the chunk ends in a null terminator.
If the chunk ends in a null terminator the system will interpret it as the end of the file.)
DOC_PARAM(full_size, If the final chunk is not null terminated this parameter should specify the length of the
file in bytes. To rely on an eventual null terminator use HAS_NULL_TERM for this parameter.)
DOC_PARAM(token_array_out, The token array structure that will receive the tokens output by the lexer.)
DOC_PARAM(max_tokens_out, The maximum number of tokens to be output to the token array. To rely on the
max built into the token array pass NO_OUT_LIMIT here.)
DOC(This call is the primary interface of the lexing system. It is quite general so it can be used in
a lot of different ways. I will explain the general rules first, and then give some examples of common
ways it might be used.
First a lexing state, Cpp_Lex_Data, must be initialized. The file to lex must be read into N contiguous chunks
of memory. An output Cpp_Token_Array must be allocated and initialized with the appropriate count and max_count values.
Then each chunk of the file must be passed to cpp_lex_nonalloc in order using the same lexing state for each call.
Every time a call to cpp_lex_nonalloc returns LexResult_NeedChunk, the next call to cpp_lex_nonalloc should use the
next chunk. If the return is some other value, the lexer hasn't finished with the current chunk and it sopped for some
other reason, so the same chunk should be used again in the next call.
If the file chunks contain a null terminator the lexer will return LexResult_Finished when it finds this character.
At this point calling the lexer again with the same state will result in an error. If you do not have a null
terminated chunk to end the file, you may instead pass the exact size in bytes of the entire file to the full_size
parameter and it will automatically handle the termination of the lexing state when it has read that many bytes.
If a full_size is specified and the system terminates for having seen that many bytes, it will return LexResult_Finished.
If a full_size is specified and a null character is read before the total number of bytes have been read the system will
still terminate as usual and return LexResult_Finished.
If the system has filled the entire output array it will return LexResult_NeedTokenMemory. When this happens if you
want to continue lexing the file you can grow the token array, or switch to a new output array and then call
cpp_lex_nonalloc again with the chunk that was being lexed and the new output. You can also specify a max_tokens_out
which is limits how many new tokens will be added to the token array. Even if token_array_out still had more space
to hold tokens, if the max_tokens_out limit is hit, the lexer will stop and return LexResult_HitTokenLimit. If this
happens there is still space left in the token array, so you can resume simply by calling cpp_lex_nonalloc again with
the same chunk and the same output array. Also note that, unlike the chunks which must only be replaced when the system
says it needs a chunk. You may switch to or modify the output array in between calls as much as you like.
The most basic use of this system is to get it all done in one big chunk and try to allocate a nearly "infinite" output
array so that it will not run out of memory. This way you can get the entire job done in one call and then just assert
to make sure it returns LexResult_Finished to you:
CODE_EXAMPLE(
Cpp_Token_Array lex_file(char *file_name){
File_Data file = read_whole_file(file_name);
Cpp_Lex_Data lex_state =
cpp_lex_data_init((char*)malloc(4096)); // hopefully big enough
Cpp_Token_Array array = {0};
array.tokens = (Cpp_Token*)malloc(1 << 20); // hopefully big enough
array.max_count = (1 << 20)/sizeof(Cpp_Token);
Cpp_Lex_Result result =
cpp_lex_nonalloc(&lex_state, file.data, file.size, file.size,
&array, NO_OUT_LIMIT);
Assert(result == LexResult_Finished);
free(lex_state.tb);
return(array);
})
)
DOC_SEE(Cpp_Lex_Data)
DOC_SEE(cpp_lex_file)
DOC_SEE(cpp_lex_nonalloc_null_end_no_limit)
DOC_SEE(cpp_lex_nonalloc_no_null_no_limit)
DOC_SEE(cpp_lex_nonalloc_null_end_out_limit)
DOC_SEE(cpp_lex_nonalloc_no_null_out_limit)
*/{
Cpp_Lex_Result result = 0;
if (full_size == HAS_NULL_TERM){
if (max_tokens_out == NO_OUT_LIMIT){
result = cpp_lex_nonalloc_null_end_no_limit(S_ptr, chunk, size, token_array_out);
}
else{
result = cpp_lex_nonalloc_null_end_out_limit(S_ptr, chunk, size, token_array_out, max_tokens_out);
}
}
else{
if (max_tokens_out == NO_OUT_LIMIT){
result = cpp_lex_nonalloc_no_null_no_limit(S_ptr, chunk, size, full_size, token_array_out);
}
else{
result = cpp_lex_nonalloc_no_null_out_limit(S_ptr, chunk, size, full_size, token_array_out, max_tokens_out);
}
}
return(result);
}
// TODO(allen): Get the relex system ready to work in chunks.
FCPP_INTERNAL Cpp_Relex_State
cpp_relex_nonalloc_start(char *data, int32_t size, Cpp_Token_Array *array,
int32_t start, int32_t end, int32_t amount, int32_t tolerance){
Cpp_Relex_State state;
state.data = data;
state.size = size;
state.stack = stack;
state.array = array;
state.start = start;
state.end = end;
state.amount = amount;
state.tolerance = tolerance;
Cpp_Get_Token_Result result = cpp_get_token(stack, start);
Cpp_Get_Token_Result result = cpp_get_token(array, start);
state.start_token_i = result.token_index-1;
if (state.start_token_i < 0){
state.start_token_i = 0;
}
result = cpp_get_token(stack, end);
result = cpp_get_token(array, end);
state.end_token_i = result.token_index;
if (end > stack->tokens[state.end_token_i].start){
if (end > array->tokens[state.end_token_i].start){
++state.end_token_i;
}
if (state.end_token_i < 0){
state.end_token_i = 0;
}
state.relex_start = stack->tokens[state.start_token_i].start;
state.relex_start = array->tokens[state.start_token_i].start;
if (start < state.relex_start){
state.relex_start = start;
}
@ -1032,18 +1087,18 @@ cpp_relex_nonalloc_start(char *data, int32_t size, Cpp_Token_Stack *stack,
return(state);
}
FCPP_LINK char
FCPP_INTERNAL char
cpp_token_get_pp_state(uint16_t bitfield){
return (char)(bitfield);
}
// TODO(allen): Eliminate this once we actually store the EOF token
// in the token stack.
FCPP_LINK Cpp_Token
cpp_index_stack(Cpp_Token_Stack *stack, int32_t file_size, int32_t index){
FCPP_INTERNAL Cpp_Token
cpp_index_array(Cpp_Token_Array *array, int32_t file_size, int32_t index){
Cpp_Token result;
if (index < stack->count){
result = stack->tokens[index];
if (index < array->count){
result = array->tokens[index];
}
else{
result.start = file_size;
@ -1055,60 +1110,68 @@ cpp_index_stack(Cpp_Token_Stack *stack, int32_t file_size, int32_t index){
return(result);
}
FCPP_LINK int32_t
FCPP_INTERNAL void
cpp_shift_token_starts(Cpp_Token_Array *array, int32_t from_token_i, int32_t shift_amount){
Cpp_Token *token = array->tokens + from_token_i;
int32_t count = array->count, i = 0;
for (i = from_token_i; i < count; ++i, ++token){
token->start += shift_amount;
}
}
FCPP_INTERNAL int32_t
cpp_relex_nonalloc_main(Cpp_Relex_State *state,
Cpp_Token_Stack *relex_stack,
Cpp_Token_Array *relex_array,
int32_t *relex_end,
char *spare){
Cpp_Token_Stack *stack = state->stack;
Cpp_Token *tokens = stack->tokens;
Cpp_Token_Array *array = state->array;
Cpp_Token *tokens = array->tokens;
cpp_shift_token_starts(stack, state->end_token_i, state->amount);
cpp_shift_token_starts(array, state->end_token_i, state->amount);
Lex_Data lex = lex_data_init(spare);
Cpp_Lex_Data lex = cpp_lex_data_init(spare);
lex.pp_state = cpp_token_get_pp_state(tokens[state->start_token_i].state_flags);
lex.pos = state->relex_start;
int32_t relex_end_i = state->end_token_i;
Cpp_Token match_token = cpp_index_stack(stack, state->size, relex_end_i);
Cpp_Token match_token = cpp_index_array(array, state->size, relex_end_i);
Cpp_Token end_token = match_token;
int32_t went_too_far = false;
// TODO(allen): This can be better I suspect.
for (;;){
int32_t result =
cpp_lex_size_nonalloc(&lex,
state->data,
state->size,
state->size,
relex_stack, 1);
cpp_lex_nonalloc_no_null_out_limit(&lex, state->data,
state->size, state->size,
relex_array, 1);
switch (result){
case LexHitTokenLimit:
case LexResult_HitTokenLimit:
{
Cpp_Token token = relex_stack->tokens[relex_stack->count-1];
Cpp_Token token = relex_array->tokens[relex_array->count-1];
if (token.start == end_token.start &&
token.size == end_token.size &&
token.flags == end_token.flags &&
token.state_flags == end_token.state_flags){
--relex_stack->count;
--relex_array->count;
goto double_break;
}
while (lex.pos > end_token.start && relex_end_i < stack->count){
while (lex.pos > end_token.start && relex_end_i < array->count){
++relex_end_i;
end_token = cpp_index_stack(stack, state->size, relex_end_i);
end_token = cpp_index_array(array, state->size, relex_end_i);
}
}
break;
case LexNeedChunk: Assert(!"Invalid path"); break;
case LexResult_NeedChunk: Assert(!"Invalid path"); break;
case LexNeedTokenMemory:
case LexResult_NeedTokenMemory:
went_too_far = true;
goto double_break;
case LexFinished:
case LexResult_Finished:
goto double_break;
}
}
@ -1118,78 +1181,117 @@ cpp_relex_nonalloc_main(Cpp_Relex_State *state,
*relex_end = relex_end_i;
}
else{
cpp_shift_token_starts(stack, state->end_token_i, -state->amount);
cpp_shift_token_starts(array, state->end_token_i, -state->amount);
}
return(went_too_far);
}
#if defined(FCPP_ALLOW_MALLOC)
#include <stdlib.h>
#include <string.h>
FCPP_LINK Cpp_Token_Stack
cpp_make_token_stack(int32_t starting_max){
Cpp_Token_Stack token_stack;
token_stack.count = 0;
token_stack.max_count = starting_max;
token_stack.tokens = (Cpp_Token*)malloc(sizeof(Cpp_Token)*starting_max);
return(token_stack);
FCPP_LINK Cpp_Token_Array
cpp_make_token_array(int32_t starting_max)/*
DOC_PARAM(starting_max, The number of tokens to initialize the array with.)
DOC_RETURN(An empty Cpp_Token_Array with memory malloc'd for storing tokens.)
DOC(This call allocates a Cpp_Token_Array with malloc for use in other
convenience functions. Stacks that are not allocated this way should not be
used in the convenience functions.
This call is a part of the FCPP_ALLOW_MALLOC convenience functions.
If you want to use it defined the macro FCPP_ALLOW_MALLOC before including 4cpp_lexer.h)
*/{
Cpp_Token_Array token_array;
token_array.tokens = (Cpp_Token*)malloc(sizeof(Cpp_Token)*starting_max);
token_array.count = 0;
token_array.max_count = starting_max;
return(token_array);
}
FCPP_LINK void
cpp_free_token_stack(Cpp_Token_Stack token_stack){
free(token_stack.tokens);
cpp_free_token_array(Cpp_Token_Array token_array)/*
DOC_PARAM(token_array, An array previously allocated by cpp_make_token_array)
DOC(This call frees a Cpp_Token_Array.
This call is a part of the FCPP_ALLOW_MALLOC convenience functions.
If you want to use it defined the macro FCPP_ALLOW_MALLOC before including 4cpp_lexer.h)
DOC_SEE(cpp_make_token_array)
*/{
free(token_array.tokens);
}
FCPP_LINK void
cpp_resize_token_stack(Cpp_Token_Stack *token_stack, int32_t new_max){
Cpp_Token *new_tokens = (Cpp_Token*)malloc(sizeof(Cpp_Token)*new_max);
if (new_tokens){
memcpy(new_tokens, token_stack->tokens, sizeof(Cpp_Token)*token_stack->count);
free(token_stack->tokens);
token_stack->tokens = new_tokens;
token_stack->max_count = new_max;
}
}
cpp_resize_token_array(Cpp_Token_Array *token_array, int32_t new_max)/*
DOC_PARAM(token_array, An array previously allocated by cpp_make_token_array.)
DOC_PARAM(new_max, The new maximum size the array should support. If this is not greater
than the current size of the array the operation is ignored.)
DOC(This call allocates a new memory chunk and moves the existing tokens in the array
over to the new chunk.
FCPP_LINK void
cpp_push_token(Cpp_Token_Stack *token_stack, Cpp_Token token){
if (!cpp_push_token_nonalloc(token_stack, token)){
int32_t new_max = 2*token_stack->max_count + 1;
cpp_resize_token_stack(token_stack, new_max);
cpp_push_token_nonalloc(token_stack, token);
This call is a part of the FCPP_ALLOW_MALLOC convenience functions.
If you want to use it defined the macro FCPP_ALLOW_MALLOC before including 4cpp_lexer.h)
DOC_SEE(cpp_make_token_array)
*/{
if (new_max > token_array->count){
Cpp_Token *new_tokens = (Cpp_Token*)malloc(sizeof(Cpp_Token)*new_max);
if (new_tokens){
memcpy(new_tokens, token_array->tokens, sizeof(Cpp_Token)*token_array->count);
free(token_array->tokens);
token_array->tokens = new_tokens;
token_array->max_count = new_max;
}
}
}
FCPP_LINK void
cpp_lex_file(char *data, int32_t size, Cpp_Token_Stack *token_stack_out){
Lex_Data S = {0};
cpp_lex_file(char *data, int32_t size, Cpp_Token_Array *token_array_out)/*
DOC_PARAM(data, The file data to be lexed in a single contiguous block.)
DOC_PARAM(size, The number of bytes in data.)
DOC_PARAM(token_array_out, The token array where the output tokens will be pushed.
This token array must be previously allocated with cpp_make_token_array)
DOC(Lexes an entire file and manages the interaction with the lexer system so that
it is quick and convenient to lex files.
This call is a part of the FCPP_ALLOW_MALLOC convenience functions.
If you want to use it defined the macro FCPP_ALLOW_MALLOC before including 4cpp_lexer.h)
DOC_SEE(cpp_make_token_array)
*/{
Cpp_Lex_Data S = {0};
S.tb = (char*)malloc(size);
int32_t quit = 0;
token_stack_out->count = 0;
token_array_out->count = 0;
for (;!quit;){
int32_t result = cpp_lex_nonalloc(&S, data, size, token_stack_out);
int32_t result = cpp_lex_nonalloc(&S, data, size, HAS_NULL_TERM, token_array_out, NO_OUT_LIMIT);
switch (result){
case LexFinished:
case LexResult_Finished:
{
quit = 1;
}break;
case LexNeedChunk:
case LexResult_NeedChunk:
{
Assert(token_array_out->count < token_array_out->max_count);
// NOTE(allen): We told the system we would provide the null
// terminator, but we didn't actually, so provide the null
// terminator via this one byte chunk.
char empty = 0;
cpp_lex_nonalloc(&S, &empty, 1, token_stack_out);
quit = 1;
cpp_lex_nonalloc(&S, &empty, 1, HAS_NULL_TERM, token_array_out, NO_OUT_LIMIT);
}break;
case LexNeedTokenMemory:
case LexResult_NeedTokenMemory:
{
int32_t new_max = 2*token_stack_out->max_count + 1;
cpp_resize_token_stack(token_stack_out, new_max);
// NOTE(allen): We told the system to use all of the output memory
// but we ran out anyway, so allocate more memory. We hereby assume
// the stack was allocated using cpp_make_token_array.
int32_t new_max = 2*token_array_out->max_count + 1;
cpp_resize_token_array(token_array_out, new_max);
}break;
}
}

97
4cpp_lexer_fsms.h
View File

@ -1,97 +0,0 @@
/*
* FSMs for 4cpp lexer
*
* 23.03.2016 (dd.mm.yyyy)
*
*/
// TOP
#if !defined(FCPP_LEXER_FSMS_H)
#define FCPP_LEXER_FSMS_H
enum Lex_State{
LS_default,
LS_identifier,
LS_pound,
LS_pp,
LS_ppdef,
LS_char,
LS_char_multiline,
LS_char_slashed,
LS_string,
LS_string_multiline,
LS_string_slashed,
LS_number,
LS_number0,
LS_float,
LS_crazy_float0,
LS_crazy_float1,
LS_hex,
LS_comment_pre,
LS_comment,
LS_comment_slashed,
LS_comment_block,
LS_comment_block_ending,
LS_dot,
LS_ellipsis,
LS_less,
LS_less_less,
LS_more,
LS_more_more,
LS_minus,
LS_arrow,
LS_and,
LS_or,
LS_plus,
LS_colon,
LS_star,
LS_modulo,
LS_caret,
LS_eq,
LS_bang,
LS_error_message,
//
LS_count
};
enum Lex_Int_State{
LSINT_default,
LSINT_u,
LSINT_l,
LSINT_L,
LSINT_ul,
LSINT_uL,
LSINT_ll,
LSINT_extra,
//
LSINT_count
};
enum Lex_PP_State{
LSPP_default,
LSPP_include,
LSPP_macro_identifier,
LSPP_identifier,
LSPP_body_if,
LSPP_body,
LSPP_number,
LSPP_error,
LSPP_junk,
//
LSPP_count
};
struct Lex_FSM{
uint8_t state;
uint8_t int_state;
uint8_t emit_token;
uint8_t multi_line;
};
static Lex_FSM null_lex_fsm = {0};
#endif
// BOTTOM

116
4cpp_lexer_types.h
View File

@ -8,6 +8,10 @@
#define ENUM(type,name) typedef type name; enum name##_
#endif
#ifndef INTERNAL_ENUM
#define INTERNAL_ENUM(type,name) typedef type name; enum name##_
#endif
/* DOC(A Cpp_Token_Type classifies a token to make parsing easier. Some types are not
actually output by the lexer, but exist because parsers will also make use of token
types in their own output.) */
@ -255,11 +259,11 @@ ENUM(uint16_t, Cpp_Preprocessor_State){
CPP_LEX_PP_COUNT
};
struct Cpp_Token_Stack{
struct Cpp_Token_Array{
Cpp_Token *tokens;
int32_t count, max_count;
};
static Cpp_Token_Stack null_cpp_token_stack = {0};
static Cpp_Token_Array null_cpp_token_array = {0};
struct Cpp_Get_Token_Result{
int32_t token_index;
@ -270,7 +274,7 @@ struct Cpp_Relex_State{
char *data;
int32_t size;
Cpp_Token_Stack *stack;
Cpp_Token_Array *array;
int32_t start, end, amount;
int32_t start_token_i;
int32_t end_token_i;
@ -279,6 +283,112 @@ struct Cpp_Relex_State{
int32_t space_request;
};
struct Cpp_Lex_FSM{
uint8_t state;
uint8_t int_state;
uint8_t emit_token;
uint8_t multi_line;
};
static Cpp_Lex_FSM null_lex_fsm = {0};
struct Cpp_Lex_Data{
char *tb;
int32_t tb_pos;
int32_t token_start;
int32_t pos;
int32_t pos_overide;
int32_t chunk_pos;
Cpp_Lex_FSM fsm;
uint8_t white_done;
uint8_t pp_state;
uint8_t completed;
Cpp_Token token;
int32_t __pc__;
};
ENUM(int32_t, Cpp_Lex_Result){
LexResult_Finished,
LexResult_NeedChunk,
LexResult_NeedTokenMemory,
LexResult_HitTokenLimit,
};
INTERNAL_ENUM(uint8_t, Cpp_Lex_State){
LS_default,
LS_identifier,
LS_pound,
LS_pp,
LS_ppdef,
LS_char,
LS_char_multiline,
LS_char_slashed,
LS_string,
LS_string_multiline,
LS_string_slashed,
LS_number,
LS_number0,
LS_float,
LS_crazy_float0,
LS_crazy_float1,
LS_hex,
LS_comment_pre,
LS_comment,
LS_comment_slashed,
LS_comment_block,
LS_comment_block_ending,
LS_dot,
LS_ellipsis,
LS_less,
LS_less_less,
LS_more,
LS_more_more,
LS_minus,
LS_arrow,
LS_and,
LS_or,
LS_plus,
LS_colon,
LS_star,
LS_modulo,
LS_caret,
LS_eq,
LS_bang,
LS_error_message,
//
LS_count
};
INTERNAL_ENUM(uint8_t, Cpp_Lex_Int_State){
LSINT_default,
LSINT_u,
LSINT_l,
LSINT_L,
LSINT_ul,
LSINT_uL,
LSINT_ll,
LSINT_extra,
//
LSINT_count
};
INTERNAL_ENUM(uint8_t, Cpp_Lex_PP_State){
LSPP_default,
LSPP_include,
LSPP_macro_identifier,
LSPP_identifier,
LSPP_body_if,
LSPP_body,
LSPP_number,
LSPP_error,
LSPP_junk,
//
LSPP_count
};
#endif
// BOTTOM

2
4ed.cpp
View File

@ -502,7 +502,7 @@ case_change_range(System_Functions *system,
}
}
if (file->state.token_stack.tokens)
if (file->state.token_array.tokens)
file_relex_parallel(system, mem, file, range.start, range.end, 0);
}
#endif

10
4ed_api_implementation.cpp
View File

@ -541,7 +541,7 @@ DOC_SEE(Access_Flag)
}
internal i32
seek_token_left(Cpp_Token_Stack *tokens, i32 pos){
seek_token_left(Cpp_Token_Array *tokens, i32 pos){
Cpp_Get_Token_Result get = cpp_get_token(tokens, pos);
if (get.token_index == -1){
get.token_index = 0;
@ -556,7 +556,7 @@ seek_token_left(Cpp_Token_Stack *tokens, i32 pos){
}
internal i32
seek_token_right(Cpp_Token_Stack *tokens, i32 pos){
seek_token_right(Cpp_Token_Array *tokens, i32 pos){
Cpp_Get_Token_Result get = cpp_get_token(tokens, pos);
if (get.in_whitespace){
++get.token_index;
@ -609,7 +609,7 @@ DOC_SEE(4coder_Buffer_Positioning_System)
if (flags & (1 << 1)){
if (file->state.tokens_complete){
pos[1] = seek_token_right(&file->state.token_stack, start_pos);
pos[1] = seek_token_right(&file->state.token_array, start_pos);
}
else{
pos[1] = buffer_seek_whitespace_right(&file->state.buffer, start_pos);
@ -640,7 +640,7 @@ DOC_SEE(4coder_Buffer_Positioning_System)
if (flags & (1 << 1)){
if (file->state.tokens_complete){
pos[1] = seek_token_left(&file->state.token_stack, start_pos);
pos[1] = seek_token_left(&file->state.token_array, start_pos);
}
else{
pos[1] = buffer_seek_whitespace_left(&file->state.buffer, start_pos);
@ -942,7 +942,7 @@ DOC_SEE(4coder_Buffer_Positioning_System)
bool32 result = false;
Editing_File *file = imp_get_file(cmd, buffer);
if (file && file->state.token_stack.tokens &&
if (file && file->state.token_array.tokens &&
file->state.tokens_complete && !file->state.still_lexing){
result = true;

4
4ed_file.cpp
View File

@ -135,8 +135,8 @@ struct Editing_File_State{
Undo_Data undo;
Cpp_Token_Stack token_stack;
Cpp_Token_Stack swap_stack;
Cpp_Token_Array token_array;
Cpp_Token_Array swap_array;
u32 lex_job;
b32 tokens_complete;
b32 still_lexing;

162
4ed_file_view.cpp
View File

@ -1083,13 +1083,13 @@ internal void
file_close(System_Functions *system, General_Memory *general, Editing_File *file){
if (file->state.still_lexing){
system->cancel_job(BACKGROUND_THREADS, file->state.lex_job);
if (file->state.swap_stack.tokens){
general_memory_free(general, file->state.swap_stack.tokens);
file->state.swap_stack.tokens = 0;
if (file->state.swap_array.tokens){
general_memory_free(general, file->state.swap_array.tokens);
file->state.swap_array.tokens = 0;
}
}
if (file->state.token_stack.tokens){
general_memory_free(general, file->state.token_stack.tokens);
if (file->state.token_array.tokens){
general_memory_free(general, file->state.token_array.tokens);
}
Buffer_Type *buffer = &file->state.buffer;
@ -1136,25 +1136,25 @@ Job_Callback_Sig(job_full_lex){
char *tb = (char*)memory->data;
Cpp_Token_Stack tokens;
Cpp_Token_Array tokens;
tokens.tokens = (Cpp_Token*)((char*)memory->data + buffer_size);
tokens.max_count = (memory->size - buffer_size) / sizeof(Cpp_Token);
tokens.count = 0;
b32 still_lexing = 1;
Lex_Data lex = lex_data_init(tb);
Cpp_Lex_Data lex = cpp_lex_data_init(tb);
do{
i32 result =
cpp_lex_size_nonalloc(&lex,
text_data, text_size, text_size,
&tokens, 2048);
cpp_lex_nonalloc(&lex,
text_data, text_size, text_size,
&tokens, 2048);
switch (result){
case LexNeedChunk: Assert(!"Invalid Path"); break;
case LexResult_NeedChunk: Assert(!"Invalid Path"); break;
case LexNeedTokenMemory:
case LexResult_NeedTokenMemory:
if (system->check_cancel(thread)){
return;
}
@ -1164,13 +1164,13 @@ Job_Callback_Sig(job_full_lex){
tokens.max_count = (memory->size - buffer_size) / sizeof(Cpp_Token);
break;
case LexHitTokenLimit:
case LexResult_HitTokenLimit:
if (system->check_cancel(thread)){
return;
}
break;
case LexFinished: still_lexing = 0; break;
case LexResult_Finished: still_lexing = 0; break;
}
} while (still_lexing);
@ -1178,27 +1178,27 @@ Job_Callback_Sig(job_full_lex){
system->acquire_lock(FRAME_LOCK);
{
Assert(file->state.swap_stack.tokens == 0);
file->state.swap_stack.tokens = (Cpp_Token*)
Assert(file->state.swap_array.tokens == 0);
file->state.swap_array.tokens = (Cpp_Token*)
general_memory_allocate(general, new_max*sizeof(Cpp_Token));
}
system->release_lock(FRAME_LOCK);
u8 *dest = (u8*)file->state.swap_stack.tokens;
u8 *dest = (u8*)file->state.swap_array.tokens;
u8 *src = (u8*)tokens.tokens;
memcpy(dest, src, tokens.count*sizeof(Cpp_Token));
system->acquire_lock(FRAME_LOCK);
{
Cpp_Token_Stack *file_stack = &file->state.token_stack;
file_stack->count = tokens.count;
file_stack->max_count = new_max;
if (file_stack->tokens){
general_memory_free(general, file_stack->tokens);
Cpp_Token_Array *file_token_array = &file->state.token_array;
file_token_array->count = tokens.count;
file_token_array->max_count = new_max;
if (file_token_array->tokens){
general_memory_free(general, file_token_array->tokens);
}
file_stack->tokens = file->state.swap_stack.tokens;
file->state.swap_stack.tokens = 0;
file_token_array->tokens = file->state.swap_array.tokens;
file->state.swap_array.tokens = 0;
}
system->release_lock(FRAME_LOCK);
@ -1216,16 +1216,16 @@ file_kill_tokens(System_Functions *system,
file->settings.tokens_exist = 0;
if (file->state.still_lexing){
system->cancel_job(BACKGROUND_THREADS, file->state.lex_job);
if (file->state.swap_stack.tokens){
general_memory_free(general, file->state.swap_stack.tokens);
file->state.swap_stack.tokens = 0;
if (file->state.swap_array.tokens){
general_memory_free(general, file->state.swap_array.tokens);
file->state.swap_array.tokens = 0;
}
}
if (file->state.token_stack.tokens){
general_memory_free(general, file->state.token_stack.tokens);
if (file->state.token_array.tokens){
general_memory_free(general, file->state.token_array.tokens);
}
file->state.tokens_complete = 0;
file->state.token_stack = null_cpp_token_stack;
file->state.token_array = null_cpp_token_array;
}
#if BUFFER_EXPERIMENT_SCALPEL <= 0
@ -1235,7 +1235,7 @@ file_first_lex_parallel(System_Functions *system,
file->settings.tokens_exist = 1;
if (file->is_loading == 0 && file->state.still_lexing == 0){
Assert(file->state.token_stack.tokens == 0);
Assert(file->state.token_array.tokens == 0);
file->state.tokens_complete = 0;
file->state.still_lexing = 1;
@ -1256,7 +1256,7 @@ file_relex_parallel(System_Functions *system,
General_Memory *general = &mem->general;
Partition *part = &mem->part;
if (file->state.token_stack.tokens == 0){
if (file->state.token_array.tokens == 0){
file_first_lex_parallel(system, general, file);
return(false);
}
@ -1267,15 +1267,15 @@ file_relex_parallel(System_Functions *system,
char *data = file->state.buffer.data;
i32 size = file->state.buffer.size;
Cpp_Token_Stack *stack = &file->state.token_stack;
Cpp_Token_Array *array = &file->state.token_array;
Cpp_Relex_State state =
cpp_relex_nonalloc_start(data, size, stack,
cpp_relex_nonalloc_start(data, size, array,
start_i, end_i, amount, 100);
Temp_Memory temp = begin_temp_memory(part);
i32 relex_end;
Cpp_Token_Stack relex_space;
Cpp_Token_Array relex_space;
relex_space.count = 0;
relex_space.max_count = state.space_request;
relex_space.tokens = push_array(part, Cpp_Token, relex_space.max_count);
@ -1289,27 +1289,27 @@ file_relex_parallel(System_Functions *system,
i32 shift_amount = relex_space.count - delete_amount;
if (shift_amount != 0){
i32 new_count = stack->count + shift_amount;
if (new_count > stack->max_count){
i32 new_count = array->count + shift_amount;
if (new_count > array->max_count){
i32 new_max = LargeRoundUp(new_count, Kbytes(1));
stack->tokens = (Cpp_Token*)
general_memory_reallocate(general, stack->tokens,
stack->count*sizeof(Cpp_Token),
array->tokens = (Cpp_Token*)
general_memory_reallocate(general, array->tokens,
array->count*sizeof(Cpp_Token),
new_max*sizeof(Cpp_Token));
stack->max_count = new_max;
array->max_count = new_max;
}
i32 shift_size = stack->count - relex_end;
i32 shift_size = array->count - relex_end;
if (shift_size > 0){
Cpp_Token *old_base = stack->tokens + relex_end;
Cpp_Token *old_base = array->tokens + relex_end;
memmove(old_base + shift_amount, old_base,
sizeof(Cpp_Token)*shift_size);
}
stack->count += shift_amount;
array->count += shift_amount;
}
memcpy(state.stack->tokens + state.start_token_i, relex_space.tokens,
memcpy(state.array->tokens + state.start_token_i, relex_space.tokens,
sizeof(Cpp_Token)*relex_space.count);
}
@ -1317,17 +1317,21 @@ file_relex_parallel(System_Functions *system,
}
if (!inline_lex){
Cpp_Token_Stack *stack = &file->state.token_stack;
Cpp_Get_Token_Result get_token_result = cpp_get_token(stack, end_i);
Cpp_Token_Array *array = &file->state.token_array;
Cpp_Get_Token_Result get_token_result = cpp_get_token(array, end_i);
i32 end_token_i = get_token_result.token_index;
if (end_token_i < 0) end_token_i = 0;
else if (end_i > stack->tokens[end_token_i].start) ++end_token_i;
if (end_token_i < 0){
end_token_i = 0;
}
else if (end_i > array->tokens[end_token_i].start){
++end_token_i;
}
cpp_shift_token_starts(stack, end_token_i, amount);
cpp_shift_token_starts(array, end_token_i, amount);
--end_token_i;
if (end_token_i >= 0){
Cpp_Token *token = stack->tokens + end_token_i;
Cpp_Token *token = array->tokens + end_token_i;
if (token->start < end_i && token->start + token->size > end_i){
token->size += amount;
}
@ -1914,9 +1918,9 @@ file_pre_edit_maintenance(System_Functions *system,
Editing_File *file){
if (file->state.still_lexing){
system->cancel_job(BACKGROUND_THREADS, file->state.lex_job);
if (file->state.swap_stack.tokens){
general_memory_free(general, file->state.swap_stack.tokens);
file->state.swap_stack.tokens = 0;
if (file->state.swap_array.tokens){
general_memory_free(general, file->state.swap_array.tokens);
file->state.swap_array.tokens = 0;
}
file->state.still_lexing = 0;
}
@ -2175,7 +2179,7 @@ file_do_batch_edit(System_Functions *system, Models *models, Editing_File *file,
case BatchEdit_PreserveTokens:
{
if (file->state.tokens_complete){
Cpp_Token_Stack tokens = file->state.token_stack;
Cpp_Token_Array tokens = file->state.token_array;
Cpp_Token *token = tokens.tokens;
Cpp_Token *end_token = tokens.tokens + tokens.count;
Cpp_Token original = {(Cpp_Token_Type)0};
@ -2531,21 +2535,21 @@ struct Make_Batch_Result{
};
internal Cpp_Token*
get_first_token_at_line(Buffer *buffer, Cpp_Token_Stack tokens, i32 line){
Cpp_Token *result = 0;
i32 start_pos = 0;
Cpp_Get_Token_Result get_token = {0};
get_first_token_at_line(Buffer *buffer, Cpp_Token_Array tokens, i32 line){
i32 start_pos = buffer->line_starts[line];
Cpp_Get_Token_Result get_token = cpp_get_token(&tokens, start_pos);
start_pos = buffer->line_starts[line];
get_token = cpp_get_token(&tokens, start_pos);
if (get_token.in_whitespace) get_token.token_index += 1;
result = tokens.tokens + get_token.token_index;
if (get_token.in_whitespace){
get_token.token_index += 1;
}
Cpp_Token *result = tokens.tokens + get_token.token_index;
return(result);
}
internal Cpp_Token*
seek_matching_token_backwards(Cpp_Token_Stack tokens, Cpp_Token *token,
seek_matching_token_backwards(Cpp_Token_Array tokens, Cpp_Token *token,
Cpp_Token_Type open_type, Cpp_Token_Type close_type){
int32_t nesting_level = 0;
if (token <= tokens.tokens){
@ -2667,7 +2671,7 @@ compute_this_indent(Buffer *buffer, Indent_Parse_State indent,
}
internal i32*
get_line_indentation_marks(Partition *part, Buffer *buffer, Cpp_Token_Stack tokens,
get_line_indentation_marks(Partition *part, Buffer *buffer, Cpp_Token_Array tokens,
i32 line_start, i32 line_end, i32 tab_width){
i32 indent_mark_count = line_end - line_start;
@ -2935,7 +2939,7 @@ file_auto_tab_tokens(System_Functions *system, Models *models,
Buffer *buffer = &file->state.buffer;
Assert(file && !file->is_dummy);
Cpp_Token_Stack tokens = file->state.token_stack;
Cpp_Token_Array tokens = file->state.token_array;
Assert(tokens.tokens);
i32 line_start = buffer_get_line_index(buffer, start);
@ -3138,7 +3142,7 @@ init_normal_file(System_Functions *system, Models *models, Editing_File *file,
String val = make_string(buffer, size);
file_create_from_string(system, models, file, val);
if (file->settings.tokens_exist && file->state.token_stack.tokens == 0){
if (file->settings.tokens_exist && file->state.token_array.tokens == 0){
file_first_lex_parallel(system, general, file);
}
@ -3156,7 +3160,7 @@ init_read_only_file(System_Functions *system, Models *models, Editing_File *file
String val = null_string;
file_create_from_string(system, models, file, val, 1);
if (file->settings.tokens_exist && file->state.token_stack.tokens == 0){
if (file->settings.tokens_exist && file->state.token_array.tokens == 0){
file_first_lex_parallel(system, general, file);
}
@ -3467,7 +3471,7 @@ update_highlighting(View *view){
}
else if (file->state.tokens_complete){
Cpp_Token_Stack *tokens = &file->state.token_stack;
Cpp_Token_Stack *tokens = &file->state.token_array;
Cpp_Get_Token_Result result = cpp_get_token(tokens, pos);
Cpp_Token token = tokens->tokens[result.token_index];
if (!result.in_whitespace){
@ -5299,10 +5303,10 @@ draw_file_loaded(View *view, i32_Rect rect, b32 is_active, Render_Target *target
Assert(view->edit_pos);
b32 tokens_use = 0;
Cpp_Token_Stack token_stack = {};
Cpp_Token_Array token_array = {};
if (file){
tokens_use = file->state.tokens_complete && (file->state.token_stack.count > 0);
token_stack = file->state.token_stack;
tokens_use = file->state.tokens_complete && (file->state.token_array.count > 0);
token_array = file->state.token_array;
}
Partition *part = &models->mem.part;
@ -5372,8 +5376,8 @@ draw_file_loaded(View *view, i32_Rect rect, b32 is_active, Render_Target *target
u32 main_color = style->main.default_color;
u32 special_color = style->main.special_character_color;
if (tokens_use){
Cpp_Get_Token_Result result = cpp_get_token(&token_stack, items->index);
main_color = *style_get_color(style, token_stack.tokens[result.token_index]);
Cpp_Get_Token_Result result = cpp_get_token(&token_array, items->index);
main_color = *style_get_color(style, token_array.tokens[result.token_index]);
token_i = result.token_index + 1;
}
@ -5388,13 +5392,13 @@ draw_file_loaded(View *view, i32_Rect rect, b32 is_active, Render_Target *target
i32 ind = item->index;
highlight_this_color = 0;
if (tokens_use && ind != prev_ind){
Cpp_Token current_token = token_stack.tokens[token_i-1];
Cpp_Token current_token = token_array.tokens[token_i-1];
if (token_i < token_stack.count){
if (ind >= token_stack.tokens[token_i].start){
if (token_i < token_array.count){
if (ind >= token_array.tokens[token_i].start){
main_color =
*style_get_color(style, token_stack.tokens[token_i]);
current_token = token_stack.tokens[token_i];
*style_get_color(style, token_array.tokens[token_i]);
current_token = token_array.tokens[token_i];
++token_i;
}
else if (ind >= current_token.start + current_token.size){

312
4ed_metagen.cpp
View File

@ -410,7 +410,7 @@ typedef struct Item_Set{
typedef struct Parse{
String code;
Cpp_Token_Stack tokens;
Cpp_Token_Array tokens;
int32_t item_count;
} Parse;
@ -440,10 +440,10 @@ get_lexeme(Cpp_Token token, char *code){
}
static Parse_Context
setup_parse_context(char *data, Cpp_Token_Stack stack){
setup_parse_context(char *data, Cpp_Token_Array array){
Parse_Context context;
context.token_s = stack.tokens;
context.token_e = stack.tokens + stack.count;
context.token_s = array.tokens;
context.token_e = array.tokens + array.count;
context.token = context.token_s;
context.data = data;
return(context);
@ -536,7 +536,7 @@ static Parse
meta_lex(char *filename){
Parse result = {0};
result.code = file_dump(filename);
result.tokens = cpp_make_token_stack(1024);
result.tokens = cpp_make_token_array(1024);
cpp_lex_file(result.code.str, result.code.size, &result.tokens);
return(result);
}
@ -584,9 +584,19 @@ typedef enum Doc_Note_Type{
DOC_PARAM,
DOC_RETURN,
DOC,
DOC_SEE
DOC_SEE,
DOC_HIDE
} Doc_Note_Type;
static String
doc_note_string[] = {
make_lit_string("DOC_PARAM"),
make_lit_string("DOC_RETURN"),
make_lit_string("DOC"),
make_lit_string("DOC_SEE"),
make_lit_string("DOC_HIDE"),
};
static int32_t
check_and_fix_docs(String *doc_string){
int32_t result = false;
@ -628,14 +638,6 @@ get_doc_string_from_prev(Parse_Context *context, String *doc_string){
return(result);
}
static String
doc_note_string[] = {
make_lit_string("DOC_PARAM"),
make_lit_string("DOC_RETURN"),
make_lit_string("DOC"),
make_lit_string("DOC_SEE"),
};
static String
doc_parse_note(String source, int32_t *pos){
String result = {0};
@ -1781,6 +1783,7 @@ print_macro_html(String *out, String name, Argument_Breakdown breakdown){
#define BACK_COLOR "#FAFAFA"
#define TEXT_COLOR "#0D0D0D"
#define CODE_BACK "#DFDFDF"
#define EXAMPLE_BACK "#EFEFDF"
#define POP_COLOR_1 "#309030"
#define POP_BACK_1 "#E0FFD0"
@ -1790,9 +1793,12 @@ print_macro_html(String *out, String name, Argument_Breakdown breakdown){
#define CODE_STYLE "font-family: \"Courier New\", Courier, monospace; text-align: left;"
#define DESCRIPT_SECTION_STYLE \
"margin-top: 3mm; margin-bottom: 3mm; font-size: .95em; " \
"background: "CODE_BACK"; padding: 0.25em;"
#define CODE_BLOCK_STYLE(back) \
"margin-top: 3mm; margin-bottom: 3mm; font-size: .95em; " \
"background: "back"; padding: 0.25em;"
#define DESCRIPT_SECTION_STYLE CODE_BLOCK_STYLE(CODE_BACK)
#define EXAMPLE_CODE_STYLE CODE_BLOCK_STYLE(EXAMPLE_BACK)
#define DOC_HEAD_OPEN "<div style='margin-top: 3mm; margin-bottom: 3mm; color: "POP_COLOR_1";'><b><i>"
#define DOC_HEAD_CLOSE "</i></b></div>"
@ -1808,6 +1814,227 @@ print_macro_html(String *out, String name, Argument_Breakdown breakdown){
#define DOC_ITEM_OPEN "<div style='margin-left: 5mm; margin-right: 5mm;'>"
#define DOC_ITEM_CLOSE "</div>"
#define EXAMPLE_CODE_OPEN "<div style='"CODE_STYLE EXAMPLE_CODE_STYLE"'>"
#define EXAMPLE_CODE_CLOSE "</div>"
static String
get_first_double_line(String source){
String line = {0};
int32_t pos0 = find_substr_s(source, 0, make_lit_string("\n\n"));
int32_t pos1 = find_substr_s(source, 0, make_lit_string("\r\n\r\n"));
if (pos1 < pos0){
pos0 = pos1;
}
line = substr(source, 0, pos0);
return(line);
}
static String
get_next_double_line(String source, String line){
String next = {0};
int32_t pos = (int32_t)(line.str - source.str) + line.size;
int32_t start = 0, pos0 = 0, pos1 = 0;
if (pos < source.size){
assert(source.str[pos] == '\n' || source.str[pos] == '\r');
start = pos + 1;
if (start < source.size){
pos0 = find_substr_s(source, start, make_lit_string("\n\n"));
pos1 = find_substr_s(source, start, make_lit_string("\r\n\r\n"));
if (pos1 < pos0){
pos0 = pos1;
}
next = substr(source, start, pos0 - start);
}
}
return(next);
}
static String
get_next_word(String source, String prev_word){
String word = {0};
int32_t pos0 = (int32_t)(prev_word.str - source.str) + prev_word.size;
int32_t pos1 = 0;
char c = 0;
for (; pos0 < source.size; ++pos0){
c = source.str[pos0];
if (!(char_is_whitespace(c) || c == '(' || c == ')')){
break;
}
}
if (pos0 < source.size){
for (pos1 = pos0; pos1 < source.size; ++pos1){
c = source.str[pos1];
if (char_is_whitespace(c) || c == '(' || c == ')'){
break;
}
}
word = substr(source, pos0, pos1 - pos0);
}
return(word);
}
static String
get_first_word(String source){
String start_str = make_string(source.str, 0);
String word = get_next_word(source, start_str);
return(word);
}
enum Doc_Chunk_Type{
DocChunk_PlainText,
DocChunk_CodeExample,
DocChunk_Count
};
static String doc_chunk_headers[] = {
make_lit_string(""),
make_lit_string("CODE_EXAMPLE"),
};
static String
get_next_doc_chunk(String source, String prev_chunk, Doc_Chunk_Type *type){
String chunk = {0};
String word = {0};
int32_t pos = source.size;
int32_t word_index = 0;
Doc_Chunk_Type t = DocChunk_PlainText;
int32_t start_pos = (int32_t)(prev_chunk.str - source.str) + prev_chunk.size;
String source_tail = substr_tail(source, start_pos);
Assert(DocChunk_Count == ArrayCount(doc_chunk_headers));
for (word = get_first_word(source_tail);
word.str;
word = get_next_word(source_tail, word), ++word_index){
for (int32_t i = 1; i < DocChunk_Count; ++i){
if (match_ss(word, doc_chunk_headers[i])){
pos = (int32_t)(word.str - source.str);
t = (Doc_Chunk_Type)i;
goto doublebreak;
}
}
}
doublebreak:;
*type = DocChunk_PlainText;
if (word_index == 0){
*type = t;
int32_t nest_level = 1;
int32_t i = find_s_char(source, pos, '(');
for (++i; i < source.size; ++i){
if (source.str[i] == '('){
++nest_level;
}
else if (source.str[i] == ')'){
--nest_level;
if (nest_level == 0){
break;
}
}
}
pos = i+1;
}
chunk = substr(source, start_pos, pos - start_pos);
int32_t is_all_white = 1;
for (int32_t i = 0; i < chunk.size; ++i){
if (!char_is_whitespace(chunk.str[i])){
is_all_white = 0;
break;
}
}
if (is_all_white){
chunk = null_string;
}
return(chunk);
}
static String
get_first_doc_chunk(String source, Doc_Chunk_Type *type){
String start_str = make_string(source.str, 0);
String chunk = get_next_doc_chunk(source, start_str, type);
return(chunk);
}
static void
print_doc_description(String *out, Partition *part, String src){
Doc_Chunk_Type type;
for (String chunk = get_first_doc_chunk(src, &type);
chunk.str;
chunk = get_next_doc_chunk(src, chunk, &type)){
switch (type){
case DocChunk_PlainText:
{
for (String line = get_first_double_line(chunk);
line.str;
line = get_next_double_line(chunk, line)){
append_ss(out, line);
append_sc(out, "<br><br>");
}
}break;
case DocChunk_CodeExample:
{
int32_t start = 0;
int32_t end = chunk.size-1;
while (start < end && chunk.str[start] != '(') ++start;
start += 1;
while (end > start && chunk.str[end] != ')') --end;
append_sc(out, EXAMPLE_CODE_OPEN);
if (start < end){
String code_example = substr(chunk, start, end - start);
int32_t first_line = 1;
for (String line = get_first_line(code_example);
line.str;
line = get_next_line(code_example, line)){
if (!(first_line && line.size == 0)){
int32_t space_i = 0;
for (; space_i < line.size; ++space_i){
if (line.str[space_i] == ' '){
append_sc(out, "&nbsp;");
}
else{
break;
}
}
String line_tail = substr_tail(line, space_i);
append_ss(out, line_tail);
append_sc(out, "<br>");
}
first_line = 0;
}
}
append_sc(out, EXAMPLE_CODE_CLOSE);
}break;
}
}
}
static void
print_struct_docs(String *out, Partition *part, Item_Node *member){
for (Item_Node *member_iter = member->first_child;
@ -1829,7 +2056,8 @@ print_struct_docs(String *out, Partition *part, Item_Node *member){
append_sc(out, DOC_ITEM_HEAD_INL_CLOSE"</div>");
append_sc(out, "<div style='margin-bottom: 6mm;'>"DOC_ITEM_OPEN);
append_ss(out, doc.main_doc);
// TODO(allen): append_ss(out, doc.main_doc);
print_doc_description(out, part, doc.main_doc);
append_sc(out, DOC_ITEM_CLOSE"</div>");
append_sc(out, "</div>");
@ -1854,30 +2082,6 @@ print_see_also(String *out, Documentation *doc){
}
}
static void
print_see_also(FILE *file, Documentation *doc){
int32_t doc_see_count = doc->see_also_count;
if (doc_see_count > 0){
fprintf(file, DOC_HEAD_OPEN"See Also"DOC_HEAD_CLOSE);
for (int32_t j = 0; j < doc_see_count; ++j){
String see_also = doc->see_also[j];
fprintf(file,
DOC_ITEM_OPEN"<a href='#%.*s_doc'>%.*s</a>"DOC_ITEM_CLOSE,
see_also.size, see_also.str,
see_also.size, see_also.str
);
}
}
}
static void
print_str(FILE *file, String str){
if (str.size > 0){
fprintf(file, "%.*s", str.size, str.str);
}
}
static void
print_function_body_code(String *out, Parse_Context *context, int32_t start){
String pstr = {0}, lexeme = {0};
@ -1968,7 +2172,8 @@ print_function_docs(String *out, Partition *part, String name, String doc_string
String main_doc = doc.main_doc;
if (main_doc.size != 0){
append_sc(out, DOC_HEAD_OPEN"Description"DOC_HEAD_CLOSE DOC_ITEM_OPEN);
append_ss(out, main_doc);
// TODO(allen): append_ss(out, main_doc);
print_doc_description(out, part, main_doc);
append_sc(out, DOC_ITEM_CLOSE);
}
@ -2076,9 +2281,13 @@ print_item(String *out, Partition *part, Used_Links *used,
append_sc(out, DOC_HEAD_OPEN"Description"DOC_HEAD_CLOSE);
append_sc(out, DOC_ITEM_OPEN);
append_ss(out, main_doc);
// TODO(allen): append_ss(out, main_doc);
print_doc_description(out, part, main_doc);
append_sc(out, DOC_ITEM_CLOSE);
}
else{
fprintf(stderr, "warning: no documentation string for %.*s\n", name.size, name.str);
}
print_see_also(out, &doc);
@ -2104,9 +2313,13 @@ print_item(String *out, Partition *part, Used_Links *used,
append_sc(out, DOC_HEAD_OPEN"Description"DOC_HEAD_CLOSE);
append_sc(out, DOC_ITEM_OPEN);
append_ss(out, main_doc);
// TODO(allen): append_ss(out, main_doc);
print_doc_description(out, part, main_doc);
append_sc(out, DOC_ITEM_CLOSE);
}
else{
fprintf(stderr, "warning: no documentation string for %.*s\n", name.size, name.str);
}
if (item->first_child){
append_sc(out, DOC_HEAD_OPEN"Values"DOC_HEAD_CLOSE);
@ -2132,7 +2345,8 @@ print_item(String *out, Partition *part, Used_Links *used,
append_sc(out, "</span></div>");
append_sc(out, "<div style='margin-bottom: 6mm;'>"DOC_ITEM_OPEN);
append_ss(out, doc.main_doc);
// TODO(allen): append_ss(out, doc.main_doc);
print_doc_description(out, part, doc.main_doc);
append_sc(out, DOC_ITEM_CLOSE"</div>");
append_sc(out, "</div>");
@ -2164,9 +2378,13 @@ print_item(String *out, Partition *part, Used_Links *used,
append_sc(out, DOC_HEAD_OPEN"Description"DOC_HEAD_CLOSE);
append_sc(out, DOC_ITEM_OPEN);
append_ss(out, main_doc);
// TODO(allen): append_ss(out, main_doc);
print_doc_description(out, part, main_doc);
append_sc(out, DOC_ITEM_CLOSE);
}
else{
fprintf(stderr, "warning: no documentation string for %.*s\n", name.size, name.str);
}
if (member->first_child){
append_sc(out, DOC_HEAD_OPEN"Fields"DOC_HEAD_CLOSE);

7
build.c
View File

@ -713,12 +713,9 @@ build_main(char *cdir, uint32_t flags){
static void
standard_build(char *cdir, uint32_t flags){
fsm_generator(cdir);
metagen(cdir);
//do_buildsuper(cdir);
//build_main(cdir, flags);
do_buildsuper(cdir);
build_main(cdir, flags);
}
#define PACK_DIR "../distributions"

17
fsm_table_generator.cpp
View File

@ -15,7 +15,6 @@
#define ArrayCount(a) (sizeof(a)/sizeof(*a))
#include "4cpp_lexer_types.h"
#include "4cpp_lexer_fsms.h"
#include "4ed_mem_ansi.c"
typedef struct Whitespace_FSM{
@ -34,8 +33,8 @@ whitespace_skip_fsm(Whitespace_FSM wfsm, char c){
return(wfsm);
}
Lex_FSM
int_fsm(Lex_FSM fsm, char c){
Cpp_Lex_FSM
int_fsm(Cpp_Lex_FSM fsm, char c){
switch (fsm.int_state){
case LSINT_default:
switch (c){
@ -98,8 +97,8 @@ int_fsm(Lex_FSM fsm, char c){
return(fsm);
}
Lex_FSM
main_fsm(Lex_FSM fsm, unsigned char pp_state, unsigned char c){
Cpp_Lex_FSM
main_fsm(Cpp_Lex_FSM fsm, uint8_t pp_state, uint8_t c){
if (c == 0){
fsm.emit_token = 1;
}
@ -610,8 +609,8 @@ generate_int_table(){
allocate_full_tables(&table, state_count);
int32_t i = 0;
Lex_FSM fsm = {0};
Lex_FSM new_fsm;
Cpp_Lex_FSM fsm = {0};
Cpp_Lex_FSM new_fsm = {0};
for (uint16_t c = 0; c < 256; ++c){
for (uint8_t state = 0; state < state_count; ++state){
fsm.int_state = state;
@ -633,8 +632,8 @@ generate_fsm_table(uint8_t pp_state){
allocate_full_tables(&table, state_count);
int32_t i = 0;
Lex_FSM fsm = {0};
Lex_FSM new_fsm;
Cpp_Lex_FSM fsm = {0};
Cpp_Lex_FSM new_fsm = {0};
for (uint16_t c = 0; c < 256; ++c){
for (uint8_t state = 0; state < state_count; ++state){
fsm.state = state;

4
win32_api_impl.cpp
View File

@ -213,7 +213,7 @@ Stream mode can be enabled with -S or -F flags on the command line to 4ed.)
API_EXPORT bool32
Is_Fullscreen(Application_Links *app)/*
DOC_SEE(This call returns true if the 4coder is in full screen mode. This call
DOC(This call returns true if the 4coder is in full screen mode. This call
takes toggles that have already occured this frame into account. So it may return
true even though the frame has not ended and actually put 4coder into full screen. If
it returns true though, 4coder will definitely be full screen by the beginning of the next
@ -229,7 +229,7 @@ frame if the state is not changed.)
API_EXPORT void
Send_Exit_Signal(Application_Links *app)/*
DOC_SEE(This call sends a signal to 4coder to attempt to exit. If there are unsaved
DOC(This call sends a signal to 4coder to attempt to exit. If there are unsaved
files this triggers a dialogue ensuring you're okay with closing.)
*/{
win32vars.send_exit_signal = 1;