290 lines
10 KiB
C
290 lines
10 KiB
C
|
#include "arc/std/parser/csv.h"
|
||
|
|
#include "arc/std/parser/helpers.h"
|
||
|
|
#include "arc/std/bool.h"
|
||
|
|
#include "arc/std/parser.h"
|
||
|
|
#include <stddef.h>
|
||
|
|
#include <stdint.h>
|
||
|
|
#include <stdlib.h>
|
||
|
|
|
||
|
|
typedef struct ARC_ParserCSVUserData {
|
||
|
|
ARC_Bool header;
|
||
|
|
ARC_ParserCSV_CastTypeFn castTypeFn;
|
||
|
|
ARC_ParserCSV_DestroyTypeFn destroyTypeFn;
|
||
|
|
} ARC_ParserCSVUserData;
|
||
|
|
|
||
|
|
void ARC_ParserCSV_InitLexerRulesFn(ARC_Lexer *lexer){
|
||
|
|
|
||
|
|
}
|
||
|
|
|
||
|
|
uint32_t ARC_ParserCSV_GetStringIdFn(ARC_String *string){
|
||
|
|
if(ARC_String_EqualsCStringWithStrlen(string, "LAMBDA")){
|
||
|
|
return ARC_PARSER_TAG_LAMBDA;
|
||
|
|
}
|
||
|
|
|
||
|
|
return ~(uint32_t)0;
|
||
|
|
}
|
||
|
|
|
||
|
|
void ARC_ParserCSVData_VectorDestroyVectorFn(void *data){
|
||
|
|
ARC_Vector *vector = (ARC_Vector *)data;
|
||
|
|
ARC_Vector_Destroy(vector);
|
||
|
|
}
|
||
|
|
|
||
|
|
void ARC_ParserCSVData_GetDataTag(ARC_Vector *dataVector, ARC_ParserTagToken *tagToken, ARC_ParserCSVUserData *userData){
|
||
|
|
//cast the csv data back to its original type
|
||
|
|
ARC_ParserCSVUserData *csvUserData = (ARC_ParserCSVUserData *)userData;
|
||
|
|
|
||
|
|
//loop through the tags either recursing to next body or adding data to vector
|
||
|
|
for(uint32_t index = 0; index < ARC_Vector_GetSize(tagToken->tagTokens); index++){
|
||
|
|
ARC_ParserTagToken *childTagToken = (ARC_ParserTagToken *)ARC_Vector_Get(tagToken->tagTokens, index);
|
||
|
|
|
||
|
|
ARC_String *stringData = NULL;
|
||
|
|
ARC_Vector *row = NULL;
|
||
|
|
uint32_t rowIndex = 0;
|
||
|
|
|
||
|
|
switch(childTagToken->id){
|
||
|
|
//recuse to run the next line
|
||
|
|
case ARC_PARSER_CSV_STRING:
|
||
|
|
//get the string of the data
|
||
|
|
ARC_String_Create(&stringData, NULL, 0);
|
||
|
|
ARC_ParserData_HelperRecurseStringAdd(&stringData, childTagToken);
|
||
|
|
|
||
|
|
//move data string and cleanup
|
||
|
|
void *data = (void *)stringData;
|
||
|
|
|
||
|
|
//get the last row vector
|
||
|
|
rowIndex = ARC_Vector_GetSize(dataVector) - 1;
|
||
|
|
row = (ARC_Vector *)ARC_Vector_Get(dataVector, rowIndex);
|
||
|
|
|
||
|
|
//only call the userData cast type callback if it is not a header
|
||
|
|
if(rowIndex != 0 || csvUserData->header == ARC_False){
|
||
|
|
data = NULL;
|
||
|
|
userData->castTypeFn(&data, stringData);
|
||
|
|
ARC_String_Destroy(stringData);
|
||
|
|
}
|
||
|
|
|
||
|
|
//add the data to the row vector
|
||
|
|
ARC_Vector_Add(row, data);
|
||
|
|
continue;
|
||
|
|
|
||
|
|
case ARC_PARSER_CSV_DATA:
|
||
|
|
ARC_ParserCSVData_GetDataTag(dataVector, childTagToken, userData);
|
||
|
|
continue;
|
||
|
|
|
||
|
|
default:
|
||
|
|
//this should only be the case for the comma
|
||
|
|
continue;
|
||
|
|
}
|
||
|
|
}
|
||
|
|
}
|
||
|
|
|
||
|
|
//<line> -> <data> NEWLINE <line> | <data> | NEWLINE <line> | LAMBDA
|
||
|
|
void ARC_ParserCSVData_RunLineTag(ARC_Vector *dataVector, ARC_ParserTagToken *tagToken, ARC_ParserCSVUserData *userData){
|
||
|
|
//loop through the tags either going to the next line or the next body
|
||
|
|
for(uint32_t index = 0; index < ARC_Vector_GetSize(tagToken->tagTokens); index++){
|
||
|
|
ARC_ParserTagToken *childTagToken = (ARC_ParserTagToken *)ARC_Vector_Get(tagToken->tagTokens, index);
|
||
|
|
|
||
|
|
ARC_Vector *row = NULL;
|
||
|
|
uint32_t rowIndex = 0;
|
||
|
|
|
||
|
|
switch(childTagToken->id){
|
||
|
|
//recuse to run the next line
|
||
|
|
case ARC_PARSER_CSV_LINE:
|
||
|
|
ARC_ParserCSVData_RunLineTag(dataVector, childTagToken, userData);
|
||
|
|
continue;
|
||
|
|
|
||
|
|
//get the row data
|
||
|
|
case ARC_PARSER_CSV_DATA:
|
||
|
|
ARC_ParserCSVData_GetDataTag(dataVector, childTagToken, userData);
|
||
|
|
continue;
|
||
|
|
|
||
|
|
//add a new row for each new line
|
||
|
|
case ARC_PARSER_CSV_CHAR_NEWLINE:
|
||
|
|
//check if current is not empty (no need to create another empty row)
|
||
|
|
rowIndex = ARC_Vector_GetSize(dataVector) - 1;
|
||
|
|
row = (ARC_Vector *)ARC_Vector_Get(dataVector, rowIndex);
|
||
|
|
if(ARC_Vector_GetSize(row) == 0){
|
||
|
|
continue;
|
||
|
|
}
|
||
|
|
|
||
|
|
//create and add a new row
|
||
|
|
ARC_Vector_Create(&row, NULL, NULL);
|
||
|
|
ARC_Vector_Add(dataVector, (void *)row);
|
||
|
|
continue;
|
||
|
|
|
||
|
|
default:
|
||
|
|
continue;
|
||
|
|
}
|
||
|
|
}
|
||
|
|
}
|
||
|
|
|
||
|
|
void ARC_ParserCSVData_CreateFn(void **data, ARC_ParserTagToken *parsedData, void *userData){
|
||
|
|
*data = NULL;
|
||
|
|
if(data == NULL || userData == NULL){
|
||
|
|
//TODO: error here?
|
||
|
|
*data = NULL;
|
||
|
|
return;
|
||
|
|
}
|
||
|
|
|
||
|
|
//cast the csv data back to its original type
|
||
|
|
ARC_ParserCSVUserData *csvUserData = (ARC_ParserCSVUserData *)userData;
|
||
|
|
|
||
|
|
//crate a vector for the data to be stored in before being copied to a 2d array
|
||
|
|
ARC_Vector *dataVector;
|
||
|
|
ARC_Vector_DestroyDataFn destroyVectorFn = ARC_ParserCSVData_VectorDestroyVectorFn;
|
||
|
|
ARC_Vector_Create(&dataVector, NULL, &destroyVectorFn);
|
||
|
|
|
||
|
|
//create the first row vector, does not need a destroy function as its contents will be moved
|
||
|
|
ARC_Vector *dataRowVector;
|
||
|
|
ARC_Vector_Create(&dataRowVector, NULL, NULL);
|
||
|
|
|
||
|
|
//add the row to the dataVector
|
||
|
|
ARC_Vector_Add(dataVector, (void *)dataRowVector);
|
||
|
|
|
||
|
|
//recursively add data from the parsedData to the data vector
|
||
|
|
ARC_ParserCSVData_RunLineTag(dataVector, parsedData, csvUserData);
|
||
|
|
|
||
|
|
//get the first line to check if it has any values
|
||
|
|
dataRowVector = ARC_Vector_Get(dataVector, 0);
|
||
|
|
if(ARC_Vector_GetSize(dataRowVector) == 0){
|
||
|
|
//TODO: iterate and clear the vector
|
||
|
|
ARC_Vector_Destroy(dataVector);
|
||
|
|
return;
|
||
|
|
}
|
||
|
|
|
||
|
|
//create the data that will be saved
|
||
|
|
ARC_ParserCSVData *csvData = (ARC_ParserCSVData *)malloc(sizeof(ARC_ParserCSVData));
|
||
|
|
csvData->hasHeader = csvUserData->header;
|
||
|
|
csvData->headers = NULL;
|
||
|
|
csvData->height = 0;
|
||
|
|
csvData->width = ARC_Vector_GetSize(dataRowVector);
|
||
|
|
csvData->data = NULL;
|
||
|
|
|
||
|
|
//create the starting index (to offset the header if it exists)
|
||
|
|
uint32_t heightStartIndex = 0;
|
||
|
|
|
||
|
|
//store the headers if they exist
|
||
|
|
if(csvData->hasHeader == ARC_True){
|
||
|
|
heightStartIndex++;
|
||
|
|
|
||
|
|
//copy the headers
|
||
|
|
csvData->headers = (ARC_String **)malloc(sizeof(ARC_String *) * csvData->width);
|
||
|
|
for(uint32_t headerIndex = 0; headerIndex < csvData->width; headerIndex++){
|
||
|
|
csvData->headers[headerIndex] = (ARC_String *)ARC_Vector_Get(dataRowVector, headerIndex);
|
||
|
|
}
|
||
|
|
}
|
||
|
|
|
||
|
|
//check that all the rows are the same size
|
||
|
|
for(uint32_t rowIndex = heightStartIndex; rowIndex < ARC_Vector_GetSize(dataVector); rowIndex++){
|
||
|
|
ARC_Vector *currentRowVector = (ARC_Vector *)ARC_Vector_Get(dataVector, rowIndex);
|
||
|
|
|
||
|
|
//skip an empty line (this should only be the case for the last line)
|
||
|
|
uint32_t currentRowVectorSize = ARC_Vector_GetSize(currentRowVector);
|
||
|
|
if(currentRowVectorSize == 0){
|
||
|
|
continue;
|
||
|
|
}
|
||
|
|
|
||
|
|
//a row was found so update the height
|
||
|
|
csvData->height++;
|
||
|
|
|
||
|
|
//TODO: probs want to error
|
||
|
|
//cleanup and exit if they don't match
|
||
|
|
if(csvData->width != currentRowVectorSize){
|
||
|
|
//TODO: iterate and clear the vector
|
||
|
|
ARC_Vector_Destroy(dataVector);
|
||
|
|
return;
|
||
|
|
}
|
||
|
|
}
|
||
|
|
|
||
|
|
//init location to copy data to
|
||
|
|
csvData->data = (void ***)malloc(sizeof(void **) * csvData->height);
|
||
|
|
|
||
|
|
//copy the data
|
||
|
|
uint32_t rowIndex = 0;
|
||
|
|
for(uint32_t y = 0; y < csvData->height; y++, rowIndex++){
|
||
|
|
ARC_Vector *currentRowVector = (ARC_Vector *)ARC_Vector_Get(dataVector, rowIndex + heightStartIndex);
|
||
|
|
|
||
|
|
//skip an empty line (should only be the case for the last line)
|
||
|
|
uint32_t currentRowVectorSize = ARC_Vector_GetSize(currentRowVector);
|
||
|
|
if(currentRowVectorSize == 0){
|
||
|
|
y--;
|
||
|
|
continue;
|
||
|
|
}
|
||
|
|
|
||
|
|
//create the current row
|
||
|
|
csvData->data[y] = (void **)malloc(sizeof(void *) * csvData->width);
|
||
|
|
|
||
|
|
for(uint32_t x = 0; x < csvData->width; x++){
|
||
|
|
csvData->data[y][x] = ARC_Vector_Get(currentRowVector, x);
|
||
|
|
}
|
||
|
|
}
|
||
|
|
|
||
|
|
ARC_Vector_Destroy(dataVector);
|
||
|
|
|
||
|
|
*data = (void *)csvData;
|
||
|
|
}
|
||
|
|
|
||
|
|
void ARC_ParserCSVData_DestroyFn(void *data, ARC_Bool clear, void *userData){
|
||
|
|
if(userData == NULL){
|
||
|
|
return;
|
||
|
|
}
|
||
|
|
|
||
|
|
ARC_ParserCSVUserData *csvUserData = (ARC_ParserCSVUserData *)userData;
|
||
|
|
|
||
|
|
if(data != NULL){
|
||
|
|
ARC_ParserCSVData *csvData = (ARC_ParserCSVData *)data;
|
||
|
|
|
||
|
|
//cleanup the headers if they exist
|
||
|
|
if(csvData->hasHeader == ARC_True){
|
||
|
|
for(uint32_t x = 0; x < csvData->width; x++){
|
||
|
|
ARC_String *string = csvData->headers[x];
|
||
|
|
ARC_String_Destroy(string);
|
||
|
|
}
|
||
|
|
free(csvData->headers);
|
||
|
|
}
|
||
|
|
|
||
|
|
for(uint32_t y = 0; y < csvData->height; y++){
|
||
|
|
//cleanup each element in the data
|
||
|
|
for(uint32_t x = 0; x < csvData->width; x++){
|
||
|
|
csvUserData->destroyTypeFn(csvData->data[y][x]);
|
||
|
|
}
|
||
|
|
|
||
|
|
//cleanup each row of data
|
||
|
|
free(csvData->data[y]);
|
||
|
|
}
|
||
|
|
|
||
|
|
//cleanup the data
|
||
|
|
free(csvData->data);
|
||
|
|
free(csvData);
|
||
|
|
}
|
||
|
|
|
||
|
|
if(clear == ARC_False){
|
||
|
|
free(csvUserData);
|
||
|
|
}
|
||
|
|
}
|
||
|
|
|
||
|
|
void ARC_ParserCSV_CreateAsParser(ARC_Parser **parser, ARC_Bool header, ARC_ParserCSV_CastTypeFn castTypeFn, ARC_ParserCSV_DestroyTypeFn destroyTypeFn){
|
||
|
|
/* ~ define the language as a string ~ */
|
||
|
|
char *languageCString =
|
||
|
|
"<line> -> <data> NEWLINE <line> | <data> | NEWLINE <line> | LAMBDA\n"
|
||
|
|
"<data> -> <string> COMMA <data> | <string>\n"
|
||
|
|
"<string> -> <nonCommaChar> <string> | <nonCommaChar>\n"
|
||
|
|
"<nonCommaChar> -> CHAR_BEFORE_COMMA | CHAR_AFTER_COMMA\n";
|
||
|
|
|
||
|
|
ARC_String *languageString;
|
||
|
|
ARC_String_CreateWithStrlen(&languageString, languageCString);
|
||
|
|
|
||
|
|
/* ~ init the userdata ~ */
|
||
|
|
ARC_ParserCSVUserData *userdata = (ARC_ParserCSVUserData *)malloc(sizeof(ARC_ParserCSVUserData));
|
||
|
|
userdata->header = header;
|
||
|
|
userdata->castTypeFn = castTypeFn;
|
||
|
|
userdata->destroyTypeFn = destroyTypeFn;
|
||
|
|
|
||
|
|
/* ~ create the language ~ */
|
||
|
|
ARC_ParserData_CreateFn createCharFn = ARC_ParserCSVData_CreateFn;
|
||
|
|
ARC_ParserData_DestroyFn destroyCharFn = ARC_ParserCSVData_DestroyFn;
|
||
|
|
ARC_Parser_CreateFromString(parser, languageString, ARC_ParserCSV_InitLexerRulesFn, ARC_ParserCSV_GetStringIdFn, &createCharFn, &destroyCharFn, userdata);
|
||
|
|
|
||
|
|
//cleanup
|
||
|
|
ARC_String_Destroy(languageString);
|
||
|
|
}
|