How to return a Tcl dictionary in a script variable? - tcl

I have an SQL callback function after querying the database and I'm then trying to stuff that data into a TCL dictionary so I can capture it in a TCL script.
I have this callback function which gets called for each select and it adds the column as key with it's value. I pass the Tcl_Interp* as the data arg.
int callback(void *interp, int argc, char **argv, char **azColName)
{
int i = 0;
Tcl_Obj *dictPtr = Tcl_NewDictObj();
for (i = 0; i < argc; i++) {
Tcl_Obj *key = Tcl_NewStringObj(azColName[i], -1);
Tcl_Obj *value = Tcl_NewStringObj((argv[i] ? argv[i] : "null"), -1);
Tcl_DictObjPut((Tcl_Interp*) interp, dictPtr, key, value);
}
return TCL_OK;
}
The problem is that when running this command like ...
set res [mycommand getrow]
foreach i $res {
puts "i = $i"
}
I see that the callback does get called and the correct key/val's are being added, but the result that gets printed is ...
i = 0
Am I missing anything here? I'm expecting my key (column) and val's (row) to be listed.

You're missing a call to Tcl_SetObjResult, which is probably best immediately before the function returns TCL_OK.
int callback(void *interp, int argc, char **argv, char **azColName)
{
int i = 0;
Tcl_Obj *dictPtr = Tcl_NewDictObj();
for (i = 0; i < argc; i++) {
Tcl_Obj *key = Tcl_NewStringObj(azColName[i], -1);
Tcl_Obj *value = Tcl_NewStringObj((argv[i] ? argv[i] : "null"), -1);
// Should handle refcounts correctly for keys
Tcl_IncrRefCount(key);
// Don't need to pass an interp; that's just for errors
Tcl_DictObjPut(NULL, dictPtr, key, value);
Tcl_DecrRefCount(key);
}
// This is the critical line
Tcl_SetObjResult((Tcl_Interp*) interp, dictPtr);
return TCL_OK;
}
The reason why Tcl_DictObjPut doesn't need an interp here? We can (trivially) prove that it never produces an error; the dictPtr argument is always a well-formed dictionary (and unshared, but getting that wrong would cause a panic).

Related

How can we read data file based on column in C [duplicate]

I have a .csv file:
lp;imie;nazwisko;ulica;numer;kod;miejscowosc;telefon;email;data_ur
1;Jan;Kowalski;ul. Nowa;1a;11-234;Budry;123-123-456;jan#go.xxx;1980.05.13
2;Jerzy;Nowak;ul. Konopnicka;13a/3;00-900;Lichowice;(55)333-44-55;jer#wu.to;1990.03.23
And I need to read this in C. I have some code, but only for the connection.
Hopefully this would get you started
See it live on http://ideone.com/l23He (using stdin)
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
const char* getfield(char* line, int num)
{
const char* tok;
for (tok = strtok(line, ";");
tok && *tok;
tok = strtok(NULL, ";\n"))
{
if (!--num)
return tok;
}
return NULL;
}
int main()
{
FILE* stream = fopen("input", "r");
char line[1024];
while (fgets(line, 1024, stream))
{
char* tmp = strdup(line);
printf("Field 3 would be %s\n", getfield(tmp, 3));
// NOTE strtok clobbers tmp
free(tmp);
}
}
Output:
Field 3 would be nazwisko
Field 3 would be Kowalski
Field 3 would be Nowak
The following code is in plain c language and handles blank spaces.
It only allocates memory once, so one free() is needed, for each processed line.
http://ideone.com/mSCgPM
/* Tiny CSV Reader */
/* Copyright (C) 2015, Deligiannidis Konstantinos
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program. If not, see <http://w...content-available-to-author-only...u.org/licenses/>. */
#include <stdio.h>
#include <string.h>
#include <stdlib.h>
/* For more that 100 columns or lines (when delimiter = \n), minor modifications are needed. */
int getcols( const char * const line, const char * const delim, char ***out_storage )
{
const char *start_ptr, *end_ptr, *iter;
char **out;
int i; //For "for" loops in the old c style.
int tokens_found = 1, delim_size, line_size; //Calculate "line_size" indirectly, without strlen() call.
int start_idx[100], end_idx[100]; //Store the indexes of tokens. Example "Power;": loc('P')=1, loc(';')=6
//Change 100 with MAX_TOKENS or use malloc() for more than 100 tokens. Example: "b1;b2;b3;...;b200"
if ( *out_storage != NULL ) return -4; //This SHOULD be NULL: Not Already Allocated
if ( !line || !delim ) return -1; //NULL pointers Rejected Here
if ( (delim_size = strlen( delim )) == 0 ) return -2; //Delimiter not provided
start_ptr = line; //Start visiting input. We will distinguish tokens in a single pass, for good performance.
//Then we are allocating one unified memory region & doing one memory copy.
while ( ( end_ptr = strstr( start_ptr, delim ) ) ) {
start_idx[ tokens_found -1 ] = start_ptr - line; //Store the Index of current token
end_idx[ tokens_found - 1 ] = end_ptr - line; //Store Index of first character that will be replaced with
//'\0'. Example: "arg1||arg2||end" -> "arg1\0|arg2\0|end"
tokens_found++; //Accumulate the count of tokens.
start_ptr = end_ptr + delim_size; //Set pointer to the next c-string within the line
}
for ( iter = start_ptr; (*iter!='\0') ; iter++ );
start_idx[ tokens_found -1 ] = start_ptr - line; //Store the Index of current token: of last token here.
end_idx[ tokens_found -1 ] = iter - line; //and the last element that will be replaced with \0
line_size = iter - line; //Saving CPU cycles: Indirectly Count the size of *line without using strlen();
int size_ptr_region = (1 + tokens_found)*sizeof( char* ); //The size to store pointers to c-strings + 1 (*NULL).
out = (char**) malloc( size_ptr_region + ( line_size + 1 ) + 5 ); //Fit everything there...it is all memory.
//It reserves a contiguous space for both (char**) pointers AND string region. 5 Bytes for "Out of Range" tests.
*out_storage = out; //Update the char** pointer of the caller function.
//"Out of Range" TEST. Verify that the extra reserved characters will not be changed. Assign Some Values.
//char *extra_chars = (char*) out + size_ptr_region + ( line_size + 1 );
//extra_chars[0] = 1; extra_chars[1] = 2; extra_chars[2] = 3; extra_chars[3] = 4; extra_chars[4] = 5;
for ( i = 0; i < tokens_found; i++ ) //Assign adresses first part of the allocated memory pointers that point to
out[ i ] = (char*) out + size_ptr_region + start_idx[ i ]; //the second part of the memory, reserved for Data.
out[ tokens_found ] = (char*) NULL; //[ ptr1, ptr2, ... , ptrN, (char*) NULL, ... ]: We just added the (char*) NULL.
//Now assign the Data: c-strings. (\0 terminated strings):
char *str_region = (char*) out + size_ptr_region; //Region inside allocated memory which contains the String Data.
memcpy( str_region, line, line_size ); //Copy input with delimiter characters: They will be replaced with \0.
//Now we should replace: "arg1||arg2||arg3" with "arg1\0|arg2\0|arg3". Don't worry for characters after '\0'
//They are not used in standard c lbraries.
for( i = 0; i < tokens_found; i++) str_region[ end_idx[ i ] ] = '\0';
//"Out of Range" TEST. Wait until Assigned Values are Printed back.
//for ( int i=0; i < 5; i++ ) printf("c=%x ", extra_chars[i] ); printf("\n");
// *out memory should now contain (example data):
//[ ptr1, ptr2,...,ptrN, (char*) NULL, "token1\0", "token2\0",...,"tokenN\0", 5 bytes for tests ]
// |__________________________________^ ^ ^ ^
// |_______________________________________| | |
// |_____________________________________________| These 5 Bytes should be intact.
return tokens_found;
}
int main()
{
char in_line[] = "Arg1;;Th;s is not Del;m;ter;;Arg3;;;;Final";
char delim[] = ";;";
char **columns;
int i;
printf("Example1:\n");
columns = NULL; //Should be NULL to indicate that it is not assigned to allocated memory. Otherwise return -4;
int cols_found = getcols( in_line, delim, &columns);
for ( i = 0; i < cols_found; i++ ) printf("Column[ %d ] = %s\n", i, columns[ i ] ); //<- (1st way).
// (2nd way) // for ( i = 0; columns[ i ]; i++) printf("start_idx[ %d ] = %s\n", i, columns[ i ] );
free( columns ); //Release the Single Contiguous Memory Space.
columns = NULL; //Pointer = NULL to indicate it does not reserve space and that is ready for the next malloc().
printf("\n\nExample2, Nested:\n\n");
char example_file[] = "ID;Day;Month;Year;Telephone;email;Date of registration\n"
"1;Sunday;january;2009;123-124-456;jitter#go.xyz;2015-05-13\n"
"2;Monday;March;2011;(+30)333-22-55;buffer#wl.it;2009-05-23";
char **rows;
int j;
rows = NULL; //getcols() requires it to be NULL. (Avoid dangling pointers, leaks e.t.c).
getcols( example_file, "\n", &rows);
for ( i = 0; rows[ i ]; i++) {
{
printf("Line[ %d ] = %s\n", i, rows[ i ] );
char **columnX = NULL;
getcols( rows[ i ], ";", &columnX);
for ( j = 0; columnX[ j ]; j++) printf(" Col[ %d ] = %s\n", j, columnX[ j ] );
free( columnX );
}
}
free( rows );
rows = NULL;
return 0;
}
A complete example which leaves the fields as NULL-terminated strings in the original input buffer and provides access to them via an array of char pointers. The CSV processor has been confirmed to work with fields enclosed in "double quotes", ignoring any delimiter chars within them.
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
// adjust BUFFER_SIZE to suit longest line
#define BUFFER_SIZE 1024 * 1024
#define NUM_FIELDS 10
#define MAXERRS 5
#define RET_OK 0
#define RET_FAIL 1
#define FALSE 0
#define TRUE 1
// char* array will point to fields
char *pFields[NUM_FIELDS];
// field offsets into pFields array:
#define LP 0
#define IMIE 1
#define NAZWISKo 2
#define ULICA 3
#define NUMER 4
#define KOD 5
#define MIEJSCOw 6
#define TELEFON 7
#define EMAIL 8
#define DATA_UR 9
long loadFile(FILE *pFile, long *errcount);
static int loadValues(char *line, long lineno);
static char delim;
long loadFile(FILE *pFile, long *errcount){
char sInputBuf [BUFFER_SIZE];
long lineno = 0L;
if(pFile == NULL)
return RET_FAIL;
while (!feof(pFile)) {
// load line into static buffer
if(fgets(sInputBuf, BUFFER_SIZE-1, pFile)==NULL)
break;
// skip first line (headers)
if(++lineno==1)
continue;
// jump over empty lines
if(strlen(sInputBuf)==0)
continue;
// set pFields array pointers to null-terminated string fields in sInputBuf
if(loadValues(sInputBuf,lineno)==RET_FAIL){
(*errcount)++;
if(*errcount > MAXERRS)
break;
} else {
// On return pFields array pointers point to loaded fields ready for load into DB or whatever
// Fields can be accessed via pFields, e.g.
printf("lp=%s, imie=%s, data_ur=%s\n", pFields[LP], pFields[IMIE], pFields[DATA_UR]);
}
}
return lineno;
}
static int loadValues(char *line, long lineno){
if(line == NULL)
return RET_FAIL;
// chop of last char of input if it is a CR or LF (e.g.Windows file loading in Unix env.)
// can be removed if sure fgets has removed both CR and LF from end of line
if(*(line + strlen(line)-1) == '\r' || *(line + strlen(line)-1) == '\n')
*(line + strlen(line)-1) = '\0';
if(*(line + strlen(line)-1) == '\r' || *(line + strlen(line)-1 )== '\n')
*(line + strlen(line)-1) = '\0';
char *cptr = line;
int fld = 0;
int inquote = FALSE;
char ch;
pFields[fld]=cptr;
while((ch=*cptr) != '\0' && fld < NUM_FIELDS){
if(ch == '"') {
if(! inquote)
pFields[fld]=cptr+1;
else {
*cptr = '\0'; // zero out " and jump over it
}
inquote = ! inquote;
} else if(ch == delim && ! inquote){
*cptr = '\0'; // end of field, null terminate it
pFields[++fld]=cptr+1;
}
cptr++;
}
if(fld > NUM_FIELDS-1){
fprintf(stderr, "Expected field count (%d) exceeded on line %ld\n", NUM_FIELDS, lineno);
return RET_FAIL;
} else if (fld < NUM_FIELDS-1){
fprintf(stderr, "Expected field count (%d) not reached on line %ld\n", NUM_FIELDS, lineno);
return RET_FAIL;
}
return RET_OK;
}
int main(int argc, char **argv)
{
FILE *fp;
long errcount = 0L;
long lines = 0L;
if(argc!=3){
printf("Usage: %s csvfilepath delimiter\n", basename(argv[0]));
return (RET_FAIL);
}
if((delim=argv[2][0])=='\0'){
fprintf(stderr,"delimiter must be specified\n");
return (RET_FAIL);
}
fp = fopen(argv[1] , "r");
if(fp == NULL) {
fprintf(stderr,"Error opening file: %d\n",errno);
return(RET_FAIL);
}
lines=loadFile(fp,&errcount);
fclose(fp);
printf("Processed %ld lines, encountered %ld error(s)\n", lines, errcount);
if(errcount>0)
return(RET_FAIL);
return(RET_OK);
}
Use fscanf to read the file until you encounter ';' or \n, then just skip it with fscanf(f, "%*c").
int main()
{
char str[128];
int result;
FILE* f = fopen("test.txt", "r");
/*...*/
do {
result = fscanf(f, "%127[^;\n]", str);
if(result == 0)
{
result = fscanf(f, "%*c");
}
else
{
//Put here whatever you want to do with your value.
printf("%s\n", str);
}
} while(result != EOF);
return 0;
}
This code is fairly simple, but effective. It parses comma-separated files with parenthesis. You can easily modify it to suit your needs.
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
int main(int argc, char *argv[])
{
// argv[1] path to csv file
// argv[2] number of lines to skip
// argv[3] length of longest value (in characters)
FILE *pfinput;
unsigned int nSkipLines, currentLine, lenLongestValue;
char *pTempValHolder;
int c;
unsigned int vcpm; // Value character marker
int QuotationOnOff; // 0 - off, 1 - on
nSkipLines = atoi(argv[2]);
lenLongestValue = atoi(argv[3]);
pTempValHolder = (char*)malloc(lenLongestValue);
if(pfinput = fopen(argv[1], "r")) {
rewind(pfinput);
currentLine = 1;
vcpm = 0;
QuotationOnOff = 0;
// currentLine > nSkipLines condition
// skips / ignores first argv[2] lines
while((c = fgetc(pfinput)) != EOF)
{
switch(c)
{
case ',':
if(!QuotationOnOff && currentLine > nSkipLines)
{
pTempValHolder[vcpm] = '\0';
printf("%s,", pTempValHolder);
vcpm = 0;
}
break;
case '\n':
if(currentLine > nSkipLines)
{
pTempValHolder[vcpm] = '\0';
printf("%s\n", pTempValHolder);
vcpm = 0;
}
currentLine++;
break;
case '\"':
if(currentLine > nSkipLines)
{
if(!QuotationOnOff) {
QuotationOnOff = 1;
pTempValHolder[vcpm] = c;
vcpm++;
} else {
QuotationOnOff = 0;
pTempValHolder[vcpm] = c;
vcpm++;
}
}
break;
default:
if(currentLine > nSkipLines)
{
pTempValHolder[vcpm] = c;
vcpm++;
}
break;
}
}
fclose(pfinput);
free(pTempValHolder);
}
return 0;
}
#include <conio.h>
#include <stdio.h>
#include <string.h>
// Driver Code
int main()
{
// Substitute the full file path
// for the string file_path
FILE* fp = fopen("Movie.csv", "r");
char *wrds[40];
if (!fp)
printf("Can't open file\n");
else {
// Here we have taken size of
// array 1024 you can modify it
char buffer[1024];
int row = 0;
int column = 0;
while (fgets(buffer, 1024, fp)) {
column = 0;
row++;
// To avoid printing of column
// names in file can be changed
// according to need
if (row == 1)
continue;
// Splitting the data
char* value = strtok(buffer, ", ");
while (value) {
// Column 1
if (column == 0) {
printf("Name :");
}
// Column 2
if (column == 1) {
printf("\tAccount No. :");
}
// Column 3
if (column == 2) {
printf("\tAmount :");
}
printf("%s", value);
wrds[column] = value;
value = strtok(NULL, ", ");
column++;
}
printf("\n");
}
// Close the file
fclose(fp);
}
getchar();
return 0;
}

trying to deserialize JSON unsuccesfully

I am trying to deserialize an array of chars using the ArduinoJson Lib. The chars are coming from a I2C connection using Wire library. They come successfully so I don't think there is an issue there.
When trying to de-serialize at run time I get the message "deserializeJson() failed: EmptyInput".
From ArduinoJson Troubleshooter's report
The issue happens at run time
The issue concerns deserialization
deserializeJson() returns EmptyInput
Input comes neither from an HTTP response, nor a file, nor a stream
The part of code that deserialization fails is here:
void callback(char payload[], unsigned int msg_size) {
Serial.print("Message arrived ");
for (int i=0;i<msg_size;i++) {
Serial.print((payload[i]));
}
DeserializationError error = deserializeJson(doc, payload, msg_size);
if (error) {
Serial.print(F("deserializeJson() failed: "));
Serial.println(error.f_str());
return;
}
const char* command = doc["cmd"];
Serial.print("command ");
Serial.println(command);
//rest of code
}
void setup() {
//communications
Wire.begin(0x8); // join i2c bus with address #8
Wire.onReceive(receiveEvent); // register event
Serial.begin(9600); // start serial for output
Serial.print("Start_");
Serial.println(ver);
}
void loop() {
// function that executes whenever data is received from master
// this function is registered as an event, see setup()
void receiveEvent(int howMany) {
char buff[60];
Serial.print("how many=");
Serial.println(howMany);
//initialize buffers
buff[0] = '\0';
int i = 0;
while (0 < Wire.available()) { // loop through all but the last
//char c = Wire.read(); // receive byte as a character
char c = Wire.read();
Serial.print(c); // print the character
buff[i] = c;
buff[i+1] = '\0'; // Add a NULL after each character
i++;
}
int x = Wire.read(); // receive byte as an integer
Serial.println();
Serial.print("x:");
Serial.println(x); // print the integer
Serial.print("BUFFER:");
for (int j=0; j < howMany; j++) {
Serial.print(buff[j]);
}
Serial.println();
Serial.print("size:");
Serial.println(sizeof(buff));
callback(buff, howMany);
}
The print out from serial ide is here:
Start_v.0.1
how many=25
{"cmd":"demo","seq":"1"}
x:-1
BUFFER:{"cmd":"demo","seq":"1"}
size:60
Message arrived {"cmd":"demo","seq":"1"} deserializeJson() failed: EmptyInput
But the payload variable, as you see here is not empty.
Can you provide any debugging hints? Thank you
Edit: I have added more code that shows the i2c communication
from smbus import SMBus
addr = 0x8 # bus address
bus = SMBus(1) # indicates /dev/ic2-1
numb = 1
def StringToBytes(val):
retVal = []
for c in val:
retVal.append(ord(c))
return retVal
def writeData(arrayValue, i2c_address):
byteValue = StringToBytes(arrayValue)
# examples --->
bus.write_i2c_block_data(i2c_address,0x00,byteValue)
return -1
# main
print ("Enter command:")
while 1:
cmd = input(">>>> ")
writeData(str(cmd), addr)
I filtered the incoming i2c buffer for nulls and the de serialization of JSON worked. I am not sure why the nulls are sent by rpi/python.
void receiveEvent(int howMany) {
char buff[60];
Serial.print("how many=");
Serial.println(howMany);
//initialize buffers
buff[0] = '\0';
int i = 0;
while (0 < Wire.available()) { // loop through all but the last
char c = Wire.read(); // receive byte as a character
if (c == '\0') {
//ignore
} else {
//char c = Wire.read();
Serial.print(c); // print the character
buff[i] = c;
buff[i+1] = '\0'; // Add a NULL after each character
i++;
}
}
int x = Wire.read(); // receive byte as an integer
Serial.println();
Serial.print("x:");
Serial.println(x); // print the integer
Serial.print("BUFFER:");
for (int j=0; j < howMany; j++) {
Serial.print(buff[j]);
}
Serial.println();
Serial.print("size:");
Serial.println(sizeof(buff));
callback(buff, howMany);
}

C - alternative using snprintf to prepare MySQL statements?

I've been tearing my hair out for a while on this one. The C code is called from a bash script, which loops through a command's output in a while loop and passes variables to the C script as args. It goes through a list and partitions data properly. I've been using the C MySQL api, and up until now everything has been relatively straight forward. It tries to run a SELECT(EXISTS) command to dictate whether to input a new row, or update an existing one.
I have typed the command into MySQL terminal and it works perfectly. I have even printf'd it and copied the command directly into the terminal. It works....
So why then, am I getting Syntax errors? I've tried escaping fields and input using backticks, single quotes and double quotes and I'm still getting this dumbounding error. I thought maybe it was something to do with the null space? But I'm at my witts end. Here's the code, any advice would be greatly appreciated :)
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include <mysql/mysql.h>
const int MAXLEN = 100;
/* Compile with:
gcc db.connect.c `mysql_config --libs` -O1
for the best results
*/
/* Function definitions for later */
void finish_with_error(MYSQL *con);
int send_query(MYSQL *con, char query[MAXLEN]);
/* If any SQL commands fail, return an error message */
void finish_with_error(MYSQL *con)
{
fprintf(stderr, "%s\n", mysql_error(con));
mysql_close(con);
exit(1);
}
/* Helper function to send queries to MySQL database */
int send_query(MYSQL *con, char query[MAXLEN])
{
if (mysql_query(con, query)) {
finish_with_error(con);
}
return 0;
}
int main(int argc, char ** argv)
{
// Establish MySQL API connection, if not- fail with err
MYSQL *con = mysql_init(NULL);
if (con == NULL) {
finish_with_error(con);
}
// Connection string.
if (mysql_real_connect(con, "localhost", "user", "password",
NULL, 0, NULL, 0) == NULL){
finish_with_error(con);
}
if (argv[1] == NULL){
printf("No query passed, terminating script \n");
return 1;
}
if (argv[1] != NULL) {
if( strcmp( argv[1], "--help" ) == 0 ) {
printf("This program was created to interact with MySQL, by checking and updating live network stats\n");
printf("It has 2 parameters, an IP address to look in the database for and a value to update a field by, \
if that IP address is found. ");
printf("If the value is not found, the program will insert a new row.");
return 1;
}
// Works out how much memory to allocate to buffer for snprintf
// Originally cmd_len was 65- as this was the amount of bits needed by the address string.
// This was changed to MAXLEN to prevent SEGFAULTS and give the function breathing room.
size_t cmd_len = MAXLEN;
size_t param_len = sizeof(argv[2]);
size_t q_len = cmd_len + param_len;
// Allocates that memory to a buffer, referenced as query
char *query = malloc(sizeof(char) * q_len);
snprintf(query, q_len, "SELECT EXISTS(SELECT * FROM `analytics`.`live` WHERE `foreign_addr` = `%s`)", argv[1]);
printf("%s\n", query);
send_query(con, query);
free(query);
// Used to store the result of the MySQL select commands
MYSQL_RES *result = mysql_store_result(con);
if (result == NULL) {
finish_with_error(con);
}
// num_fields stores the number of fields, i and x are counters, answer is 1 or 0
int num_fields = mysql_num_fields(result);
int i = 0;
// Loops through each row in the answer statement.
// There will only be one row in the answer, which will be 1 or 0
// Basically, if the IP is found.
MYSQL_ROW row;
while ((row = mysql_fetch_row(result))){
for (i=0; i<num_fields; i++) {
// If the IP isn't in the table
if(!atoi(row[i]))
send_query(con, argv[1]);
// If the IP is already in the table
if(atoi(row[i])) {
snprintf(query, q_len, "UPDATE analytics.live SET count=count+1 WHERE foreign_addr = '%s'", argv[1]);
printf("%s\n", query);
free(query);
snprintf(query, q_len, "UPDATE analytics.live SET dat_sent = dat_sent + %s", argv[2]);
printf("%s\n", query);
free(query);
}
}
}
mysql_close(con);
return 1;
}
mysql_close(con);
return 0;
}

C mysql_num_rows() Segmentation fault

I'm trying to execute MySQL query in C, however I get a Segmentation fault while calling mysql_num_rows().
Here's the code I'm using:
char *username = "test#mail.com";
char *password = "pass";
char query[1000];
int len;
char *q = "SELECT * FROM Users WHERE `Email` = '%s' AND `Password` = '%s'";
len = snprintf(query, strlen(q) + strlen(username) + strlen(password), q, username, password);
MYSQL_RES *result;
if (db_query(query, result))
{
if (result != NULL)
{
int test_count = mysql_num_rows(result);
printf("%d\n", test_count);
}
}
else
{
printf("Query error\n");
}
And here is the db_query() function:
bool db_query(const char *query, MYSQL_RES *result)
{
if (mysql_query(db_connection, query))
{
printf("mysql_query(): Error %u: %s\n", mysql_errno(db_connection), mysql_error(db_connection));
return false;
}
if (!(result = mysql_store_result(db_connection)))
{
printf("mysql_store_result(): Error %u: %s\n", mysql_errno(db_connection), mysql_error(db_connection));
return false;
}
return true;
}
I've tested the query and the problem isn't there, the connection is initiated too. Any ideas?
Thanks!
Your problem is here, in the db_query function:
if (!(result = mysql_store_result(db_connection)))
The assignment to result has no visible effect in the function's caller - you're passing a pointer by value, changing the value of result in the callee doesn't do anything to result in the caller.
You need to change your function to take a pointer-to-pointer, and adapt the call site and the db_query function.
bool db_query(const char *query, MYSQL_RES **result)
{
...
if (!(*result = mysql_store_result(db_connection)))
...
}
Any changes to result in your db_query function are not reflected back to the caller, hence it will still contain the arbitrary value it had when it was created (as an auto variable with no initialisation.
If you want to change the value and have it reflected back, you should pass a double pointer to it then dereference the double pointer to get at the actual value.
Even better would be to return the result value and use its NULL/non-NULL status for a success code rather than returning true/false.

How do I work with complex numbers in CUSPARSE?

I'm currently working with CUSPARSE. I'm having trouble because I don't know how to print a complex number. For example, when I write:
cuComplex a;
a.x=1.2;
a.y=2.2;
How do I print the varable a?
I've tried :
cout<< a;
but it doesn't work.
You will need to overload the << operator to take in cuComplex and cuDoubleComplex data types.
std::ostream& operator<<(std::ostream& strm, const cuComplex& in)
{
char sgn[2] = "+-"
strm << in.x << sgn[in.y < 0] << " i"<< std::abs(in.y);
return strm;
}
You can do the same for cuDoubleComplex
The data in std::complex is identical to the corresponding data in a cuComplex, i.e. you can reinterpret_cast pointers (and therefore arrays, too) of one type to the other – it works in practise and is, I think, actually guaranteed by C++11, you can test it like this:
namespace check_stdComplexdouble_to_cuDoubleComplex_binary_compatibility{
using std::complex;
const complex<double> testarr[] = { complex<double>(0.,.5)
, complex<double>(1.,1.5) };
const cuDoubleComplex* cucomplexd
= reinterpret_cast<const cuDoubleComplex*>(testarr);
auto tester() -> bool {
assert( cuCreal(cucomplexd[0])==0. && cuCimag(cucomplexd[0])==.5
&& cuCreal(cucomplexd[1])==1. && cuCimag(cucomplexd[1])==1.5 );
return true;
}
const bool ok = tester();
bool good(){return ok;}
};
If you call a CUBLAS function that's supposed to read/write from/to an std::complex<float>, you can just give it a reinterpret_casted pointer, e.g.
std::complex<double> result;
xhandle->cublasstat = yhandle->cublasstat
= cublasZdotc( *xhandle->cublashandle
, xhandle->vect_dimension
, xhandle->vector, 1
, yhandle->vector, 1
, reinterpret_cast<cuDoubleComplex*>(&result) );