I'm querying a mysql table which then loops through the results.
One of the fields has a value of "0" in it, so when I try the following it doesn't work!
while ((row2 = mysql_fetch_row(resultset2)) != NULL) {
if (row2[2] != "0") {
// the field has a value of 0, but it's still executing the code here!
} else {
// should be executing this code
}
}
I know C/C++ is very strict when it comes variables (unlink php), but I can't figure this one out. Anyone have any ideas why?
You're comparing row2[2], a pointer to char, with a pointer to the constant char array "0".
Use strcmp(row2[2], "0") != 0 (C solution), std::string(row2[2]) != "0" (C++ solution), or atoi(row2[2]) != 0 if you know row2[2] is always the string representation of an integer (and cannot be a SQL NULL value).
You cannot compare string literal like this :
if (row2[2] != "0") //wrong
Either you do this :
if (strcmp(row2[2], "0")) //correct
Or this:
if (std::string(row2[2]) != "0") //correct
For this particular case, when there is only one character you can also do this:
if (row2[2][0] != '0') //correct - not the single quote around 0!
Related
I want to do two string compare and used two different if condition. Is there any better way to do string compare in one if condition
if (strcmp(Buff1(), Config1) == 0)
{
if (strcmp(Buff2, Config2) == 0)
{
// my code goes here
}
}
The equivalent code is:
if ((strcmp(Buff1(), Config1) == 0)) &&
(strcmp(Buff2, Config2) == 0))
{
// my code goes here
}
Note: The compiler should generate the same machine code for both code samples. The difference is cosmetic and primarily aimed at the reader of the code.
You do get a difference when you add else clauses:
if (strcmp(Buff1(), Config1) == 0)
{
if (strcmp(Buff2, Config2) == 0)
{
// my code goes here
}
else
{
// else 1
}
}
else
{
// else 2
}
Compared to:
if ((strcmp(Buff1(), Config1) == 0)) &&
(strcmp(Buff2, Config2) == 0))
{
// my code goes here
}
else
{
// Single else clause
}
In addition to Klas's answer(just in case you're not familiar with the AND operator) - the AND operator ('&&') checks the first condition and it continues to check the second condition -only if- the first condition is true.
So in your specific question, it checks if the first couple of strings are equal and only if true (are equal), it checks if the second couple are also equal.
The obvious optimization (not mentioned yet), if you know anything about those strings, is to first perform the compare that is more likely to fail.
So I've created my own function to compare two C Strings:
bool list::compareString(const char array1[], const char array2[])
{
unsigned char count;
for (count = 0; array1[count] != '\0' && array2[count] != '\0' && (array1[count] == array2[count] || array1[count + 32] == array2[count] || array1[count] == array2[count+32]); count++);
if (array1[count] == '\0' && array2[count] == '\0')
return true;
else
return false;
}
The parameter of my for loop is very long because it brings count to the end of at least one of the strings, and compares each char in each array in such a way that it their case won't matter (adding 32 to an uppercase char turns that char into its lowercase counterpart).
Now, I'm guessing that this is the most efficient way to go about comparing two C Strings, but that for loop is hard to read because of its length. What I've been told is to use a for loop instead of a while loop whenever possible because a for loop has the starting, ending, and incrementing conditions in its starting parameter, but for this, that seems like it may not apply.
What I'm asking is, how should I format this loop, and is there a more efficient way to do it?
Instead of indexing into the arrays with count, which you don't know the size of, you can instead operate directly on the pointers:
bool list::compareString(const char* array1, const char* array2)
{
while (*array1 != '\0' || *array2 != '\0')
if (*array1++ != *array2++) return false; // not the same character
return true;
}
For case insensitive comparison, replace the if condition with:
if (tolower(*array1++) != tolower(*array2++)) return false;
This does a safe character conversion to lower case.
The while loop checks if the strings are terminated. It continues while one of the strings is not yet terminated. If only 1 string has terminated, the next line - the if statement, will realize that the characters don't match (since only 1 character is '\0', and returns false.
If the strings differ at any point, the if statement returns false.
The if statement also post-increments the pointers so that it tests the next character in the next iteration of the while loop.
If both strings are equal, and terminate at the same time, at some point, the while condition will become false. In this case, the return true statement will execute.
If you want to write the tolower function yourself, you need to check that the character is a capital letter, and not a different type of character (eg. a number of symbol).
This would be:
inline char tolower(char ch)
{
return (ch >= 'A' && ch <= 'Z' ? (ch + 'a' - 'A') : ch);
}
I guess you are trying to do a case-insensitive comparison here. If you just need the fastest version, use a library function: strcasecmp or stricmp or strcmpi (name depends on your platform).
If you need to understand how to do it (I mean, is your question for learning purpose?), start with a readable version, something like this:
for (index = 0; ; ++index)
{
if (array1[index] == '\0' && array2[index] == '\0')
return true; // end of string reached
if (tolower(array1[index]) != tolower(array2[index]))
return false; // different characters discovered
}
Then measure its performance. If it's good enough, done. If not, investigate why (by looking at the machine code generated by the compiler). The first step in optimization might be replacing the tolower library function by a hand-crafted piece of code (which disregards non-English characters - is it what you want to do?):
int tolower(int c)
{
if (c >= 'A' && c <= 'Z')
return c + 'a' - 'A';
}
Note that I am still keeping the code readable. Readable code can be fast, because the compiler is going to optimize it.
array1[count + 32] == array2[count]
can lead to an OutOfRangeException, if the length of the array is smaller than 32.
You can use strcmp for comparing two strings
You have a few problems with your code.
What I'd do here is move some of your logic into the body of the for loop. Cramming everything into the for loop expression massively reduces readability without giving you any performance boosts that I can think of. The code just ends up being messy. Keep the conditions of the loop to testing incrementation and put the actual task in the body.
I'd also point out that you're not adding 32 to the character at all. You're adding it to the index of the array putting you at risk of running out of bounds. You need to test the value at the index, not the index itself.
Using an unsigned char to index an array gives you no benefits and only serves to reduce the maximum length of the strings that you can compare. Use an int.
You could restructure the code so that it looks like this:
bool list::compareString(const char array1[], const char array2[])
{
// Iterate over the strings until we find the string termination character
for (int count = 0; array1[count] != '\0' && array2[count] != '\0'; count++) {
// Note 0x20 is hexadecimal 32. We're comparing two letters for
// equality in a case insensitive way.
if ( (array1[count] | 0x20) != (array2[count] | 0x20) ) {
// Return false if the letters aren't equal
return false;
}
}
// We made it to the end of the loop. Strings are equal.
return true;
}
As for efficiency, it looks to me like you were trying to reduce:
The size of the variables that you're using to store data in
memory
The number of individual lines of code in your solution
Neither of these are worth your time. Efficiency is about how many steps (not lines of code, mind you) it will take to perform a task and how those steps scale as the inputs get bigger. For instance, how much slower would it be to compare the content of two novels for equality than two single word strings?
I hope that helps :)
I've written the code bits below. I have a constructor which takes five arguments. Unfortunately, the setGender method spits out a default 'M' for all instances of a class rather than setting the gender to the specified parameter. What am I doing incorrectly? Any advice would be greatly appreciated. Thank you.
DateProfile::DateProfile(char gdr,
char searchGdr, int romanceScale, int financeScale, string theName)
bool DateProfile::setGender(char gdr)
{
if (gdr != 'M' || gdr != 'F')
return false;
gender = gdr;
return true;
}
if (gdr != 'M' || gdr != 'F') is always true, irrespective of input. If you're passing 'M', the second part of the expression is true. If you're passing anything else, the first part of the expression becomes true.
What you meant to write is if (gdr != 'M' && gdr != 'F') instead.
Increasing the warning level of your compiler may have helped you spot the error. Most compilers will warn about expressions always evaluating to a single value, or at least about the unreachable code following it.
This is the snippet of code that I'm puzzled about. I'm checking for how long an incoming string is. I've appended * in order to have a sentinel value to stop the while loop. Yet, I'm consistently getting a length value that is inclusive of the * and I don't understand why, since the while loop with the nested if ought to stop prior to the *. Can someone point out what I'm doing wrong and why I'm having this issue?
void conversion(string romanIn)
{
length=0;
romanIn.append("*");
while(item!="*")
{
if(item != "*")
{
item = romanIn[length];
length++;
}
cout<<item;
}
you are naturally going to get a +1 the first time through the loop because you aren't initializing the variable "item". Also make it a do while instead of a while loop.
Try this:
do
{
// This line moves out of the if statement
item = romanIn[length];
if(item != "*")
{
length++;
}
cout<<item;
}while(item!="*")
What is the initial value of item?
Let's assume it's 0. You enter the loop
item == 0 != marker, so you enter the if as well, and you say
item = romanIn[0], length++
If romanIn[0] == "*" you will exit the loop, but your length now says 1 which includes the marker
I have a quick question regarding proper use of syntax, basically i am trying to summarise these two if statements into one if statement.
if (sc.LastCallToFunction) {
if (p_LowRectanglesList != NULL) {
free(p_LowRectanglesList);
sc.PersistVars->i1 = 0;
}
if (p_HighRectanglesList != NULL) {
free(p_HighRectanglesList);
sc.PersistVars->i2 = 0;
}
return;
Would it be syntactically correct to rewrite this as:
if (sc.LastCallToFunction) {
if (p_LowRectanglesList || p_HighrectangleList != NULL) {
free(p_LowRectanglesList && p_HighRectanglesList);
sc.PersistVars->i1 && sc.PersistVars->i2 = 0;
}
return;
Or would the compiler not accept this / Is my Logic faulty?
you can't do it the way you have given
if (p_LowRectanglesList || p_HighrectangleList != NULL)
this logically ORs the first pointer ( treats it as true or false ) with the comparison of the seond pointer to NULL
free(p_LowRectanglesList && p_HighRectanglesList);
this logically &&s the pointers together and then tries to free the result of that operation. ie, you are trying to free "true" or "false"
sc.PersistVars->i1 && sc.PersistVars->i2 = 0;
this logically ands the two things together, which will result in true or false and then trys to assign 0 to it..... doesn't make any sense at all.
Also, in your original code....after the free, you should put p_LowRectanglesList=NULL;