Finding carriage returns (\c) in a line

On second thought, it's a bit unclear what you are trying to achieve. Do you want to get only the line that ends with a particular character/substring?

In this case you could use a combination of
getline() to get a line from a text file,
rfind() to find the last occurence of a character/substring,
and size() (for example) to test if the substring is at the very end of the string.

Quote:

---

Originally Posted by Wraithan

Bumping is against the rules.

---

Sorry :o.

Quote:

---

Originally Posted by anon

On second thought, it a bit unclear what you are trying to achieve. Do you want to get only the line that ends with a particular character/substring?

In this case you could use a combination of
getline() to get a line from a text file,
rfind() to find the last occurence of a character/substring,
and size() (for example) to test if the substring is at the very end of the string.

Code:

---

getline(FileIn, line);
pos = line.rfind(substring);
if (pos == npos && substring.size()+pos == line.size()
//you got it

---

---

Please forgive my awful explanation. So here it is, I'll attempt again:

Essentially, I'm looking for occurrences of a particular string within a line. For example, imagine the following file:

Code:

---

Line 1
Line 2
Line 3

---

Assuming this file is saved in Windows format, if I searched for "Line 1\c" then I should return the line in which the string I searched for is found - namely, "Line 1" should be returned.

Does this make more sense?

Thanks.

But what on earth is \c?

Quote:

---

Originally Posted by Mario F.

But what on earth is \c?

---

We're using it as the carriage return character. As I understand now, the C++ representation of character 13 is \r. In our program, the user would type "\c"

Thanks.

Quote:

---

Originally Posted by JizJizJiz

We're using it as the carriage return character. As I understand now, the C++ representation of character 13 is \r. In our program, the user would type "\c"

---

So the input file you would be searching would look like this?

Code:

---

Line 1\cLine 2\cLine 3

---

As a txt file everything would be on one line? And now you want to break it to actual lines? (- Replace all "\\c"-s with "\r\n")
Or search for "Line 1"? (s.find("Line 1")
Or search for "Line 1\c" (s.find("Line 1\\c")?

Why do you make the user type "\c" in the first place instead of just letting them hit Enter?

Quote:

---

Originally Posted by anon

So the input file you would be searching would look like this?

Code:

---

Line 1\cLine 2\cLine 3

---

As a txt file everything would be on one line? And now you want to break it to actual lines? Or search for "Line 1"? Or search for "Line 1\c"?

Why do you make the user type "\c" in the first place instead of just letting them hit Enter?

---

The text file itself would have the actual new lines. So like, it would look just like what I posted above. However, say a customer wanted to search for a carriage return following a word, such as "Line 1\c"

The Windows binary representation for Line 1 should be something like Line 1\r\l. Since I'm replacing \c with \r in the actual search.

The problem with \c is that the user would actually have to input "\\c".

Quote:

---

Originally Posted by Mario F.

The problem with \c is that the user would actually have to input "\\c".

---

Well, I parse the user's input. Say the user writes: "Line 1\c" - what I actually pass to the function that does the file searching is: "Line 1\r"

I think this works, right?

Lol.. look Jiz... you are either confusing yourself to death, or us :)

Forget \c ok... Lets just pretend \c doesn't exist. Better, let's not pretend. \c doesn't exist. ok?

Now...

What you want is to search for a line that ends with a given string. And anon already told you how to do that. With getline and such... look at that post of his.

Now... you also want to return the line number at which that happens. For that you use anon code and wrap it inside a loop. You increment a counter inside the loop. Everytime the string is found, you output the counter.

Ok! Why you have to replace '\r' with '\c' just beats me, but OK.

Are you sure you understand that '\r' as a carriage return symbol is one character, as opposed to "\\r" which is two characters (backslash+r). If you understand that, there should be no problems...

I think I'm confusing myself to death, and in the process, everyone else too. I'm sorry :P. I kinda feel like an idiot now.

So essentially, I'm writing a program that works on both Windows and Unix. It needs to search through a file and return lines that contain a certain string. I'm beginning to question myself why I should use \r. I'm going to try to figure things out and get back to this thread.

If I'm not wrong, getline() get's a line from a text file, no matter what the OS is? So this should be the way to go? It ignores line delimiters, right, so you wouldn't have to worry about them at all. And if you want to put a newline in a text file, you'd just use endl or '\n' ?

Quote:

---

Originally Posted by anon

If I'm not wrong, getline() get's a line from a text file, no matter what the OS is? So this should be the way to go? It ignores line delimiters, right, so you wouldn't have to worry about them at all. And if you want to put a newline in a text file, you'd just use endl or '\n' ?

---

Ah, if getline() ignores line delimiters, that may be why my original method isn't working...is there something else I should try using?

And anon has pointed out, it's just one character, so all you need is:

Code:

---

else if (contains[contains.size() - 1] == '\r')

---

Or if you prefer hex:

Code:

---

else if (contains[contains.size() - 1] == 0xd)

---