removing a single string/line from a text file

Hi

I have a text file containing one string on each line. My problem is I want to surgically remove just one line (or one string - same thing) without leaving a blank line.
ie, if I had the following lines in my text file:
thisisline1
thisisline2
thisisline3
thisisline4

I would like to get rid of, say, line 3, leaving:
thisisline1
thisisline2
thisisline4

I'm sure there must be an easy way to do this - a function hidden somewhere or something?! As far as I can tell, fscanf and fgets etc. can only copy the text, not 'cut' or remove it. I have scoured the man pages and searched the net and come up blank.

If anyone could shed some light on this for me, it would be much appreciated.
Thanks.

Originally Posted by turmoil

Hi

I have a text file containing one string on each line. My problem is I want to surgically remove just one line (or one string - same thing) without leaving a blank line.
ie, if I had the following lines in my text file:
thisisline1
thisisline2
thisisline3
thisisline4

I would like to get rid of, say, line 3, leaving:
thisisline1
thisisline2
thisisline4

I'm sure there must be an easy way to do this - a function hidden somewhere or something?! As far as I can tell, fscanf and fgets etc. can only copy the text, not 'cut' or remove it. I have scoured the man pages and searched the net and come up blank.

If anyone could shed some light on this for me, it would be much appreciated.
Thanks.

Nope, no hidden function (unless you write one yourself). Something like this requires you to bang out a solution yourself. Typically you would do something like this:

Code:

1.  Open the input file containing your lines of text.
2.  Open an output file to write to.
3.  Read a line from input file.
4.  If this is the line you want to remove, don't write it to output file.
    Otherwise write the line to the output file.
5.  While got more lines from input file go to step 4.
6.  Close both files.
7.  Delete the old file used as input.
8.  Rename the output to have same name as old input file.

If you were to store all the lines in a data structure such as a linked-list (or were perhaps inclined to use C++ for one of its handy container classes), then you could eliminate the deleting and renaming:

Code:

1.  Open input file.
2.  Read all lines and store in your linked list.
3.  Close input file.
4.  Remove line you don't want from the linked list (or simply don't store it in the first place).
    (You could also store everything and then skip the writing of the line you don't want to
    end up in the file during the output step below).
5.  Reopen the file previously used as input but now open it in write mode.
6.  Output everything in linked-list to file (minus the line in question, however you
    may have dealt with that issue).
7.  Close output file.

Hey thanks man. I'm supprised it takes such an effort to achieve that result, but hey, if it works... Well, I know it works, I just tried it. :-)

Thanks again.

Why would you be surprised? Just think of how a text editor works. You open a file. You change something. You save the file.

Text editors don't directly interface with the disk. They load everything into memory, you change it, then it saves over top of the file.

Quzah.

Why would you be surprised?

When you press delete at the beginning of a line, it does the same thing, but it appears to the user that one letter was removed and everything else automatically shifted back - you don't actually see all the data being written one column earlier.

And for all the OP knows, there may be a standard function already written to do this when given a certain element number as a parameter.

Which is entirely different than deleting a portion of the file. You haven't saved anything when you press the delete key. It's still in memory. Nothing saves unless you tell it to. Which means, it follows common sense that that's how you'd do it yourself. Load it up, edit it, save it.

But I guess they don't teach logic, and you can't teach common sense.

Quzah.