Display last record

[trailmonk]

We are hoping that an end user will be able to click on a URL link and for that link to point to the last record of our Perl database.

I'm not sure if this helps but, right now, our display.cgi code reads:

#  Read in Data
###############################
open (DATABASE, "$data_file");
@data=<DATABASE>;    
close (DATABASE);
chomp @data;

We could reverse the way the data is displayed by adding:

@reverse = reverse @data;        
@data = @reverse;

but this is a quick fix and it is not what we truly desire. We desire that the end user has an option to click on a URL link so that only the last record is displayed. Do you have an perls of wisdom about this?

jdporter added code tags

[davidrw]

There's no need for a temporary array when doing:

  @reverse = reverse @data;
  @data = @reverse;

You can just do: @data = reverse @data and then print $data[0];. But to get the last item, you don't need to reverse the list -- just use the $#data or -1 index .. these are all the same:

  print $data[-1];
  print $data[$#data];
  print pop @data;    # NOTE: THIS SHORTENS THE ARRAY
  print scalar splice(@data,-1);   # NOTE: SAME AS pop
# or:
  @data = reverse @data;  # OF COURSE, @data IS NOW MODIFIED
  print $data[0];
  print shift @data;  # NOTE: THIS SHORTENS THE ARRAY
  print scalar splice(@data,0,1);  # NOTE: SAME AS shift



All of that is assuming you read your whole file into an array in memory, which is obviously costly for memory if the file is large ... A quick search revealed Read Last Line of A File Only

[ikegami]

Don't forget @data can be eliminated totally with (<DATABASE>)[-1]. It still loads the whole file into memory, though.

[ikegami]

Have you looked at File::ReadBackwards? It reads a file line by line from the end of the file. It does so efficiently. It doesn't read the entire file, much less load it entirely into memory.

[davido]

Just to add to what ikegami said, File::ReadBackwards will not only read the file line by line, starting at the end of the file and working backwards, but it will read record by record (which usually is line by line, but doesn't have to be). This is a pretty flexible solution. In fact, unlike the plain old <> operator, File::ReadBackwards allows you to set the record separator based on either literal text, or a regular expression. Nice!

[TedPride]

Just move the file pointer to the end of the file minus whatever the maximum record size is, then read everything to the end of the file and extract the last record. This should be quite efficient even with huge files.

[Happy-the-monk]

Just move the file pointer to the end of the file minus whatever the maximum record size is, then read everything

Without example code to back it up, this must look quite a bit obscure to a newbee.

Being concerned of large files I would opt for the answer in perlfaq5 to perldoc -q "a line in a file" and use Tie::File.

Cheers, Sören

[trailmonk]

Thank you for your help. I think that I did not state my question accurately. What I *believe* will have to happen is that the end user will click on the "URL" and it will point to a subroutine. The subroutine will be an if->then->else, n+1 that searches line-by-line until it reaches the end of the file. At that point it will "print" or display the last record. I'm just guessing. You are awesome and I am grateful for your help.