Seek and delete

Anonymous Monk has asked for the wisdom of the Perl Monks concerning the following question:

Replies are listed 'Best First'.
Re: Seek and delete by ikegami (Patriarch) on May 22, 2008 at 09:36 UTC
The simplest way is to copy the desired lines to a temporary store (an array or a temporary file) then write the contents of the store over the existing file (or rename the temporary file). One specific implementation: `perl -i -ne"next if $.<100; print" file` [download]	[reply] [d/l]
Re: Seek and delete by apl (Monsignor) on May 22, 2008 at 09:54 UTC
I am trying to achieve a very simple task, or so I thought till I started trying to accomplish it. Could we see what you tried to implement? What was difficult about this "simple task"?	[reply]
Re: Seek and delete by pc88mxer (Vicar) on May 22, 2008 at 13:14 UTC
Here's a solution using `Tie::File`: `use Tie::File; tie @array, ’Tie::File’, "filename" or die "tie failed: $!"; # remove the first 100 lines: splice(@array, 0, 100); untie @array;` [download] If you don't know the exact line number, then you could do something like this: `tie @array, 'Tie::File', "filename" or die ...; my $i; for (@array) { if (m/some regex/) { last } $i++; } splice(@array, 0, $i);` [download]	[reply] [d/l] [select]
Re: Seek and delete by toolic (Bishop) on May 22, 2008 at 16:32 UTC
If you are just doing quick and dirty stuff on nix, there's always sed*. This deletes the 1st 5 lines of a file: `sed 1,5d file.txt` [download] This deletes the all lines of a file up until the line including the string "end_of_header" (inclusive): `sed 1,/end_of_header/d file.txt` [download]	[reply] [d/l] [select]
Re: Seek and delete by johngg (Canon) on May 22, 2008 at 21:36 UTC
You could use the flip-flop operator (`..`). `$ cat lines line 1 line 2 line 3 line 4 line 5 line 6 $ perl -ne 'print if $. == 3 .. eof;' lines line 3 line 4 line 5 line 6 $ perl -ne 'print if /5/ .. eof;' lines line 5 line 6 $` [download] I hope this is useful. Cheers, JohnGG	[reply] [d/l] [select]