On Mon, 2 Sep 2002, Tim Booher wrote:
> Hello, I am currently frustrated trying to do a simple find and replace
> on a file, but the three open options: Read, Write and Append don't seem
> to give me what I want. For example, Read, works on the finding, but
> nothing will replace. Write erases all the files and leaves me with
> blank files and append adds to the end.
>
> I tried something like:
>
> open(MYFILE,'+<',$myfile);
>
> and that didn't work. If you are interested here is my code:
> .. . .
> open(HTMFILE,'+<', $filename) or die "Can't Open: $!"; # now
> go through all .htm files and make changes
> while (<HTMFILE>) {
> $_ =~ s/leftmargin="0".*bgcolor="#3c3939"//g;
> $_ =~ s/bgcolor="\#5A5D4E"/class="topOfPic"/ig;
You are modifying the line that you have read in, but you are not writing
it back. You will have to open a temporary file write the modified
contents to it and copy it back to original file for a find and replace.
There are other ways to this, for e.g. using $^I (perldoc perlvar)
Here is an e.g. where I find and replace 'abcd' with 'efgh' in the file
'test.txt'. Actually perl internally renames the original file, opens a
new file with the name of the original file to do the processing. In
effect it is pretty much the same.
my $myfile = 'test.txt';
{
local @ARGV = ($myfile);
local $^I = '~'; # perldoc perlvar,
# perldoc perlrun, read through section explaining the -i switch
while (<>) { # perldoc perlsyn
s/abcd/efgh/g;
print;
}
}
The same can also be done as an oneliner
perl -i~ -pe 's/abcd/efgh/g' test.txt
You can use any one of these methods for your job.
Note: For HTML parsing you might also want to take look at modules that
might do this for you. http://search.cpan.org
> }
> close(HTMFILE);
> . . .
>
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
