Chad Perrin wrote:
Y'know, I was playing around with this to see if I could come up with a
reasonably elegant solution, and I noticed a problem:
Your code doesn't seem to actually work. Am I missing something?
I went back to check, and it seems to only be working in very specific
cases. For example, it changed <META NAME="... to <meta name="... and
left the contents of the "..." part alone. However, it did not change
<TITLE> to <title>, presumably because there were no '""' to match. I
tweaked it a little, to make sure that it picked up everything beginning
with a '<'. It now will catch things like <TITLE> and <HTML>, but I had
to cheat (as you can see in the code below).
As far as I can tell, this will work for the cases mentioned above, as
well as cases that have multiple tags on one line (something that
previous snippit didn't do). Now see if it works for you.
Adam
Here's the full code:
while (<$fh>) {
if ($. == 1) {
print "$new_head\n" unless /^<\!/; # "$new_head" contains
# DOCTYPE info
s/<\!(.*?)>/$new_head/; # puts in new DOCTYPE
} # info
if (!/<(img|meta|link|(h|b)r)(.*?)\/>/) { # finds
s/<(img|meta|link|(h|b)r)(.*?)>/<$1$3 \/>/; # self-closing
# tags and adds
# an '/' if
# it's lacking
}
if (/<(.*?)>/g && !/<\!/g) { # Should find all tags unless they
# are <! DOCTYPE... or comments
# (is this even necessary?).
s/<(HTML|HEAD|BODY|TITLE)>/<\L$1>/ig; # Cheating to get easy
# ones.
s/<(.*?)"(.*?)"/<\L$1\E"$2"/g; # Replace anything
# outside of ' "..." '
# with lowercase.
}
print;
}
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
<http://learn.perl.org/> <http://learn.perl.org/first-response>
