TCL & Perl Web scraper from The Practice of Programming by Brian W. Kernighan and Rob Pike

geturl.tcl

# geturl.tcl: retrieve document from URL
# input has form [http://]abc.def.com[/whatever...]

regsub "http://" $argv "" argv      ;# remove http:// if present
regsub "/" $argv " " argv           ;# replace leading / with blank

set so [socket [lindex $argv 0] 80] ;# make network connection
set q "/[lindex $argv 1]"

puts $so "GET $q HTTP/1.0\n\n"      ;# send request
flush $so
while {[gets $so line] >= 0 && $line != ""} {}  ;# skip header
puts [read $so]

unhtml.pl

# unhtml.pl: delete HTML tags

while (<>) {            # collect all input into single string
    $str .= $_;         # by concatenating input lines
}

$str =~ s/<[^>]*//g;    # delete <...>
$str =~ s/&nbsp;/ /g;   # replace &nbsp; by blank
$str =~ s/\s+/\n/g;     # compress white space
print $str