#!/usr/bin/perl #Simple script that pulls the content off of #computerjobs.com and sto +res it #in a hash. This data is stored in a file (that is structured like a +hash) #so that you can later use the data to draw graphs.(like I do on atlan +tageek.com) I should have used Tie::DBI but I didnt know about it wh +en I first wrote this. use LWP::Simple; use Data::Dumper; my $cities=[ "Atlanta", "Carolina", "Texas", "Boston", "Chicago", "D.C. Metro", "Denver", "Detroit", "Florida", "Los Angeles", "New York", "Ohio", "Philadelphia", "Phoenix", "Portland", "Seattle", "Silicon Valley", "St. Louis", "Texas", "Twin Cities" ]; # Get Hash Data my $file_name = "jobs.dat"; my $hash_str = slurp($file_name); my $data_hash = eval($hash_str); my $content = &initContent(); foreach $city (@$cities) { my $day = getDay(); $data_hash->{$day}->{$city} = &getCount($city, $content); } print Dumper($data_hash); dump($file_name, Dumper($data_hash)); sub initContent { my $content = get("http://www.computerjobs.com/homepage.asp"); $content =~ s/\n//g; # Get rid of newlines $content =~ s/\r//g; # Get rid of carriage returns $content =~ s/<script.*?\/script.*?>//g; # get rid of JavaScript $content =~ s/<.*?>//g; # get rid of html tags return($content); } sub getCount { my $city = shift; my $city_content =shift; $city_content =~ /(${city}.*?([0-9,]+))/g; $city_count = $2; $city_count =~ s/,//g; return $city_count; } sub getDay(){ print time() . "\n"; $t = time() - time() % 86400; return $t/86400; } sub dump { $file = shift; $content = shift; open FIL, ">$file"; print FIL $content; close FIL; } sub slurp { $file =shift; undef $/; open FIL, "$file"; $output= <FIL>; close FIL; return $output; }
Edit Masem 2001-11-16 - CODE tags added resulting from section move
In reply to Job Counts by atlantageek
| For: | Use: | ||
| & | & | ||
| < | < | ||
| > | > | ||
| [ | [ | ||
| ] | ] |