#!/afs/isis/pkg/isis/bin/perl
# Last modified:  Time-stamp: <2003-12-09 13:37:10 haines>
#
# Abstract:  Get NOS data for each file in nos_locs.txt 
#      from http call dependent on station ID. Screen scrape data from html.
#
# Usage:  % get_noaa_nos_by_station.pl [-d]
#
# Author: Sara Haines (2003-09) 

# Processing:
#    (0) read in station list
# For each station
#    (1) Capture station data from http
#    (2) Screen scrape data from html and store to temporary file
#    (?) further limit it by glob (?)

$delimiter = "\n";

# if debugging requested print messages to STDOUT
if (grep /[debug|DEBUG|d]/, @ARGV) {
  $debug = 1;
  # remove it from the list of input args
  @ARGV = grep !/([debug|DEBUG|d])/, @ARGV;
}

# add libraries needed for this function
use POSIX qw(strftime);
use LWP::Simple;

# 
$now = strftime("%Y:%m:%d %H:%M:%S",gmtime);
if ($debug) { print "\n==== Starting: $now UTC ==== Perl Version: $]\n"; }

$starttime = time;

# (0) read in station list
$doc = get("file:/opt/local/seacoos/bin/nos_locs.txt");

# splitting a new line drops new lines.
@station_data = split(/$delimiter/, $doc);

# if (debug) { print $doc;}

# extract station_ids
# there's probably a better, cleaner way to do this in perl, but ...
foreach $_ (@station_data) {
	push @station_list, substr($_, 0, 7);    
}


$starttime2 = time;

for ($i=0; $i<=$#station_list; $i++) {

    @all_data = (); # clear all_data 
    $station_id = $station_list[$i];

    if ($debug) { print "$station_id\n";}

    $temp_data_dir = "/seacoos/data/nos/raw_data/temp";
    $temp_data_file = "$temp_data_dir/$station_id.txt";
    $nos_url = "http://tidesonline.nos.noaa.gov/data_read.shtml?station_info=$station_id";

    # (1) Get html document
    $doc = get($nos_url);
    # print $doc;

    # (2) Screen scrape http for data lines
    # some other possible html scrapes
    # @all_data = $doc =~ m{>(.*?)<}sgi;

    # match everything between <PRE> and </PRE> tag to scrape data
    # m{}s = match even if it includes newlines (test string as single line)
    # m{}g = do match more than once until end of stream (match globally)
    # m{}i = case insenstive match 
    @all_data = $doc =~ m{<PRE><FONT.*>(.*?)</PRE>}sgi;

    # 
    # if ($debug) {
    # 	print "------------------------------" . "\n";
    # 	print "Data scraped from " . "\n";
    # 	print "$nos_url:". "\n";
    # 	for (@all_data) { 
    # 	    print($_); 
    # 	  }
    #   }

    #print "Saving data to $temp_data_file ...\n\n\n";
    open(OUTFILE,">$temp_data_file")||
	die('Unable to write temporary data file...');
    $output_data=join("\n",@all_data);
    print OUTFILE "$output_data";

} # for ($i=0; $i<=$#station_list; $i++)

$subtime = time - $starttime2;
if ($debug) { print (" ... archive time = $subtime (seconds)\n"); }

$cummtime = time - $starttime;
if ($debug) { print ("Total script time = $cummtime (seconds)\n"); }