note to self....

#!/usr/bin/perl -w
#
# getazlyric.pl - gets an lyric from azlyrics.com 
#

#  SYNTAX
#
#   getazlyric.pl <artist> <song> <file>
#
#  OPERATION
#   This is a wrapper for html2text, azlyricfilter, and tee.
#   The lyric is retrieved by html2text with HTML markup
#   converted to text formating.  The azlyricfilter program
#   strips the remaining text formatting, including text
#   tagging added by azlyrics.com ("[name] LYRICS" header

#   and www.azlyrics.com link text at the bottom).
#
#   The cleaned text file is saved to <file> and dumped to 
#   stdout. 
#
#   The "Done" message that is sent to stdout is not included 
#   in the saved file.
#
#  PARAMETERS
#
#   The format of artist names and and lyric file/song title

#   file names at azlyrics.com is all lower case, no spaces
#   or special characters.
#
#   The base URL for azlyrics.com lyric files is
#
#    http://www.azlyrics.com/lyrics/<artist>/<song>.html
#
#   so to get a file, pick pass the name of the artist and the
#   name of the song formatted as per above (no spaces, lower
#   case - do not include the .html on the song name).
#
#   Note that it might be a good idea to confirm that the 
#   artist/song file exists before trying to retreive it, 

#   since the default behaviour by azlyrics.ccom when a 
#   file is not found is to return an index of some kind.
#
#   The filename passed as <file> should be a writable filename
#   - no warning is given if <file> is being overwritten.
#
#  Revision History:
#   2005-11-01: pdwilso@gmail.com - initial revision
#
#  TO DO:
#

#   Add command line switch processing to use a "quiet" mode
#   where no output is sent to stdout. 
#
#   Add a switch to output only to stdout - for now,
#   use /dev/null as localfile to suppress file output.
#
#   Add a test to assure that <artist>/<song>.html exists on
#   azlyrics.com - fail with warning if URL doesn't exist.
#
#   Add a switch to issue a warning/query if <file> exists.

#
# ##############################################################
#
# "constants" - command variables
$tee = "tee";
$html2text = "html2text -nobs";
$azfilter="/usr/local/bin/azlyricfilter.pl";
#

# get parameter values
unless (scalar(@ARGV)==3) {
    print "Usage: ".__FILE__." artist song localfile\n";
    exit;
}
$artist = shift;

$song = shift;
$localfile = shift;
# 
# construct the URL of the song from
$url = "http://www.azlyrics.com/lyrics/$artist/$song.html";
#
# backtick execution causes the output of the command to be 
# returned by the command -  the tee command causes the output
# of the command string to be returned by the backtick command
# execution mechanism - the print statement places the output
# of the command in backticks on std out 

print `$html2text $url | $azfilter | $tee $localfile`;
#
# done - might want to remove this 
print "\n *** Done.\n";

posted by 0x0000  # 23:02