Created
June 13, 2010 15:51
-
-
Save wtnabe/436764 to your computer and use it in GitHub Desktop.
A grep for Skype HTML log written in Ruby
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#! /usr/bin/env ruby | |
# -*- coding: utf-8 -*- | |
require 'optparse' | |
require 'rubygems' unless defined? Gem | |
require 'nokogiri' | |
require 'highline' | |
$KCODE = 'u' unless defined? Encode | |
class Skype_Html_Log_Parser | |
def initialize | |
@date = nil | |
@datelist = nil | |
@key = nil | |
@regexp = nil | |
@regexo = [] | |
end | |
def run | |
args.parse!( ARGV ) | |
expand( grep( messages( detect_chatdate() ) ) ) | |
end | |
def grep( messages ) | |
regex = Regexp.new( @regexp, @regexo.inject( 0 ) { |r, i| r | i } ) | |
messages.find_all { |e| | |
e[@key.to_sym].match( regex ) | |
} | |
end | |
# | |
# [ { :name => name, | |
# :message => message, | |
# :time => time }, | |
# {}, | |
# {} | |
# ] | |
# | |
def messages( date ) | |
result = [] | |
ele = parser( html_log ).search( 'h3' ).find { |e| | |
e.text == date | |
} | |
message = {} | |
while ( ele = ele.next_sibling ) | |
content = text( ele ) | |
if ( content ) | |
content.keys.each { |k| | |
if ( message.has_key?( k ) ) | |
result << message | |
message = {} | |
end | |
message[k] = content[k] | |
} | |
else | |
break | |
end | |
end | |
return result | |
end | |
def text( ele ) | |
case ( ele.name ) | |
when 'dt' | |
ele.text.match( /\A(.+): (.+)\z/ ) | |
{ :name => $1, | |
:time => $2 } | |
when 'dd' | |
{ :message => ele.text } | |
when 'h3' | |
false | |
end | |
end | |
# | |
# [Return] Hash | |
# | |
def datelist | |
if ( !@datelist ) | |
list = parser( html_log ).search( 'h3' ).reverse.map { |e| | |
e.inner_text().strip | |
} | |
@datelist = Hash[*(1..list.size).to_a.zip( list ).flatten] | |
end | |
return @datelist | |
end | |
def detect_chatdate | |
if ( STDIN.tty? ) | |
return _chatdate_from_tty | |
else | |
return _chatdate_from_cmdopt | |
end | |
end | |
def _chatdate_from_tty | |
puts "logged days below:" | |
datelist.each_pair{ |k, v| | |
printf( "% 3d : %s\n", k, v ) | |
} | |
answer = HighLine.new.ask( "choose a day [ 1 - #{datelist.size} ] :", | |
Integer ) { |num| | |
num.in = datelist.keys | |
} | |
return datelist[answer] | |
end | |
def _chatdate_from_cmdopt | |
if ( datelist.has_value?( @date ) ) | |
return @date | |
else | |
return _default_chatdate() | |
end | |
end | |
def _default_chatdate | |
return datelist[datelist.keys.min] | |
end | |
def html_log | |
ARGV.size > 0 ? open( ARGV.shift ) : STDIN | |
end | |
def parser( strio = nil ) | |
if ( !@parser and strio ) | |
@parser = Nokogiri( strio.read ) | |
end | |
return @parser | |
end | |
def args | |
OptionParser.new { |opt| | |
opt.on( '-d DATE' ) { |date| | |
@date = date | |
} | |
opt.on( '-k KEY' ) { |key| | |
@key = key | |
} | |
opt.on( '-r REGEXP' ) { |regexp| | |
@regexp = regexp | |
} | |
opt.on( '-i', '--ignorecase' ) { | |
@regexo << Regexp::IGNORECASE | |
} | |
opt.on( '-m', '--multiline' ) { | |
@regexo << Regexp::MULTILINE | |
} | |
opt.on( '-x', '--extended' ) { | |
@regexo << Regexp::EXTENDED | |
} | |
} | |
end | |
def expand( messages ) | |
messages.map { |e| | |
sprintf( '%s : %s : %s', e[:time], e[:name], e[:message] ) | |
} | |
end | |
end | |
if ( __FILE__ == $0 ) | |
puts Skype_Html_Log_Parser.new.run | |
end |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment