#
#  WWR Robots.txt
#  Eliminate things we don't want indexed to save load on site. 
#  Make sure good stuff (e.g. artists, House concerts, podcasts etc) is indexed
#
#
# Google allows the following extensions to robots.txt
#   Wildcard 
#	*  - Match sequence of characters
#	       e.g. all urls beginning with private /private*/ 
#	$  - Matching the end characters of a URL 
#		e.g. all urls ending in .asp  /*.asp$ 
#   Allow: 
#	-  Allow all URLS containing a ? 
#		Allow: /*?*
#   
# REMOVING in order to let 'em index 'em since they are loading them anyway
# Disallow:/*action=edit$

# JCK ADDED 1/22/2010 per Google Sitemaps info - broken in to 2 files because of size
Sitemap: http://wholewheatradio.org/wiki/sitemap1.xml
Sitemap: http://wholewheatradio.org/wiki/sitemap2.xml

User-agent: Googlebot
Disallow:/photos/
Disallow:/audio_am/
Disallow:/emoticons/
Disallow:/GIFS/
Disallow:/video/
Disallow:/promo/
Disallow:/feeds/
Disallow:/audio/
Disallow:/jbb/
Disallow:/WWR2/
Disallow:/houseconcerts/
Disallow:/wiki/index.php/Super_Tagger
Disallow:/wiki/index.php/Template:*
Disallow:/wiki/index.php/What%27s_playing/  
Disallow:/wiki/index.php/Archives*  
Disallow:/wiki/index.php/Reports:*
Disallow:/wiki/index.php/System_Variables* 
Disallow:/wiki/index.php/Special:*
Disallow:/wiki/index.php/Emoticon_cheat_sheet/  
Disallow:/wiki/index.php/Beta_requests/
Disallow:/wiki/index.php/Mediawiki*
Disallow:/wiki/images
Disallow:/wiki/index.php/Image:*
Disallow:/wiki/index.php/File:*
Disallow:/wiki/index.php/Category:*
Disallow:/wiki/index.php/System%20List*
Disallow:/wiki/index.php/Chat*
Disallow:/wiki/index.php/Keywords*
Disallow:/wiki/index.php/Emoticon*
Disallow:/wiki/index.php/Share_links*
Disallow:/wwemail.php
Disallow:/widgets.php*
Crawl-delay: 20

User-agent: Googlebot-Image
Disallow:/photos
Disallow:/audio_am
Disallow:/emoticons
Disallow:/GIFS
Disallow:/video
Disallow:/promo
Disallow:/wiki/index.php/Super_Tagger
Disallow:/wiki/index.php/Template:*
Disallow:/wiki/index.php/What%27s_playing
Disallow:/wiki/index.php/Archives*
Disallow:/wiki/index.php/Reports:*
Disallow:/wiki/index.php/System_Variables*
Disallow:/wiki/index.php/Special:*
Disallow:/wiki/index.php/Emoticon_cheat_sheet
Disallow:/wiki/index.php/Beta_requests
Disallow:/wiki/index.php/Mediawiki*
Disallow:/wiki/images
Disallow:/wiki/index.php/Image:*
Disallow:/wiki/index.php/File:*
Disallow:/wiki/index.php/Category:*
Disallow:/wiki/index.php/System%20List*
Disallow:/wiki/index.php/Chat*
Disallow:/wiki/index.php/Keywords*
Disallow:/wiki/index.php/Emoticon*
Disallow:/wiki/index.php/Share_links*
Disallow:/widgets.php*
Crawl-delay: 20

User-agent: Googlebot-Mobile
Disallow: /

User-agent: Mediapartners-Google
Disallow: /

User-agent: Adsbot-Google
Disallow: /

################ FOLLOWING WAS SNARFED FROM WIKIPEDIA'S ROBOTS.TXT #####################
#
# robots.txt for http://www.wikipedia.org/ and friends
#
# Please note: There are a lot of pages on this site, and there are
# some misbehaved spiders out there that go _way_ too fast. If you're
# irresponsible, your access to the site may be blocked.
#

# advertising-related bots:
User-agent: Mediapartners-Google*
Disallow: /

# Wikipedia work bots:
User-agent: IsraBot
Disallow:

User-agent: Orthogaffe
Disallow:

# Crawlers that are kind enough to obey, but which we'd rather not have
# unless they're feeding search engines.
User-agent: UbiCrawler
Disallow: /

User-agent: DOC
Disallow: /

User-agent: Zao
Disallow: /

# Some bots are known to be trouble, particularly those designed to copy
# entire sites. Please obey robots.txt.
User-agent: sitecheck.internetseer.com
Disallow: /

User-agent: Zealbot
Disallow: /

User-agent: MSIECrawler
Disallow: /

User-agent: SiteSnagger
Disallow: /

User-agent: WebStripper
Disallow: /

User-agent: WebCopier
Disallow: /

User-agent: Fetch
Disallow: /

User-agent: Offline Explorer
Disallow: /

User-agent: Teleport
Disallow: /

User-agent: TeleportPro
Disallow: /

User-agent: WebZIP
Disallow: /

User-agent: linko
Disallow: /

User-agent: HTTrack
Disallow: /

User-agent: Microsoft.URL.Control
Disallow: /

User-agent: Xenu
Disallow: /

User-agent: larbin
Disallow: /

User-agent: libwww
Disallow: /

User-agent: ZyBORG
Disallow: /

User-agent: Download Ninja
Disallow: /

#
# Sorry, wget in its recursive mode is a frequent problem.
# Please read the man page and use it properly; there is a
# --wait option you can use to set the delay between hits,
# for instance.
#
User-agent: wget
Disallow: /

#
# The 'grub' distributed client has been *very* poorly behaved.
#
User-agent: grub-client
Disallow: /

#
# Doesn't follow robots.txt anyway, but...
#
User-agent: k2spider
Disallow: /

#
# Hits many times per second, not acceptable
# http://www.nameprotect.com/botinfo.html
User-agent: NPBot
Disallow: /

# A capture bot, downloads gazillions of pages with no public benefit
# http://www.webreaper.net/
User-agent: WebReaper
Disallow: /

############### END OF WIKIPEDIA SNARF ########################

#Jim added - worthless engine - per http://www.majestic12.co.uk/projects/dsearch/mj12bot.php 
User-agent: MJ12bot
Disallow: /

#Jim added per http://www.commoncrawl.org/faq.htm
User-agent: ccbot
Disallow: /

#Jim added per http://discoveryengine.com/discobot.html
User-agent: discobot
Disallow: / 

#Jim added per http://www.exalead.com/software/services/knowledgebase/websearchengine/webmasterguide/
User-agent: Exabot
Disallow: /

#Jim added per http://www.metadatalabs.com/mlbot/#block
User-agent: MLBot
Disallow: / 

#Jim added per no page but guessed from this:  http://www.baidu.com/search/spider.htm
User-agent: Baiduspider
Disallow: / 

#Jim added per no page but guessed from this:  http://help.naver.com/customer_webtxt_01.jsp
User-agent: Naverbot
Disallow: / 

#Jim added per http://80legs.pbworks.com/FAQ#Howdoes80legscrawlwebpages
User-agent: 008
Disallow: /

#Jim added per http://www.dotnetdotcom.org/#info
User-agent: dotbot
Disallow: / 


#
#    Standard robots.txt rules for all other agents
#	Note: must be blank line between last disallow above and next
#	User-agent: 

User-agent: *
Disallow:/photos
Disallow:/audio_am
Disallow:/emoticons
Disallow:/GIFS
Disallow:/video
Disallow:/promo
Disallow:/wiki/index.php?title=*
Disallow:/wiki/index.php/What%27s_playing
Disallow:/wiki/index.php/Listener_Console
Disallow:/wiki/index.php/Keywords
Disallow:/wiki/images
Disallow:/wiki/index.php/Image:*

Disallow:/wiki/index.php/File:*
Disallow:/wiki/index.php/Category:*
Disallow:/wiki/index.php?title=Category:*
Disallow:/wiki/index.php/Template:*
Disallow:/wiki/index.php/Share_links*
Disallow:/widgets.php*

Crawl-delay: 20