# Google Image Crawler Setup
User-agent: Googlebot-Image
Disallow: /

# Block YandexBot
User-agent: YandexBot
Disallow: /

# Block MegaIndex.ru
User-agent: MegaIndex.ru
Disallow: /

# Block dotbot
User-agent: dotbot
Disallow: /

# Block Sogou web spider
User-agent: Sogou web spider
Disallow: /

# Block SemrushBot
User-agent: SemrushBot
Disallow: /

# Block Baiduspider
User-agent: Baiduspider
Disallow: /

Sitemap: http://extension.psu.edu/sitemap/extension/sitemap.xml

# Crawlers Setup
User-agent: *
#Crawl-delay: 5

# Do not index courses pages hosted in Plone
Disallow: /courses

# # Do not index iwd preview pages
Disallow: /*?iwd_preview12345*

# # Do not index test product
User-agent: *
Disallow: /testing-reminder-iwd
Disallow: /test-product-iwd
Disallow: /york-county-iwd

# # Do not index the page subcategories that are sorted or filtered.
Disallow: /*?dir=
Disallow: /*&dir=
Disallow: /*?limit=
Disallow: /*&limit=
Disallow: /*?mode=
Disallow: /*&mode=
Disallow: /*?order=
Disallow: /*&order=
Disallow: /*?event=
Disallow: /*&event=

# Directories
Disallow: /404/
Disallow: /app/
Disallow: /cgi-bin/
Disallow: /downloader/
Disallow: /errors/
Disallow: /includes/
#Disallow: /js/
#Disallow: /lib/
Disallow: /magento/
#Disallow: /media/
Disallow: /pkginfo/
Disallow: /report/
Disallow: /scripts/
Disallow: /shell/
Disallow: /skin/
Disallow: /stats/
Disallow: /var/
Disallow: /.git/

# Paths (clean URLs)
Disallow: /index.php/
Disallow: /catalog/product_compare/
Disallow: /catalog/category/view/
Disallow: /catalog/product/view/
Disallow: /catalogsearch/
#Disallow: /checkout/
Disallow: /control/
Disallow: /contacts/
Disallow: /customer/
Disallow: /customize/
Disallow: /newsletter/
Disallow: /poll/
Disallow: /review/
Disallow: /sendfriend/
Disallow: /tag/
Disallow: /wishlist/
Disallow: /catalog/product/gallery/
Disallow: /slf-permit-training-internal

# Files
Disallow: /cron.php
Disallow: /cron.sh
Disallow: /error_log
Disallow: /install.php
Disallow: /LICENSE.html
Disallow: /LICENSE.txt
Disallow: /LICENSE_AFL.txt
Disallow: /STATUS.txt

# Paths (no clean URLs)
#Disallow: /*.js$
#Disallow: /*.css$
Disallow: /*.php$
Disallow: /*?SID=