Don't anchor the regexp to EOL to fix matching issues that were causing the entire block of crontab entries to appear twice in the crontab. This could be due to the way cron is adding newlines or some other post-processing that occurs on the cron. This bug was reproduced in the following environment: Ubuntu 10.04.2 LTS, rvm 1.0.14, ruby 1.8.7 patchlevel 253, capistrano 2.5.19, capistrano-ext 1.2.1, bundler 1.0.10

lib/whenever/command_line.rb

@@ -89,19 +89,19 @@ module Whenever
     
     def updated_crontab   
       # Check for unopened or unclosed identifier blocks
-      if read_crontab =~ Regexp.new("^#{comment_open}$") && (read_crontab =~ Regexp.new("^#{comment_close}")).nil?
+      if read_crontab =~ Regexp.new("^#{comment_open}") && (read_crontab =~ Regexp.new("^#{comment_close}")).nil?
         warn "[fail] Unclosed indentifier; Your crontab file contains '#{comment_open}', but no '#{comment_close}'"
         exit(1)
-      elsif (read_crontab =~ Regexp.new("^#{comment_open}$")).nil? && read_crontab =~ Regexp.new("^#{comment_close}")
+      elsif (read_crontab =~ Regexp.new("^#{comment_open}")).nil? && read_crontab =~ Regexp.new("^#{comment_close}")
         warn "[fail] Unopened indentifier; Your crontab file contains '#{comment_close}', but no '#{comment_open}'"
         exit(1)
       end
       
       # If an existing identier block is found, replace it with the new cron entries
-      if read_crontab =~ Regexp.new("^#{comment_open}$") && read_crontab =~ Regexp.new("^#{comment_close}$")
+      if read_crontab =~ Regexp.new("^#{comment_open}") && read_crontab =~ Regexp.new("^#{comment_close}")
         # If the existing crontab file contains backslashes they get lost going through gsub.
         # .gsub('\\', '\\\\\\') preserves them. Go figure.
-        read_crontab.gsub(Regexp.new("^#{comment_open}$.+^#{comment_close}$", Regexp::MULTILINE), whenever_cron.chomp.gsub('\\', '\\\\\\'))
+        read_crontab.gsub(Regexp.new("^#{comment_open}.+^#{comment_close}", Regexp::MULTILINE), whenever_cron.chomp.gsub('\\', '\\\\\\'))
       else # Otherwise, append the new cron entries after any existing ones
         [read_crontab, whenever_cron].join("\n\n")
       end.gsub(/\n{3,}/, "\n\n") # More than two newlines becomes just two.