Add checks for canonical names

lib/sciolyff.rb

@@ -1,5 +1,7 @@
 # frozen_string_literal: true
 
+require 'csv'
+require 'open-uri'
 require 'psych'
 require 'date'
 

lib/sciolyff/validator/canonical.rb

@@ -0,0 +1,23 @@
+# frozen_string_literal: true
+
+module SciolyFF
+  # Logic for checking against canonical lists of events and schools
+  module Validator::Canonical
+    BASE = 'https://raw.githubusercontent.com/unosmium/canonical-names/master/'
+
+    private
+
+    def canonical?(rep, file, logger)
+      return true if @canonical_warned
+
+      @canonical_list ||= CSV.parse(URI.open(BASE + file))
+      # don't try to make this more efficient, harder than it looks because of
+      # nil comparisons
+      @canonical_list.include?(rep)
+    rescue StandardError => e
+      logger.warn "could not read canonical names file: #{BASE + file}"
+      logger.debug "#{e}\n  #{e.backtrace.first}"
+      @canonical_warned = true
+    end
+  end
+end

lib/sciolyff/validator/events.rb

@@ -2,6 +2,7 @@
 
 require 'sciolyff/validator/checker'
 require 'sciolyff/validator/sections'
+require 'sciolyff/validator/canonical'
 
 module SciolyFF
   # Checks for one event in the Events section of a SciolyFF file
@@ -78,6 +79,15 @@ def places_start_at_one?(event, logger)
         "#{lowest_place} instead of 1"
     end
 
+    include Validator::Canonical
+
+    def in_canonical_list?(event, logger)
+      rep = [event[:name]]
+      return true if canonical?(rep, 'events.csv', logger)
+
+      logger.warn "non-canonical event: #{event[:name]}"
+    end
+
     private
 
     def placings_by_place(event)

lib/sciolyff/validator/teams.rb

@@ -2,6 +2,7 @@
 
 require 'sciolyff/validator/checker'
 require 'sciolyff/validator/sections'
+require 'sciolyff/validator/canonical'
 
 module SciolyFF
   # Checks for one team in the Teams section of a SciolyFF file
@@ -70,6 +71,16 @@ def in_a_subdivision_if_possible?(team, logger)
       logger.warn "missing subdivision for 'team: #{team[:number]}'"
     end
 
+    include Validator::Canonical
+
+    def in_canonical_list?(team, logger)
+      rep = [team[:school], team[:city], team[:state]]
+      return true if canonical?(rep, 'schools.csv', logger)
+
+      location = rep[1..-1].compact.join ', '
+      logger.warn "non-canonical school: #{team[:school]} in #{location}"
+    end
+
     private
 
     def initialize_teams_info(teams)