gistfile1.txt

# Import the college board schools
imported_schools = import_from(dat_file)

# Find schools matching CBIDs, combine ones with the same CBID, and update that school's data
for each college_board_school in imported_schools:
 matching_robertson_schools = find_schools_by_cbid(college_board_school.cbid) # There are sometimes several that match a CBID
 keeper_school = matching_robertson_schools.first
 keeper_school.update_info_using(college_board_school)
 duplicate_schools = matching_robertson_schools[1..-1]
 for each duplicate in duplicate_schools:
   replace_school(duplicate, keeper_school)
   
# Clear out CBIDs that don't exist
real_cbids = imported_schools.map(:cbid)
for each school in all_robertson_schools:
 unless real_cbids.includes(school.cbid):
   school.cbid = null

# Match up duplicate high schools that had no CBID entered
schools_without_cbid = all_robertson_schools.where('cbid is blank')
for each school in schools_without_cbid:
 matching_college_board_school = imported_schools.find_where:
   school.name == imported_school.name
   school.state == imported_school.state
   school.city == imported_school.city
   # Other required match criteria? I think this is sufficient, or we may have more false negatives
 if exists(matching_college_board_school):
   replace_school(shcool, matching_college_board_school)

function replace_school(duplicate, replacement):
 # A user_high_school is created for each high school attended by an applicant. Its fields include the dates and school attended.
 user_high_schools_referencing_this_duplicate = duplicate.user_high_schools
 for each user_high_school in user_high_schools_referencing_this_duplicate:
   user_high_school.high_school = replacement
 duplicate.delete