Ruby .where count only once duplicated results

Question

In my Rails 6/Grape API app I've got a model SurveyResult where I store results of survey which user filed.

It's possible to have few survey results for a given question (it can be a multiple choice question):

[31] pry(main)> SurveyResult.where(survey: survey, user: user)
 #<SurveyResult:0x00007f9283adca10
  id: 18,
  user_id: 1,
  survey_id: 1,
  cms_question_id: 5844869,
  cms_answer_id: 321,

 #<SurveyResult:0x00007f9283adc7e0
  id: 20,
  user_id: 1,
  survey_id: 1,
  cms_question_id: 5899779,
  cms_answer_id: 0987,

 #<SurveyResult:0x00007f9283adc920
  id: 21,
  user_id: 1,
  survey_id: 1,
  cms_question_id: 5899779,
  cms_answer_id: 12, 

For example there are two results for the same cms_question_id - 5899779 (as above). Now if I want to count all user SurveyResults I will get below result:

[31] pry(main)> SurveyResult.where(survey: survey, user: user).count
=> 3

How to modify this query to count SurveyResult with the same cms_question_id only once? The expected result for above example should be:

[31] pry(main)> SurveyResult.where(survey: survey, user: user).count
=> 2

Answer 1

Looking at the count documentation that would be:

SurveyResult.where(survey: survey, user: user).distinct.count(:cms_question_id)

count(column_name = nil)

Count the records.

Person.count
# => the total count of all people

Person.count(:age)
# => returns the total count of all people whose age is present in database

Person.count(:all)
# => performs a COUNT(*) (:all is an alias for '*')

Person.distinct.count(:age)
# => counts the number of different age values