Rails View model data relating to another model - ruby-on-rails

Question

So I am just learning Ruby on Rails and I have come across an issue relating to viewing model data.
I have 2 models, Product and Review. As a product can have many reviews the relationship is set to product has_many reviews and reviews has_one product.
I am trying to display all the reviews for a product on the product details page. I have added a table of reviews to the show view for products. I then added #reviews=#product.reviews to my definition for show.
What is happening is that I am receiving an error for the loop that runs through each review <% #reviews.each do |review| %> stating Unknown column 'reviews.product_id'.
In my product model, I have a column named ProductId which I thought would be how the application retrieves the list of reviews for a product but it's not. Is this product_id just a unique value created by the framework?
Just wondering if I have done something wrong or if it is something I haven't implemented.
Show.html.erb;
Show.html.erb
Show definition;
Show definition in controller
My review table definition is schema.rb;
create_table "reviews", charset: "utf8mb4", force: :cascade do |t|
t.integer "ProfileId"
t.integer "ProductId"
t.string "Author"
t.integer "ProductRating"
t.string "ReviewText"
t.date "DateofReview"
t.datetime "created_at", precision: 6, null: false
t.datetime "updated_at", precision: 6, null: false
end
My product table definition is schema.rb;
create_table "products", charset: "utf8mb4", force: :cascade do |t|
t.string "pName"
t.string "pBrand"
t.integer "pCost"
t.string "pCategory"
t.datetime "pDate"
t.string "pDescription"
t.string "pPhoto"
t.datetime "created_at", precision: 6, null: false
t.datetime "updated_at", precision: 6, null: false
end

Answer 1

the convention and default columns of a database in rails are snake_case https://edgeguides.rubyonrails.org/active_record_basics.html#schema-conventions even though it does not fail when using camelCase like in your case there is a lot of areas where you may need to change and in your case here the has_many is looking for a foreign key based on the class + _id to fix this you need to specify the foreign key on there:
has_many :reviews, foreign_key: :ProductId
and a foreign_key on the has_one, also note that this should be a belongs_to in your case as the key lives inside the reviews table https://guides.rubyonrails.org/association_basics.html#the-belongs-to-association
belongs_to :product, foreign_key: :ProductId

Answer 2

Your migration is incorrect. Rails works on conventions. It does not use StudlyCaps, it uses snake_case. It expects the column to be product_id, not ProductId. Unless you're adapting Rails to an existing schema, it's best to follow Rails conventions.
The migrations should look like so:
# Prefer Symbols to Strings.
# See https://medium.com/#lcriswell/ruby-symbols-vs-strings-248842529fd9
# force: :cascade is questionable.
create_table :reviews, charset: :utf8mb4 do |t|
# Declare profile_id and product_id as indexed foreign keys
# Declare them `not null`, a review must have a reviewer and product.
t.belongs_to :profile, foreign_key: true, null: false
t.belongs_to :product, foreign_key: true, null: false
# Assuming Profile is the author of the review, Author is redundant.
# Drop the "product" prefix, we know from context what its rating.
# Presuming a review needs a rating, set it `not null`.
t.integer :rating, null: false
# Presuming a review needs a body, set it `not null`.
# Drop the "review" prefix from "review_text". "text" is too generic.
# It's the body of the review.
t.text :body, null: false
# DateOfReview is redundant with the created_at timestamp
# The standard created_at and updated_at timestamps
t.timestamps
end
# Don't prefix columns, its confusing and obfuscating,
# especially when applied inconsistently.
# Use table aliases and fully qualified column names instead.
# In Rails you will rarely be writing SQL by hand anyway.
create_table :products, charset: :utf8mb4 do |t|
# Presuming a product requires a name and cost, make them `not null`.
t.string :name, null: false
t.integer :cost, null: false
# Consider referencing categories and brands tables instead. It would make
# listing them and querying products by them faster and easier, and
# protects against typos.
t.string :brand
t.string :category
# Large blocks of text should be "text" not "string".
# string is varchar, text is text.
t.text :description
# Do not store binaries in the database. It is slow and bloated.
# Use ActiveStorage instead.
# https://guides.rubyonrails.org/active_storage_overview.html
t.string :photo
# pDate is redundant with created_at
t.timestamps
end
has_one is inappropriate here. That's for when you could have many, but only have one. You would use it if, for example, a Product could only have one Review.
Instead, for a simple one-to-many relationship a Review belongs_to Product.
class Review < ApplicationRecord
belongs_to :product
end
class Product < ApplicationRecord
has_many :reviews
end
In general, generate your models to get the basics, and then edit them.
For example, you might generate Product and Review like so and then edit the generated files.
$ rails g model Product name:string cost:integer description:text
$ rails g model Review product:belongs_to rating:integer