1
0
Fork 0
mirror of https://github.com/maybe-finance/maybe.git synced 2025-07-19 05:09:38 +02:00

Add support for different column separator in csv import logic (#1096)

* add col_sep to import model

* add validation for col_sep column

* add col_sep option to csv import model

* make use of col_sep option in import model

* add column separator field to new/edit action of an import

* add col_sep parameter to create/update action

* fix spacing between fields

Co-authored-by: Zach Gollwitzer <zach.gollwitzer@gmail.com>
Signed-off-by: Alexander Schrot <alexander@axs-labs.com>

---------

Signed-off-by: Alexander Schrot <alexander@axs-labs.com>
Co-authored-by: Zach Gollwitzer <zach.gollwitzer@gmail.com>
This commit is contained in:
Alexander Schrot 2024-08-16 20:00:16 +02:00 committed by GitHub
parent 707c5ca0ca
commit 4527482aa2
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
11 changed files with 117 additions and 19 deletions

View file

@ -18,14 +18,14 @@ class ImportsController < ApplicationController
def update
account = Current.family.accounts.find(params[:import][:account_id])
@import.update! account: account, col_sep: params[:import][:col_sep]
@import.update! account: account
redirect_to load_import_path(@import), notice: t(".import_updated")
end
def create
account = Current.family.accounts.find(params[:import][:account_id])
@import = Import.create!(account: account)
@import = Import.create! account: account, col_sep: params[:import][:col_sep]
redirect_to load_import_path(@import), notice: t(".import_created")
end

View file

@ -2,6 +2,7 @@ class Import < ApplicationRecord
belongs_to :account
validate :raw_csv_must_be_parsable
validates :col_sep, inclusion: { in: Csv::COL_SEP_LIST }
before_save :initialize_csv, if: :should_initialize_csv?
@ -88,7 +89,7 @@ class Import < ApplicationRecord
def get_raw_csv
return nil if raw_csv_str.nil?
Import::Csv.new(raw_csv_str)
Import::Csv.new(raw_csv_str, col_sep:)
end
def should_initialize_csv?
@ -102,7 +103,7 @@ class Import < ApplicationRecord
# Uses the user-provided raw CSV + mappings to generate a normalized CSV for the import
def generate_normalized_csv(csv_str)
Import::Csv.create_with_field_mappings(csv_str, expected_fields, column_mappings)
Import::Csv.create_with_field_mappings(csv_str, expected_fields, column_mappings, col_sep)
end
def update_csv(row_idx, col_idx, value)
@ -176,7 +177,7 @@ class Import < ApplicationRecord
def raw_csv_must_be_parsable
begin
CSV.parse(raw_csv_str || "")
CSV.parse(raw_csv_str || "", col_sep:)
rescue CSV::MalformedCSVError
# i18n-tasks-use t('activerecord.errors.models.import.attributes.raw_csv_str.invalid_csv_format')
errors.add(:raw_csv_str, :invalid_csv_format)

View file

@ -1,12 +1,20 @@
class Import::Csv
def self.parse_csv(csv_str)
CSV.parse((csv_str || "").strip, headers: true, converters: [ ->(str) { str&.strip } ])
DEFAULT_COL_SEP = ",".freeze
COL_SEP_LIST = [ DEFAULT_COL_SEP, ";" ].freeze
def self.parse_csv(csv_str, col_sep: DEFAULT_COL_SEP)
CSV.parse(
csv_str&.strip || "",
headers: true,
col_sep:,
converters: [ ->(str) { str&.strip } ]
)
end
def self.create_with_field_mappings(raw_csv_str, fields, field_mappings)
raw_csv = self.parse_csv(raw_csv_str)
def self.create_with_field_mappings(raw_csv_str, fields, field_mappings, col_sep = DEFAULT_COL_SEP)
raw_csv = self.parse_csv(raw_csv_str, col_sep:)
generated_csv_str = CSV.generate headers: fields.map { |f| f.key }, write_headers: true do |csv|
generated_csv_str = CSV.generate headers: fields.map { |f| f.key }, write_headers: true, col_sep: do |csv|
raw_csv.each do |row|
row_values = []
@ -22,18 +30,19 @@ class Import::Csv
end
end
new(generated_csv_str)
new(generated_csv_str, col_sep:)
end
attr_reader :csv_str
attr_reader :csv_str, :col_sep
def initialize(csv_str, column_validators: nil)
def initialize(csv_str, column_validators: nil, col_sep: DEFAULT_COL_SEP)
@csv_str = csv_str
@col_sep = col_sep
@column_validators = column_validators || {}
end
def table
@table ||= self.class.parse_csv(csv_str)
@table ||= self.class.parse_csv(csv_str, col_sep:)
end
def update_cell(row_idx, col_idx, value)

View file

@ -1,6 +1,7 @@
<%= styled_form_with model: @import do |form| %>
<div class="mb-4">
<div class="mb-4 space-y-3">
<%= form.collection_select :account_id, Current.family.accounts.alphabetically, :id, :name, { prompt: t(".select_account"), label: t(".account"), required: true } %>
<%= form.collection_select :col_sep, Import::Csv::COL_SEP_LIST, :to_s, -> { t(".col_sep_char.#{_1.ord}") }, { prompt: t(".select_col_sep"), label: t(".col_sep"), required: true } %>
</div>
<%= form.submit t(".next"), class: "px-4 py-2 block w-full rounded-lg bg-gray-900 text-white text-sm font-medium cursor-pointer hover:bg-gray-700" %>

View file

@ -58,8 +58,13 @@ en:
new: New Import
form:
account: Account
col_sep: CSV column separator
col_sep_char:
'44': Comma (,)
'59': Semicolon (;)
next: Next
select_account: Select account
select_col_sep: Select CSV column separator
import:
complete: Complete
completed_on: Completed on %{datetime}

View file

@ -0,0 +1,5 @@
class AddColSepToImports < ActiveRecord::Migration[7.2]
def change
add_column :imports, :col_sep, :string, default: ','
end
end

5
db/schema.rb generated
View file

@ -10,7 +10,7 @@
#
# It's strongly recommended that you check this file into your version control system.
ActiveRecord::Schema[7.2].define(version: 2024_08_15_190722) do
ActiveRecord::Schema[7.2].define(version: 2024_08_16_071555) do
# These are extensions that must be enabled in order to support this database
enable_extension "pgcrypto"
enable_extension "plpgsql"
@ -118,7 +118,7 @@ ActiveRecord::Schema[7.2].define(version: 2024_08_15_190722) do
t.boolean "is_active", default: true, null: false
t.date "last_sync_date"
t.uuid "institution_id"
t.virtual "classification", type: :string, as: "\nCASE\n WHEN ((accountable_type)::text = ANY ((ARRAY['Loan'::character varying, 'CreditCard'::character varying, 'OtherLiability'::character varying])::text[])) THEN 'liability'::text\n ELSE 'asset'::text\nEND", stored: true
t.virtual "classification", type: :string, as: "\nCASE\n WHEN ((accountable_type)::text = ANY (ARRAY[('Loan'::character varying)::text, ('CreditCard'::character varying)::text, ('OtherLiability'::character varying)::text])) THEN 'liability'::text\n ELSE 'asset'::text\nEND", stored: true
t.index ["accountable_type"], name: "index_accounts_on_accountable_type"
t.index ["family_id"], name: "index_accounts_on_family_id"
t.index ["institution_id"], name: "index_accounts_on_institution_id"
@ -293,6 +293,7 @@ ActiveRecord::Schema[7.2].define(version: 2024_08_15_190722) do
t.string "normalized_csv_str"
t.datetime "created_at", null: false
t.datetime "updated_at", null: false
t.string "col_sep", default: ","
t.index ["account_id"], name: "index_imports_on_account_id"
end

View file

@ -29,7 +29,7 @@ class ImportsControllerTest < ActionDispatch::IntegrationTest
test "should create import" do
assert_difference("Import.count") do
post imports_url, params: { import: { account_id: @user.family.accounts.first.id } }
post imports_url, params: { import: { account_id: @user.family.accounts.first.id, col_sep: "," } }
end
assert_redirected_to load_import_path(Import.ordered.first)
@ -41,7 +41,7 @@ class ImportsControllerTest < ActionDispatch::IntegrationTest
end
test "should update import" do
patch import_url(@empty_import), params: { import: { account_id: @empty_import.account_id } }
patch import_url(@empty_import), params: { import: { account_id: @empty_import.account_id, col_sep: "," } }
assert_redirected_to load_import_path(@empty_import)
end

View file

@ -36,6 +36,14 @@ class Import::CsvTest < ActiveSupport::TestCase
assert_not invalid_csv.valid?
end
test "CSV with semicolon column separator" do
csv = Import::Csv.new(valid_csv_str_with_semicolon_separator, col_sep: ";")
assert_equal %w[ date name category tags amount ], csv.table.headers
assert_equal 4, csv.table.size
assert_equal "Paycheck", csv.table[3][1]
end
test "csv with additional columns and empty values" do
csv = Import::Csv.new valid_csv_with_missing_data
assert csv.valid?
@ -81,6 +89,35 @@ class Import::CsvTest < ActiveSupport::TestCase
assert_equal "Amazon stuff", csv.table[1][1]
end
test "can create CSV with expected columns, field mappings with validators and semicolon column separator" do
date_field = Import::Field.new \
key: "date",
label: "Date",
validator: method(:validate_iso_date)
name_field = Import::Field.new \
key: "name",
label: "Name"
fields = [ date_field, name_field ]
raw_csv_str = <<-ROWS
date;Custom Field Header;extra_field
invalid_date_value;Starbucks drink;Food
2024-01-02;Amazon stuff;Shopping
ROWS
mappings = {
"name" => "Custom Field Header"
}
csv = Import::Csv.create_with_field_mappings(raw_csv_str, fields, mappings, ";")
assert_equal %w[ date name ], csv.table.headers
assert_equal 2, csv.table.size
assert_equal "Amazon stuff", csv.table[1][1]
end
private
def validate_iso_date(value)

View file

@ -10,6 +10,21 @@ class ImportTest < ActiveSupport::TestCase
@loaded_import.update! raw_csv_str: valid_csv_str
end
test "validates the correct col_sep" do
assert_equal ",", @empty_import.col_sep
assert @empty_import.valid?
@empty_import.col_sep = "invalid"
assert @empty_import.invalid?
@empty_import.col_sep = ","
assert @empty_import.valid?
@empty_import.col_sep = ";"
assert @empty_import.valid?
end
test "raw csv input must conform to csv spec" do
@empty_import.raw_csv_str = malformed_csv_str
assert_not @empty_import.valid?
@ -83,4 +98,18 @@ class ImportTest < ActiveSupport::TestCase
@empty_import.reload
assert @empty_import.failed?
end
test "can create transactions from csv with custom column separator" do
loaded_import = @empty_import.dup
loaded_import.update! raw_csv_str: valid_csv_str_with_semicolon_separator, col_sep: ";"
transactions = loaded_import.dry_run
assert_equal 4, transactions.count
data = transactions.first.as_json(only: [ :name, :amount, :date ])
assert_equal data, { "amount" => "8.55", "date" => "2024-01-01", "name" => "Starbucks drink" }
assert_equal valid_csv_str, loaded_import.normalized_csv_str
end
end

View file

@ -9,6 +9,16 @@ module ImportTestHelper
ROWS
end
def valid_csv_str_with_semicolon_separator
<<~ROWS
date;name;category;tags;amount
2024-01-01;Starbucks drink;Food & Drink;Tag1|Tag2;-8.55
2024-01-01;Etsy;Shopping;Tag1;-80.98
2024-01-02;Amazon stuff;Shopping;Tag2;-200
2024-01-03;Paycheck;Income;;1000
ROWS
end
def valid_csv_with_invalid_values
<<~ROWS
date,name,category,tags,amount