pgwillia

require 'application_system_test_case'

class ItemEditTest < ApplicationSystemTestCase

  test 'download path is stable after edit' do
    admin = users(:user_admin)
    item = items(:item_admin)
    item.save

    # Need to add file to item to be able to go through editing wizard.

pgwillia

pgwillia

$ tree 05296aa6-4ced-4cf0-a612-b008fd40a404_after_file_delete/
05296aa6-4ced-4cf0-a612-b008fd40a404_after_file_delete/
└── 05296aa6-4ced-4cf0-a612-b008fd40a404
    ├── bag-info.txt
    ├── bagit.txt
    ├── data
    │   ├── logs
    │   │   ├── aipcreation.log
    │   │   └── files_logs
    │   │       └── a8592fcf-af2c-4f44-921a-b98a0be467c3

pgwillia

Running rubocop can show where this code strays from best practices/community standards. It doesn't make sense to make these changes to the code, we've already talked about the code being somewhat disposable, but could be informative to know about.

• Style/StringConcatenation: Prefer string interpolation to string concatenation.

For example

# rails/generate_report_12.rb:28:110
missing_community_collection_report[:community] << "Community Id %{community_id} does not exist" % {community_id: community_collection[0]}          

becomes "Community Id #{community_collection[0]} does not exist"

pgwillia

empty_collections = Collection.find_each.filter {|collection| collection.member_objects.size == 0 }
CSV.open('empty_collections.csv', 'wb') do |csv|
  csv << ['Community', 'Collection', 'url']
  empty_collections.each do |collection|
    community_title = Community.find(collection.community_id)
    url = Rails.application.routes.url_helpers.community_collection_url(collection.community_id, collection.id)
    csv << [community_title, collection.title, url]
  end
end

pgwillia

The Ask

March 2022, Weiwei asked,

We are looking to create an extract of around 4 dozen items in ERA as sample data for Atmire to evaluate and propose for migration charges. See Leah's message below. I wonder if this could be something you coordinate with Leah and someone from the metadata team? You are probably the one who has the most comprehensive knowledge of ERA and the history of some of these items?

Ignace Deroost of Atmire said,

A CSV format usually works best. So if you could export your samples in CSV that would be great!

From our DSpace investigation I had observed a csv export from their software (dspace_sample_export)

The Work

pgwillia

require "ostruct"
@item = OpenStruct.new(creators: ["Joe Shmoe"], title: "Test title", descripton: "Test description", sort_year: 2021) 
def datacite_metadata
    {
      data: {
        attributes: {
          creators: @item.creators.map{|creator| {name: creator}},
          titles: [{
            title: @item.title
          }],

pgwillia

Problem:

I have an object that has a csv file attached using ActiveStorage.

I'm going to do some operations based on the content of that csv file so I want to validate that the file I'm attaching has the information I will need.

So the class looks something like this:

class MySpreadsheet < ApplicationRecord

pgwillia

require "csv"

headers = ["year", "date", "time", "speakers", "title", "abstract", "type", "tags", "slides", "video", "audio"]
# create new csv for the final output with the final header:w
CSV.open("all.csv", "wb", write_headers: true, headers: headers) do |csv|
  Dir["csv/*.csv"].each do |path|  # for each of your csv files
    CSV.foreach(path, headers: true, return_headers: false) do |row| # don't output the headers in the rows
      csv << row # append to the final file
    end
  end

pgwillia

source 'https://rubygems.org'

gem 'wraith'