diff options
Diffstat (limited to 'app/services/packages/npm/create_package_service.rb')
| -rw-r--r-- | app/services/packages/npm/create_package_service.rb | 53 |
1 files changed, 23 insertions, 30 deletions
diff --git a/app/services/packages/npm/create_package_service.rb b/app/services/packages/npm/create_package_service.rb index c71ae060dd9..2c578760cc5 100644 --- a/app/services/packages/npm/create_package_service.rb +++ b/app/services/packages/npm/create_package_service.rb @@ -61,10 +61,9 @@ module Packages end def version - strong_memoize(:version) do - params[:versions].each_key.first - end + params[:versions].each_key.first end + strong_memoize_attr :version def version_data params[:versions][version] @@ -79,30 +78,27 @@ module Packages end def package_file_name - strong_memoize(:package_file_name) do - "#{name}-#{version}.tgz" - end + "#{name}-#{version}.tgz" end + strong_memoize_attr :package_file_name def attachment - strong_memoize(:attachment) do - params['_attachments'][package_file_name] - end + params['_attachments'][package_file_name] end + strong_memoize_attr :attachment # TODO (technical debt): Extract the package size calculation to its own component and unit test it separately. def calculated_package_file_size - strong_memoize(:calculated_package_file_size) do - # This calculation is based on: - # 1. 4 chars in a Base64 encoded string are 3 bytes in the original string. Meaning 1 char is 0.75 bytes. - # 2. The encoded string may have 1 or 2 extra '=' chars used for padding. Each padding char means 1 byte less in the original string. - # Reference: - # - https://blog.aaronlenoir.com/2017/11/10/get-original-length-from-base-64-string/ - # - https://en.wikipedia.org/wiki/Base64#Decoding_Base64_with_padding - encoded_data = attachment['data'] - ((encoded_data.length * 0.75) - encoded_data[-2..].count('=')).to_i - end + # This calculation is based on: + # 1. 4 chars in a Base64 encoded string are 3 bytes in the original string. Meaning 1 char is 0.75 bytes. + # 2. The encoded string may have 1 or 2 extra '=' chars used for padding. Each padding char means 1 byte less in the original string. + # Reference: + # - https://blog.aaronlenoir.com/2017/11/10/get-original-length-from-base-64-string/ + # - https://en.wikipedia.org/wiki/Base64#Decoding_Base64_with_padding + encoded_data = attachment['data'] + ((encoded_data.length * 0.75) - encoded_data[-2..].count('=')).to_i end + strong_memoize_attr :calculated_package_file_size def file_params { @@ -134,29 +130,26 @@ module Packages end def field_sizes - strong_memoize(:field_sizes) do - package_json.transform_values do |value| - value.to_s.size - end + package_json.transform_values do |value| + value.to_s.size end end + strong_memoize_attr :field_sizes def filtered_field_sizes - strong_memoize(:filtered_field_sizes) do - field_sizes.select do |_, size| - size >= ::Packages::Npm::Metadatum::MIN_PACKAGE_JSON_FIELD_SIZE_FOR_ERROR_TRACKING - end + field_sizes.select do |_, size| + size >= ::Packages::Npm::Metadatum::MIN_PACKAGE_JSON_FIELD_SIZE_FOR_ERROR_TRACKING end end + strong_memoize_attr :filtered_field_sizes def largest_fields - strong_memoize(:largest_fields) do - field_sizes + field_sizes .sort_by { |a| a[1] } .reverse[0..::Packages::Npm::Metadatum::NUM_FIELDS_FOR_ERROR_TRACKING - 1] .to_h - end end + strong_memoize_attr :largest_fields def field_sizes_for_error_tracking filtered_field_sizes.empty? ? largest_fields : filtered_field_sizes |