csv-probe 0.1.0 → 0.1.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 138cd78aa31ccab66bbaa4a629c4f2c0d3b37a547006ed635e6fed935a223051
-  data.tar.gz: 87f700fcb9c7a620f101b5a63bf095b9f6a611f72d83f876b11ca33797564a72
+  metadata.gz: 2425404adecef9f68fced84863e0acb087f3358935032c3849cab690948e74dc
+  data.tar.gz: 0bf460f9f27439cdcba3150c2f53991ab07512e6e1ffb5195315fdad5689c549
 SHA512:
-  metadata.gz: 1357f59d7446f937c3719feeafd86aa29a10844203d72dd20f30c338243ad347adeca2c5d1dca93195cd223403fb024263330f398f029455f8ae5562fd906812
-  data.tar.gz: aee4fa829429c8cc4590351da2b303b5879a163b412d987ec7dff63f5c45e5e8e9e4b5dd40af4619a19ceaa166bf09670e3f4f0b50b79f83e4c048552ec0ec7c
+  metadata.gz: 721e33de3a86d0513dcaab1d0d102454fbcf48492dd6d551ec1f5623de6c0b7ba74189ea7d6ae8825eb34674b2679f251e50c08e87fabf1ae18a993cdc72eaff
+  data.tar.gz: 042b6b18255ddbb787c7318febbc31147e4e593cbedf67047bc25c962491a9a46b3e664841a18ea3b8bfb25ef4025455391a4c7571dbf1a37072bb75c64967f8

data/.rubocop.yml

@@ -1,5 +1,6 @@
 AllCops:
   TargetRubyVersion: 2.6
+  NewCops: enable
 
 Style/StringLiterals:
   Enabled: true

data/CHANGELOG.md

@@ -1,5 +1,20 @@
 ## [Unreleased]
 
+## [0.1.4] - 2022-01-05
+- Fix `nil` handling for list + set column checks when the list or set is something like this:
+  - `a,|b|,b` leading and trailing empty items or 
+  - `a,,b` the whole list/set string is nil
+
+## [0.1.3] - 2022-01-04
+- Add better `nil` handling to list + set column checks. `nil` resp. `""` does no longer implicitly pass ColumnIsListWithDomain and ColumnIsSetWithDomain checks. `nil` or `""` has to be in the expected items in order to pass the check now.
+
+## [0.1.2] - 2022-01-03
+- Fix README.md
+
+## [0.1.1] - 2022-01-02
+- Fix rubocop warnings
+- Improve README.md
+
 ## [0.1.0] - 2022-01-02
 
 - Initial release

data/README.md

@@ -23,7 +23,10 @@ Or install it yourself as:
 
 Example how to use Probe
 
-```
+```ruby
+  # load csv-probe gem
+  require 'csv/probe'
+
   # load CSV into CSV::Table
   csv_table = CSV.parse(<<~ROWS, headers: true)
     col1,col2,col3,col4,col5,col6,col7
@@ -31,13 +34,17 @@ Example how to use Probe
     0,PA,customer,2,03-12-2021,SHA;AN;RU,badger|ant
     0,TE,guest,1000,06-11-2021,RU,spider|racoon|ant
   ROWS
+
+  # define linting rules
   checks = [Probe::ColumnIsEqualTo.new("col1", "0"),
             Probe::ColumnMatchesRegEx.new("col2", /^(NU|PA|TE)$/),
-            Probe::ColumnIsOneOf.new("col3", ["customer","guest"]),
-            Probe::ColumnMeetsCondition.new("col4", ->(val, _opts) { Integer(val, exception: false) != nil }, "Not an Integer"),
+            Probe::ColumnIsOneOf.new("col3", %w[customer guest]),
+            Probe::ColumnMeetsCondition.new("col4", lambda { |val, _opts|
+              !Integer(val, exception: false).nil?
+            }, "Not an Integer"),
             Probe::ColumnIsDate.new("col5", "%d-%m-%Y"),
-            Probe::ColumnIsListWithDomain.new("col6", ["FO", "RU", "SHA", "AN"], ";"),
-            Probe::ColumnIsSetWithDomain.new("col7", ["eagle", "hawk", "badger", "spider", "racoon", "ant"], "|")]
+            Probe::ColumnIsListWithDomain.new("col6", %w[FO RU SHA AN], ";"),
+            Probe::ColumnIsSetWithDomain.new("col7", %w[eagle hawk badger spider racoon ant], "|")]
 
   # lint table
   csv_table.lint(checks)
@@ -54,7 +61,7 @@ To install this gem onto your local machine, run `bundle exec rake install`. To
 
 ## Contributing
 
-Bug reports and pull requests are welcome on GitHub at https://github.com/[USERNAME]/csv-probe.
+Bug reports and pull requests are welcome on GitHub at https://github.com/homebase-dev/csv-probe.
 
 ## License
 

data/csv-probe.gemspec

@@ -9,12 +9,14 @@ Gem::Specification.new do |spec|
   spec.email = ["homebase.dev@gmail.com"]
 
   spec.summary = "Allows validation of CSV::Table or CSV::Rows via custom rules"
-  spec.description = "This gem provides a simple framework for CSV::Table/CSV::Row validation using custom rules for columns and rows"
+  spec.description = "This gem provides a simple framework for CSV::Table/CSV::Row linting"\
+                     "using custom rules for columns and rows"
   spec.homepage = "https://gitlab.com/homebase-dev/csv-probe"
   spec.license = "MIT"
   spec.required_ruby_version = ">= 2.6.0"
 
   # spec.metadata["allowed_push_host"] = "TODO: Set to your gem server 'https://example.com'"
+  spec.metadata["rubygems_mfa_required"] = "true"
 
   spec.metadata["homepage_uri"] = spec.homepage
   spec.metadata["source_code_uri"] = "https://gitlab.com/homebase-dev/csv-probe"

data/lib/csv/probe/checks.rb

@@ -5,30 +5,35 @@ module Probe
 
   class LintingError < Error; end
 
+  # Linting error in CSV::Row, if the rule applies to multiple columns of a row
   class RowError < LintingError
     def initialize(msg = "CheckRowError")
       super
     end
   end
 
+  # Linting warning in CSV::Row, if the rule applies to multiple columns of a row
   class RowWarning < LintingError
     def initialize(msg = "CheckRowWarning")
       super
     end
   end
 
+  # Linting error in CSV::Column, if the rule applies to a single column of a row
   class ColumnError < LintingError
     def initialize(msg = "CheckColumnError")
       super
     end
   end
 
+  # Linting warning in CSV::Column, if the rule applies to a single column of a row
   class ColumnWarning < LintingError
     def initialize(msg = "CheckColumnWarning")
       super
     end
   end
 
+  # Abstract class of a CSV::Row check
   class RowMeetsCondition
     attr_accessor :ok_condition_fn, :fail_msg, :severity, :pre_checks
 
@@ -102,6 +107,7 @@ module Probe
     end
   end
 
+  # Abstract class of a CSV::Row single field (= column) check
   class ColumnMeetsCondition < RowMeetsCondition
     attr_accessor :varname, :ok_condition_fn, :fail_msg
 
@@ -120,10 +126,14 @@ module Probe
 
     def evaluate(row, opts = {})
       evaluate_pre_checks(row, opts)
-      raise @error_classes.fetch(@severity), error_msg(row, opts) unless @ok_condition_fn.call(row.fetch(@varname), opts)
+
+      unless @ok_condition_fn.call(row.fetch(@varname), opts) # rubocop:disable Style/GuardClause
+        raise @error_classes.fetch(@severity), error_msg(row, opts)
+      end
     end
   end
 
+  # Check if a column value is equal to a provided value (Pay attention to types!)
   class ColumnIsEqualTo < ColumnMeetsCondition
     def initialize(varname, expected_val, _placeholder = nil)
       super(varname, nil, nil)
@@ -132,6 +142,7 @@ module Probe
     end
   end
 
+  # Check if a column value is one of a list of provided value
   class ColumnIsOneOf < ColumnMeetsCondition
     def initialize(varname, expected_vals_arr, _placeholder = nil)
       super(varname, nil, nil)
@@ -140,8 +151,9 @@ module Probe
     end
   end
 
+  # Check if a column value is a date with a given format
   class ColumnIsDate < ColumnMeetsCondition
-    def initialize(varname, expected_date_format, _placeholder = nil)
+    def initialize(varname, expected_date_format, _placeholder = nil) # rubocop:disable Metrics/MethodLength
       super(varname, nil, nil)
       @ok_condition_fn = lambda { |val, _cfg|
         success = true
@@ -159,11 +171,13 @@ module Probe
   # class ColumnIsInteger < ColumnMeetsCondition
   #   def initialize(varname, placeholder1=nil, placeholder=nil)
   #     super(varname, nil, nil)
-  #     @ok_condition_fn = ->(val, cfg){ val.is_a? Integer } # TODO: achtung 12abc -> integer auch 1.1 wird zu 1... ist nicht so einfach, lieber über regex prüfen
+  #     @ok_condition_fn = ->(val, cfg){ val.is_a? Integer }
+  #       # TODO: ATTENTION 12abc -> integer, also  1.1 -> 1 ! better do it over regex?
   #     @fail_msg = "expected to be an Integer"
   #   end
   # end
 
+  # Check if a column value is a date with a given format
   class ColumnMatchesRegEx < ColumnMeetsCondition
     def initialize(varname, expected_regex_pattern, _placeholder = nil)
       super(varname, nil, nil)
@@ -172,56 +186,72 @@ module Probe
     end
   end
 
-  # TODO: fixme uniq als eigenen check machen
+  # Check if a tokenized column value is a list of given values
   class ColumnIsListWithDomain < ColumnMeetsCondition
-    def initialize(varname, expected_items_arr, separator, _placeholder = nil)
+    def initialize(varname, expected_items_arr, separator, _placeholder = nil) # rubocop:disable Metrics/MethodLength
       super(varname, nil, nil)
       @ok_condition_fn = lambda { |val, _cfg|
-        items = val.split(separator)
+        expected_items_arr.map!(&:to_s) # turn nil -> ""
+
+        items = val.to_s.split(separator, -1)
+        return true if items.empty? && expected_items_arr.include?(nil.to_s) # empty str allowed
+
+        return false if items.empty?
+
         return items.all? { |e| expected_items_arr.include?(e) }
       }
       @fail_msg = lambda { |row, _opts|
-        items = row.fetch(@varname).split(separator)
+        items = row.fetch(@varname).to_s.split(separator, -1)
         diff_items = items - expected_items_arr
-        "expected that tokenized items of value #{items.inspect} are a subset of #{expected_items_arr.inspect}, but items #{diff_items.inspect} are not"
+        "expected that tokenized items of value #{items.inspect} are a subset of "\
+          "#{expected_items_arr.inspect}, but items #{diff_items.inspect} are not"
       }
     end
   end
 
+  # Check if a tokenized column value is a list of given values
   class ColumnIsSet < ColumnMeetsCondition
-    def initialize(varname, separator, _placeholder = nil)
+    def initialize(varname, separator, _placeholder = nil) # rubocop:disable Metrics/MethodLength
       super(varname, nil, nil)
       @ok_condition_fn = lambda { |val, _cfg|
-        return true if val.to_s == ""
+        return true if val.to_s == ''
 
-        items = val.split(separator)
+        items = val.to_s.split(separator, -1)
         all_uniq = (items.size == items.uniq.size)
         return all_uniq
       }
       @fail_msg = lambda { |row, _opts|
-        items = row.fetch(@varname).split(separator)
+        items = row.fetch(@varname).to_s.split(separator, -1)
         non_uniq_items = items.detect { |e| items.count(e) > 1 }
         "expected that items of tokenized value #{items.inspect} are uniqe, but items #{non_uniq_items.inspect} are not"
       }
     end
   end
 
+  # Check if a tokenized column value is a set of given values (no duplicates)
   class ColumnIsSetWithDomain < ColumnMeetsCondition
-    def initialize(varname, expected_items_arr, separator, _placeholder = nil)
+    def initialize(varname, expected_items_arr, separator, _placeholder = nil) # rubocop:disable Metrics/MethodLength
       super(varname, nil, nil)
       @pre_checks << ColumnIsSet.new(varname, separator)
       @ok_condition_fn = lambda { |val, _cfg|
-        return true if val.to_s == ""
+        expected_items_arr.map!(&:to_s) # turn nil -> ""
+
+        items = val.to_s.split(separator, -1)
+        return true if items.empty? && expected_items_arr.include?(nil.to_s) # empty str allowed
+
+        return false if items.empty?
 
-        items = val.split(separator)
         all_valid = items.all? { |i| expected_items_arr.include?(i) }
         return all_valid
       }
       @fail_msg = lambda { |row, _opts|
-        items = row.fetch(@varname).split(separator)
+        items = row.fetch(@varname).to_s.split(separator, -1)
         "expected that items of tokenized value #{items.inspect} are a subset of #{expected_items_arr.inspect}"
       }
-      # @fail_msg = ->(row, opts) { "Unexpected value:#{row.fetch(@varname).inspect} for column:#{@varname.inspect}, expected that items of tokenized value #{row.fetch(@varname).split(separator)} are uniqe and a subset of #{expected_items_arr.inspect}" }
+      # @fail_msg = lambda {|row, opts|
+      # "Unexpected value:#{row.fetch(@varname).inspect} for column:#{@varname.inspect},
+      #  expected that items of tokenized value #{row.fetch(@varname).split(separator)}
+      #  are uniqe and a subset of #{expected_items_arr.inspect}" }
     end
   end
 

data/lib/csv/probe/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Probe
-  VERSION = "0.1.0"
+  VERSION = "0.1.4"
 end

data/lib/csv/probe.rb

@@ -1,6 +1,6 @@
 # frozen_string_literal: true
 
-require "csv"
+require "csv" # TODO: should i require it here?
 require "terminal-table"
 require_relative "probe/version"
 require_relative "probe/checks"
@@ -43,7 +43,7 @@ module Probe
 end
 
 # Extend CSV::Table with .lint(...) method
-class CSV::Table
+class CSV::Table # rubocop:disable Style/ClassAndModuleChildren
   def lint(checks, opts = {})
     opts[:headers] = true unless opts.key?(:headers) # CSV::Table has always headers, hence set :headers = true
     Probe.lint_rows(self, checks, opts)
@@ -51,7 +51,7 @@ class CSV::Table
 end
 
 # Extend CSV::Row with .lint(...) method
-class CSV::Row
+class CSV::Row # rubocop:disable Style/ClassAndModuleChildren
   def lint(checks, opts = {})
     Probe.lint_row(self, checks, opts)
   end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: csv-probe
 version: !ruby/object:Gem::Version
-  version: 0.1.0
+  version: 0.1.4
 platform: ruby
 authors:
 - homebase.dev
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2022-01-02 00:00:00.000000000 Z
+date: 2022-01-05 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: csv
@@ -38,8 +38,8 @@ dependencies:
     - - "~>"
       - !ruby/object:Gem::Version
         version: '3.0'
-description: This gem provides a simple framework for CSV::Table/CSV::Row validation
-  using custom rules for columns and rows
+description: This gem provides a simple framework for CSV::Table/CSV::Row lintingusing
+  custom rules for columns and rows
 email:
 - homebase.dev@gmail.com
 executables: []
@@ -64,6 +64,7 @@ homepage: https://gitlab.com/homebase-dev/csv-probe
 licenses:
 - MIT
 metadata:
+  rubygems_mfa_required: 'true'
   homepage_uri: https://gitlab.com/homebase-dev/csv-probe
   source_code_uri: https://gitlab.com/homebase-dev/csv-probe
   changelog_uri: https://gitlab.com/homebase-dev/csv-probe