csv-probe 0.1.1 → 0.1.5

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 83d63f020bf4ed1e557518b78820b440cc24da36a7d93f6ded23b604aa7161ed
-  data.tar.gz: df4e094f0f0451c9578ac5f02c34eb967d851a4782926dc64f0b96bdcd1eb9f8
+  metadata.gz: 945175b169f28d0b77504d23cf6020d5216e66f74e8e5a8b04ef0279f681fd8c
+  data.tar.gz: a91cb9d894402b70ebfb72a342f4166a94bda0b3c6f1ba361920d0caa6b3c0c9
 SHA512:
-  metadata.gz: 9b3be13578b537577b62fd1efa51ec00f2d0a5a38755b2e8399dfeacf9f3ceaa1f30a65381336af5464cb4d08c32797bb0de5199f34f4e437041db9ec70ab681
-  data.tar.gz: 2eb6bfdeb32a7d3f819cf52b412b311984d7e45366426ed81a788ccf2f9a83504733cbe348c8da34421f23855eab2f8e5efe40e3fd4086a94a656758cdef3e4b
+  metadata.gz: 7d09e366f9e7b4692dbfb1f8389667952f5e3a913901600d77d9954a9b5685650d81b6b9bba701cfa2925c47cefb0aca2e4b4bd5d48a3547b7cd2c775e512e01
+  data.tar.gz: a076a71e6679f69848b7c30117a100986e3857eb41a3ad2bed50c0a6b320add76caf6668ae8790dce9804c82e2f7dcc5b0e0e8675972c8f81a594bc163501f73

data/CHANGELOG.md

@@ -1,5 +1,23 @@
 ## [Unreleased]
 
+## [0.1.5] - 2022-01-05
+- Fix rubocop warnings
+
+## [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

@@ -24,6 +24,9 @@ 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)

data/lib/csv/probe/checks.rb

@@ -186,16 +186,23 @@ module Probe
     end
   end
 
+  # rubocop:disable Metrics/AbcSize
   # Check if a tokenized column value is a list of given values
   class ColumnIsListWithDomain < ColumnMeetsCondition
     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"
@@ -210,12 +217,12 @@ module Probe
       @ok_condition_fn = lambda { |val, _cfg|
         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"
       }
@@ -228,14 +235,18 @@ module Probe
       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 = lambda {|row, opts|
@@ -244,6 +255,7 @@ module Probe
       #  are uniqe and a subset of #{expected_items_arr.inspect}" }
     end
   end
+  # rubocop:enable Metrics/AbcSize
 
   # TODO: nice to have
   # * if_second_not_empty_then_first_yes_check(varname1, varname2)

data/lib/csv/probe/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Probe
-  VERSION = "0.1.1"
+  VERSION = "0.1.5"
 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"

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: csv-probe
 version: !ruby/object:Gem::Version
-  version: 0.1.1
+  version: 0.1.5
 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