23 consolidate input and output handling

Gemfile.lock

@@ -2,17 +2,40 @@ PATH
   remote: .
   specs:
     csv-util (0.6a.0)
+      activesupport (~> 5.0)
 
 GEM
   remote: https://rubygems.org/
   specs:
+    activesupport (5.2.8.1)
+      concurrent-ruby (~> 1.0, >= 1.0.2)
+      i18n (>= 0.7, < 2)
+      minitest (~> 5.1)
+      tzinfo (~> 1.1)
+    ast (2.4.2)
     coderay (1.1.3)
+    concurrent-ruby (1.3.3)
     diff-lcs (1.5.0)
+    i18n (1.14.5)
+      concurrent-ruby (~> 1.0)
+    json (2.7.2)
+    language_server-protocol (3.17.0.3)
     method_source (1.0.0)
+    minitest (5.24.0)
+    parallel (1.25.1)
+    parser (3.3.3.0)
+      ast (~> 2.4.1)
+      racc
     pry (0.14.2)
       coderay (~> 1.1)
       method_source (~> 1.0)
+    racc (1.8.0)
+    rack (3.1.4)
+    rainbow (3.1.1)
     rake (13.0.6)
+    regexp_parser (2.9.2)
+    rexml (3.3.0)
+      strscan
     rspec (3.12.0)
       rspec-core (~> 3.12.0)
       rspec-expectations (~> 3.12.0)
@@ -26,6 +49,46 @@ GEM
       diff-lcs (>= 1.2.0, < 2.0)
       rspec-support (~> 3.12.0)
     rspec-support (3.12.0)
+    rubocop (1.64.1)
+      json (~> 2.3)
+      language_server-protocol (>= 3.17.0)
+      parallel (~> 1.10)
+      parser (>= 3.3.0.2)
+      rainbow (>= 2.2.2, < 4.0)
+      regexp_parser (>= 1.8, < 3.0)
+      rexml (>= 3.2.5, < 4.0)
+      rubocop-ast (>= 1.31.1, < 2.0)
+      ruby-progressbar (~> 1.7)
+      unicode-display_width (>= 2.4.0, < 3.0)
+    rubocop-ast (1.31.3)
+      parser (>= 3.3.1.0)
+    rubocop-capybara (2.21.0)
+      rubocop (~> 1.41)
+    rubocop-performance (1.21.1)
+      rubocop (>= 1.48.1, < 2.0)
+      rubocop-ast (>= 1.31.1, < 2.0)
+    rubocop-rails (2.25.0)
+      activesupport (>= 4.2.0)
+      rack (>= 1.1)
+      rubocop (>= 1.33.0, < 2.0)
+      rubocop-ast (>= 1.31.1, < 2.0)
+    rubocop-rake (0.6.0)
+      rubocop (~> 1.0)
+    rubocop-rspec (3.0.1)
+      rubocop (~> 1.61)
+    ruby-progressbar (1.13.0)
+    strscan (3.1.0)
+    thread_safe (0.3.6)
+    tzinfo (1.2.11)
+      thread_safe (~> 0.1)
+    unicode-display_width (2.5.0)
+    upennlib-rubocop (1.2.0)
+      rubocop (~> 1.24)
+      rubocop-capybara
+      rubocop-performance
+      rubocop-rails
+      rubocop-rake
+      rubocop-rspec
 
 PLATFORMS
   arm64-darwin-21
@@ -36,6 +99,7 @@ DEPENDENCIES
   pry (~> 0.14.1)
   rake (~> 13.0)
   rspec (~> 3.0)
+  upennlib-rubocop (~> 1.0)
 
 BUNDLED WITH
    2.4.3

csv-util.gemspec

@@ -1,6 +1,6 @@
 # frozen_string_literal: true
 
-require_relative 'lib/csvutil/version'
+require_relative 'lib/csv_util/version'
 
 Gem::Specification.new do |spec|
   spec.name = 'csv-util'
@@ -30,11 +30,13 @@ Gem::Specification.new do |spec|
   spec.bindir = 'exe'
   spec.executables = spec.files.grep(%r{\Aexe/}) { |f| File.basename(f) }
   spec.require_paths = ['lib']
+  spec.add_runtime_dependency 'activesupport', '~> 5.0'
 
   # Uncomment to register a new dependency of your gem
   spec.add_development_dependency 'pry', '~> 0.14.1'
   spec.add_development_dependency "rake", "~> 13.0"
   spec.add_development_dependency "rspec", "~> 3.0"
+  spec.add_development_dependency 'upennlib-rubocop', '~> 1.0'
 
   # For more information and examples about making a new gem, check out our
   # guide at: https://bundler.io/guides/creating_gem.html

exe/csv-cut

@@ -34,7 +34,7 @@ end
 ###
 
 options = {}
-options[:separator] = ','
+options[:out_col_sep] = ','
 
 OptionParser.new do |opts|
   opts.banner = "Usage: #{File.basename(__FILE__)} -c col1,col2,col3 CSV_FILE "
@@ -46,8 +46,12 @@ OptionParser.new do |opts|
     exit
   end
 
-  opts.on '-s', '--separator SEP', 'Separator for mulitple columns' do |sep|
-    options[:separator] = sep
+  opts.on '-s', '--output-sep SEPARATOR', 'Output column separator' do |sep|
+    options[:out_col_sep] = sep
+  end
+
+  opts.on '-x', '--input-sep SEPARATOR', 'CSVReader column separator' do |sep|
+    options[:in_col_sep] = sep
   end
 
   opts.on '-l' , '--list-headers', 'List the headers in CSV and quit' do
@@ -58,8 +62,8 @@ OptionParser.new do |opts|
     options[:columns] = columns.split ','
   end
 
-  opts.on '-d', '--headers', 'Print column headers' do
-    options[:headers] = true
+  opts.on '-d', '--headers', 'Output the cut column headers' do
+    options[:output_headers] = true
   end
 
   opts.on '-h', '--help', 'Prints this help' do
@@ -74,25 +78,12 @@ end.parse!
 # See if there are any arguments left; otherwise, read from stdin
 if $stdin.tty? || ARGV.any?
   csv_file = ARGV.shift
-  abort "Please provide a CSV file" unless csv_file
+  abort 'Please provide a CSV file' unless csv_file
   abort "Can't find CSV file" unless File.exist? csv_file
   input = File.open csv_file
 else
   input = $stdin
 end
 
-
-###
-# Process the data
-###
-first_row = true
-CSV col_sep: options[:separator] do |csv|
-  CSV.parse input, headers: true do |row|
-    if first_row
-      process_first_row row: row, options: options, csv: csv
-      first_row = false
-    end
-
-    csv << options[:columns].map { |column| row[column] }
-  end
-end
+csv_cut = CSVUtil::Cut.new options[:columns], options: options
+csv_cut.process input

exe/csv-filter

@@ -12,6 +12,8 @@ CMD = File.basename __FILE__
 # Methods
 ########################################################################
 
+##
+# Print the version of the command to stdout
 def print_version
   puts "#{CMD} #{CSVUtil::VERSION}"
   exit
@@ -24,40 +26,6 @@ def valid_match? options
   options[:pattern].nil? ^ options[:text].nil?
 end
 
-def matches_text? value, options
-  normal_val = options[:insensitive] ? value.downcase : value
-  test_text  = options[:insensitive] ? options[:text].to_s.downcase : options[:text].to_s
-
-  return normal_val != test_text if options[:reject]
-
-  normal_val == test_text
-end
-
-def matches_pattern? value, options
-  # Regexp options are bit values (0,1,2,4); 0 => No options
-  # https://ruby-doc.org/3.2.2/Regexp.html#class-Regexp-label-Options
-  regexp_opts = 0
-  regexp_opts |= Regexp::IGNORECASE if options[:insensitive] # bitwise or/=
-
-  regex = Regexp.new options[:pattern], regexp_opts
-
-  return value !~ regex if options[:reject]
-
-  value =~ regex
-end
-
-def match? row, options
-  value = row[options[:column]].to_s
-
-  return matches_text? value, options if options[:text]
-
-  matches_pattern? value, options
-end
-
-def process_first_row row:, options:, csv:
-  csv << row.to_h.keys
-end
-
 ########################################################################
 # Options
 ########################################################################
@@ -78,38 +46,42 @@ EOF
     options[:version] = true
   end
 
-  opts.on "-c", "--column COLUMN_NAME", "Column to filter on" do |col|
+  opts.on '-c', '--column COLUMN_NAME', 'Column to filter on' do |col|
     options[:column] = col
   end
 
-  opts.on("-r", "--regex PATTERN",
-          "Ruby regular expression; e.g., 'trade.*cards'") do |regex|
+  opts.on('-r', '--regex PATTERN',
+          'Ruby regular expression pattern; e.g., \'trad.*cards\'') do |regex|
     options[:pattern] = regex
   end
 
-  opts.on("-j", "--reject", "Reject all matching rows") do
-    options[:reject] = true
+  opts.on('-j', '--reject', 'Reject all matching rows') do
+    options[:reject_matching] = true
   end
 
-  opts.on "-i", "--case-insensitive", "Whether regex is case-insensitive" do
+  opts.on '-i', '--case-insensitive', 'Whether regex is case-insensitive' do
     options[:insensitive] = true
   end
 
-  opts.on "-t", "--text TEXT", "An exact string to match" do |text|
+  opts.on '-t', '--text TEXT', 'An exact string to match' do |text|
     options[:text] = text
   end
 
-  opts.on "-e", "--encoding ENCODING", "Input file encoding (if not ASCII or UTF-8)" do |encoding|
+  opts.on '-u', '--substring', 'Output rows containing text' do
+    options[:substring] = true
+  end
+
+  opts.on '-e', '--encoding ENCODING', 'Input file encoding (if not ASCII or UTF-8)' do |encoding|
     options[:encoding] = encoding
   end
 
   c_help = %Q{List encodings and quit (#{Encoding.list.size} encodings)}
-  opts.on "--list-encodings", c_help do
+  opts.on '--list-encodings', c_help do
     puts Encoding.list
     exit
   end
 
-  opts.on("-h", "--help", "Print this help") do
+  opts.on('-h', '--help', 'Print this help') do
     puts opts
     exit
   end
@@ -124,26 +96,16 @@ print_version if options[:version]
 # See if there are any arguments left; otherwise, read from stdin
 if $stdin.tty? || ARGV.any?
   csv_file = ARGV.shift
-  abort "Please provide a CSV file" unless csv_file
+  abort 'Please provide a CSV file' unless csv_file
   abort "Can't find CSV file" unless File.exist? csv_file
   input = File.open csv_file
 else
   input = $stdin
 end
 
 # Handle options
-encoding = options[:encoding] || DEFAULT_ENCODING
 abort 'Provide either a regex or text to match' unless valid_match? options
 abort 'Specify a column to match' unless options[:column]
 
-
-first_row = true
-CSV headers: true do |csv|
-  CSV.parse input, headers: true, encoding: "#{encoding}:utf-8" do |row|
-    if first_row
-      process_first_row row: row, options: options, csv: csv
-      first_row = false
-    end
-    csv << row if match? row, options
-  end
-end
+filter = CSVUtil::Filter.new options[:column], **options
+filter.filter input