csv-utils 0.3.21 → 0.3.23

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: bea28dcd86e140064eba5a1134d5aa6193963ddcb7632c63c2e1bfa85c4889ed
-  data.tar.gz: 1547e2cb2b937c580031205baa64c79c3c0ca062ee2aeffd578b897d10fae092
+  metadata.gz: 2241c0a8d13e6edcb21271f66523862506cb1a04fbcfd533c160bf307c3ee6eb
+  data.tar.gz: 6d2f393c77e232aa8c124a43290f6ba912f42c0d63a49c55945c159106eb8169
 SHA512:
-  metadata.gz: fe635fdc22220a0377ca386132109bebce61ba5fd41ec71625b8940b4dd7e2e5c72efb8cf163fc44f990633e18b8024b14878cbe0e852b49d4ba31fb076e44b3
-  data.tar.gz: 569285d118ead894808ca96b5e4d1c0c1dd5f4bd130452a79c1ffb4f987ede9a0cf324609f4d11a15bddf61169948799026dd2bd4836674a1aae2f2caa137063
+  metadata.gz: d988950900011db0d030d7e5534dcf1cfa0940d17ea3257de18da5fb3fdc5745dbd797007ab220a0a72c4fd5b9ed907100d9244fe30c11484de5dbf3d58c5d1c
+  data.tar.gz: 64698435d62a5bd8e0bbd3e9a480581161d74c1d14bf2e7686a39b72898e765b77aeea0d95e8a8c98c065ab9af0f5f819fb1d58d6cd453a16f95c8b24f49e1ec

data/bin/csv-splitter

@@ -0,0 +1,81 @@
+#!/usr/bin/env ruby
+
+require 'optparse'
+
+options = {
+  rows: 250_000,
+  no_header: false
+}
+OptionParser.new do |opts|
+  opts.banner = 'Usage: ' + File.basename(__FILE__) + ' [options] <csv file>'
+
+  opts.on('-h', '--help', 'Prints this help') do
+    puts opts
+    exit
+  end
+
+  opts.on('-r', '--rows ROWS', Integer, 'Max number of rows per a CSV') do |v|
+    options[:rows] = v
+  end
+
+  opts.on('--no-header', 'CSV file has no header') do
+    options[:no_header] = true
+  end
+end.parse!
+
+file = ARGV[0] || raise('no CSV file specified')
+
+require 'csv-utils'
+
+def get_split_file_name(file, num_files, total_files)
+  if file =~ /\.(?:csv|tsv)$/i
+    file.sub(/\.(?:csv|tsv)$/i) { |m| ".part-#{num_files}-of-#{total_files}" + m }
+  else
+    file + ".part-#{num_files}"
+  end
+end
+
+csv_options = CSVUtils::CSVOptions.new(file)
+
+csv = CSV.open(
+  file,
+  'rb',
+  col_sep: csv_options.col_separator,
+  liberal_parsing: true
+)
+
+headers = options[:no_header] ? nil : csv.shift
+
+total_rows = 0
+while csv.shift
+  total_rows +=1
+end
+
+csv.rewind
+csv.shift if headers
+
+total_files = total_rows / options[:rows]
+total_files += 1 if (total_rows % options[:rows]) > 0
+
+num_rows = 0
+num_files = 0
+out = nil
+
+append_row_proc = proc do |row|
+  if out.nil? || num_rows >= options[:rows]
+    out.close if out
+    num_rows = 0
+    num_files += 1
+    out = CSV.open(get_split_file_name(file, num_files, total_files), 'wb', col_sep: csv_options.col_separator)
+    out << headers if headers
+  end
+
+  out << row
+  num_rows += 1
+end
+
+while (row = csv.shift)
+  append_row_proc.call(row)
+end
+
+out.close

data/csv-utils.gemspec

@@ -2,7 +2,7 @@
 
 Gem::Specification.new do |s|
   s.name        = 'csv-utils'
-  s.version     = '0.3.21'
+  s.version     = '0.3.23'
   s.licenses    = ['MIT']
   s.summary     = 'CSV Utils'
   s.description = 'Tools for debugging malformed CSV files'

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: csv-utils
 version: !ruby/object:Gem::Version
-  version: 0.3.21
+  version: 0.3.23
 platform: ruby
 authors:
 - Doug Youch
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2022-11-14 00:00:00.000000000 Z
+date: 2023-05-02 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: inheritance-helper
@@ -34,6 +34,7 @@ executables:
 - csv-find-error
 - csv-grep
 - csv-readline
+- csv-splitter
 - csv-validator
 extensions: []
 extra_rdoc_files: []
@@ -52,6 +53,7 @@ files:
 - bin/csv-find-error
 - bin/csv-grep
 - bin/csv-readline
+- bin/csv-splitter
 - bin/csv-validator
 - csv-utils.gemspec
 - lib/csv-utils.rb