Module: SmarterCSV
- Defined in:
- lib/smarter_csv/version.rb,
lib/smarter_csv/smarter_csv.rb
Defined Under Namespace
Classes: DuplicateHeaders, HeaderSizeMismatch, IncorrectOption, MissingHeaders, SmarterCSVException
Constant Summary collapse
- VERSION =
"1.2.8"
Class Method Summary collapse
-
.process(input, options = {}, &block) ⇒ Object
first parameter: filename or input object with readline method.
Class Method Details
.process(input, options = {}, &block) ⇒ Object
first parameter: filename or input object with readline method
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 |
# File 'lib/smarter_csv/smarter_csv.rb', line 9 def SmarterCSV.process(input, ={}, &block) # first parameter: filename or input object with readline method = {:col_sep => ',' , :row_sep => $/ , :quote_char => '"', :force_simple_split => false , :verbose => false , :remove_empty_values => true, :remove_zero_values => false , :remove_values_matching => nil , :remove_empty_hashes => true , :strip_whitespace => true, :convert_values_to_numeric => true, :strip_chars_from_headers => nil , :user_provided_headers => nil , :headers_in_file => true, :comment_regexp => /\A#/, :chunk_size => nil , :key_mapping_hash => nil , :downcase_header => true, :strings_as_keys => false, :file_encoding => 'utf-8', :remove_unmapped_keys => false, :keep_original_headers => false, :value_converters => nil, :skip_lines => nil, :force_utf8 => false, :invalid_byte_sequence => '', :auto_row_sep_chars => 500, :required_headers => nil } = .merge() [:invalid_byte_sequence] = '' if [:invalid_byte_sequence].nil? = .select{|k,v| [:col_sep, :row_sep, :quote_char].include?(k)} # options.slice(:col_sep, :row_sep, :quote_char) headerA = [] result = [] old_row_sep = $/ file_line_count = 0 csv_line_count = 0 has_rails = !! defined?(Rails) begin f = input.respond_to?(:readline) ? input : File.open(input, "r:#{[:file_encoding]}") if ([:force_utf8] || [:file_encoding] =~ /utf-8/i) && ( f.respond_to?(:external_encoding) && f.external_encoding != Encoding.find('UTF-8') || f.respond_to?(:encoding) && f.encoding != Encoding.find('UTF-8') ) puts 'WARNING: you are trying to process UTF-8 input, but did not open the input with "b:utf-8" option. See README file "NOTES about File Encodings".' end if [:row_sep] == :auto [:row_sep] = line_ending = SmarterCSV.guess_line_ending( f, ) f.rewind end $/ = [:row_sep] if [:skip_lines].to_i > 0 [:skip_lines].to_i.times{f.readline} end if [:headers_in_file] # extract the header line # process the header line in the CSV file.. # the first line of a CSV file contains the header .. it might be commented out, so we need to read it anyhow header = f.readline header = header.force_encoding('utf-8').encode('utf-8', invalid: :replace, undef: :replace, replace: [:invalid_byte_sequence]) if [:force_utf8] || [:file_encoding] !~ /utf-8/i header = header.sub([:comment_regexp],'').chomp([:row_sep]) file_line_count += 1 csv_line_count += 1 header = header.gsub([:strip_chars_from_headers], '') if [:strip_chars_from_headers] if (header =~ %r{#{[:quote_char]}}) and (! [:force_simple_split]) file_headerA = begin CSV.parse( header, ** ).flatten.collect!{|x| x.nil? ? '' : x} # to deal with nil values from CSV.parse rescue CSV::MalformedCSVError => e raise $!, "#{$!} [SmarterCSV: csv line #{csv_line_count}]", $!.backtrace end else file_headerA = header.split([:col_sep]) end file_headerA.map!{|x| x.gsub(%r/#{[:quote_char]}/,'') } file_headerA.map!{|x| x.strip} if [:strip_whitespace] unless [:keep_original_headers] file_headerA.map!{|x| x.gsub(/\s+|-+/,'_')} file_headerA.map!{|x| x.downcase } if [:downcase_header] end file_header_size = file_headerA.size else raise SmarterCSV::IncorrectOption , "ERROR: If :headers_in_file is set to false, you have to provide :user_provided_headers" if [:user_provided_headers].nil? end if [:user_provided_headers] && [:user_provided_headers].class == Array && ! [:user_provided_headers].empty? # use user-provided headers headerA = [:user_provided_headers] if defined?(file_header_size) && ! file_header_size.nil? if headerA.size != file_header_size raise SmarterCSV::HeaderSizeMismatch , "ERROR: :user_provided_headers defines #{headerA.size} headers != CSV-file #{input} has #{file_header_size} headers" else # we could print out the mapping of file_headerA to headerA here end end else headerA = file_headerA end headerA.map!{|x| x.to_sym } unless [:strings_as_keys] || [:keep_original_headers] unless [:user_provided_headers] # wouldn't make sense to re-map user provided headers key_mappingH = [:key_mapping] # do some key mapping on the keys in the file header # if you want to completely delete a key, then map it to nil or to '' if ! key_mappingH.nil? && key_mappingH.class == Hash && key_mappingH.keys.size > 0 headerA.map!{|x| key_mappingH.has_key?(x) ? (key_mappingH[x].nil? ? nil : key_mappingH[x].to_sym) : ([:remove_unmapped_keys] ? nil : x)} end end # header_validations duplicate_headers = [] headerA.compact.each do |k| duplicate_headers << k if headerA.select{|x| x == k}.size > 1 end raise SmarterCSV::DuplicateHeaders , "ERROR: duplicate headers: #{duplicate_headers.join(',')}" unless duplicate_headers.empty? if [:required_headers] && [:required_headers].is_a?(Array) missing_headers = [] [:required_headers].each do |k| missing_headers << k unless headerA.include?(k) end raise SmarterCSV::MissingHeaders , "ERROR: missing headers: #{missing_headers.join(',')}" unless missing_headers.empty? end # in case we use chunking.. we'll need to set it up.. if ! [:chunk_size].nil? && [:chunk_size].to_i > 0 use_chunks = true chunk_size = [:chunk_size].to_i chunk_count = 0 chunk = [] else use_chunks = false end # now on to processing all the rest of the lines in the CSV file: while ! f.eof? # we can't use f.readlines() here, because this would read the whole file into memory at once, and eof => true line = f.readline # read one line.. this uses the input_record_separator $/ which we set previously! # replace invalid byte sequence in UTF-8 with question mark to avoid errors line = line.force_encoding('utf-8').encode('utf-8', invalid: :replace, undef: :replace, replace: [:invalid_byte_sequence]) if [:force_utf8] || [:file_encoding] !~ /utf-8/i file_line_count += 1 csv_line_count += 1 print "processing file line %10d, csv line %10d\r" % [file_line_count, csv_line_count] if [:verbose] next if line =~ [:comment_regexp] # ignore all comment lines if there are any # cater for the quoted csv data containing the row separator carriage return character # in which case the row data will be split across multiple lines (see the sample content in spec/fixtures/carriage_returns_rn.csv) # by detecting the existence of an uneven number of quote characters multiline = line.count([:quote_char])%2 == 1 while line.count([:quote_char])%2 == 1 next_line = f.readline next_line = next_line.force_encoding('utf-8').encode('utf-8', invalid: :replace, undef: :replace, replace: [:invalid_byte_sequence]) if [:force_utf8] || [:file_encoding] !~ /utf-8/i line += next_line file_line_count += 1 end print "\nline contains uneven number of quote chars so including content through file line %d\n" % file_line_count if [:verbose] && multiline line.chomp! # will use $/ which is set to options[:col_sep] if (line =~ %r{#{[:quote_char]}}) and (! [:force_simple_split]) dataA = begin CSV.parse( line, ** ).flatten.collect!{|x| x.nil? ? '' : x} # to deal with nil values from CSV.parse rescue CSV::MalformedCSVError => e raise $!, "#{$!} [SmarterCSV: csv line #{csv_line_count}]", $!.backtrace end else dataA = line.split([:col_sep]) end #### dataA.map!{|x| x.gsub(%r/#{options[:quote_char]}/,'') } # this is actually not a good idea as a default dataA.map!{|x| x.strip} if [:strip_whitespace] hash = Hash.zip(headerA,dataA) # from Facets of Ruby library # make sure we delete any key/value pairs from the hash, which the user wanted to delete: # Note: Ruby < 1.9 doesn't allow empty symbol literals! hash.delete(nil); hash.delete(''); if RUBY_VERSION.to_f > 1.8 eval('hash.delete(:"")') end # remove empty values using the same regexp as used by the rails blank? method # which caters for double \n and \r\n characters such as "1\r\n\r\n2" whereas the original check (v =~ /^\s*$/) does not if [:remove_empty_values] if has_rails hash.delete_if{|k,v| v.blank?} else hash.delete_if{|k,v| v.nil? || v !~ /[^[:space:]]/} end end hash.delete_if{|k,v| ! v.nil? && v =~ /^(\d+|\d+\.\d+)$/ && v.to_f == 0} if [:remove_zero_values] # values are typically Strings! hash.delete_if{|k,v| v =~ [:remove_values_matching]} if [:remove_values_matching] if [:convert_values_to_numeric] hash.each do |k,v| # deal with the :only / :except options to :convert_values_to_numeric next if SmarterCSV.only_or_except_limit_execution( , :convert_values_to_numeric , k ) # convert if it's a numeric value: case v when /^[+-]?\d+\.\d+$/ hash[k] = v.to_f when /^[+-]?\d+$/ hash[k] = v.to_i end end end if [:value_converters] hash.each do |k,v| converter = [:value_converters][k] next unless converter hash[k] = converter.convert(v) end end next if hash.empty? if [:remove_empty_hashes] if use_chunks chunk << hash # append temp result to chunk if chunk.size >= chunk_size || f.eof? # if chunk if full, or EOF reached # do something with the chunk if block_given? yield chunk # do something with the hashes in the chunk in the block else result << chunk # not sure yet, why anybody would want to do this without a block end chunk_count += 1 chunk = [] # initialize for next chunk of data else # the last chunk may contain partial data, which also needs to be returned (BUG / ISSUE-18) end # while a chunk is being filled up we don't need to do anything else here else # no chunk handling if block_given? yield [hash] # do something with the hash in the block (better to use chunking here) else result << hash end end end # print new line to retain last processing line message print "\n" if [:verbose] # last chunk: if ! chunk.nil? && chunk.size > 0 # do something with the chunk if block_given? yield chunk # do something with the hashes in the chunk in the block else result << chunk # not sure yet, why anybody would want to do this without a block end chunk_count += 1 chunk = [] # initialize for next chunk of data end ensure $/ = old_row_sep # make sure this stupid global variable is always reset to it's previous value after we're done! f.close if f.respond_to?(:close) end if block_given? return chunk_count # when we do processing through a block we only care how many chunks we processed else return result # returns either an Array of Hashes, or an Array of Arrays of Hashes (if in chunked mode) end end |