Project

General

Profile

Feature #11839 » csv_liberal_parsing.diff

bluej100 (Braden Anderson), 12/18/2015 01:21 AM

View differences:

lib/csv.rb
# <b><tt>:skip_blanks</tt></b>:: +false+
# <b><tt>:force_quotes</tt></b>:: +false+
# <b><tt>:skip_lines</tt></b>:: +nil+
# <b><tt>:liberal_parsing</tt></b>:: +false+
#
DEFAULT_OPTIONS = { col_sep: ",",
row_sep: :auto,
......
header_converters: nil,
skip_blanks: false,
force_quotes: false,
skip_lines: nil }.freeze
skip_lines: nil,
liberal_parsing: false }.freeze
#
# This method will return a CSV instance, just like CSV::new(), but the
......
# a comment. If the passed object does
# not respond to <tt>match</tt>,
# <tt>ArgumentError</tt> is thrown.
# <b><tt>:liberal_parsing</tt></b>:: When set to a +true+ value, CSV will
# attempt to parse input not conformant
# with RFC 4180, such as double quotes
# in unquoted fields.
#
# See CSV::DEFAULT_OPTIONS for the default settings.
#
......
def skip_blanks?() @skip_blanks end
# Returns +true+ if all output fields are quoted. See CSV::new for details.
def force_quotes?() @force_quotes end
# Returns +true+ if illegal input is handled. See CSV::new for details.
def liberal_parsing?() @liberal_parsing end
#
# The Encoding CSV is parsing or writing in. This will be the Encoding you
......
csv.last << @col_sep
end
elsif part[0] == @quote_char
# If we are staring a new quoted column
if part[-1] != @quote_char || part.count(@quote_char) % 2 != 0
# If we are starting a new quoted column
if part.count(@quote_char) % 2 == 0
if part[-1] == @quote_char
# regular quoted column
csv << part[1..-2]
if csv.last =~ @parsers[:stray_quote]
raise MalformedCSVError,
"Missing or stray quote in line #{lineno + 1}"
end
csv.last.gsub!(@quote_char * 2, @quote_char)
elsif @liberal_parsing
csv << part
else
raise MalformedCSVError,
"Missing or stray quote in line #{lineno + 1}"
end
else
# start an extended column
csv << part[1..-1]
csv.last << @col_sep
in_extended_col = true
else
# regular quoted column
csv << part[1..-2]
if csv.last =~ @parsers[:stray_quote]
raise MalformedCSVError,
"Missing or stray quote in line #{lineno + 1}"
end
csv.last.gsub!(@quote_char * 2, @quote_char)
end
elsif part =~ @parsers[:quote_or_nl]
# Unquoted field with bad characters.
......
raise MalformedCSVError, "Unquoted fields do not allow " +
"\\r or \\n (line #{lineno + 1})."
else
raise MalformedCSVError, "Illegal quoting in line #{lineno + 1}."
if @liberal_parsing
csv << part
else
raise MalformedCSVError, "Illegal quoting in line #{lineno + 1}."
end
end
else
# Regular ole unquoted field.
......
str << " encoding:" << @encoding.name
# show other attributes
%w[ lineno col_sep row_sep
quote_char skip_blanks ].each do |attr_name|
quote_char skip_blanks liberal_parsing ].each do |attr_name|
if a = instance_variable_get("@#{attr_name}")
str << " " << attr_name << ":" << a.inspect
end
......
# store the parser behaviors
@skip_blanks = options.delete(:skip_blanks)
@field_size_limit = options.delete(:field_size_limit)
@liberal_parsing = options.delete(:liberal_parsing)
# prebuild Regexps for faster parsing
esc_row_sep = escape_re(@row_sep)
test/csv/test_features.rb
assert_equal(3, count)
end
def test_liberal_parsing
input = '"Johnson, Dwayne",Dwayne "The Rock" Johnson'
assert_raise(CSV::MalformedCSVError) do
CSV.parse_line(input)
end
assert_equal(["Johnson, Dwayne", 'Dwayne "The Rock" Johnson'],
CSV.parse_line(input, liberal_parsing: true))
input = '"quoted" field'
assert_raise(CSV::MalformedCSVError) do
CSV.parse_line(input)
end
assert_equal(['"quoted" field'],
CSV.parse_line(input, liberal_parsing: true))
assert_raise(CSV::MalformedCSVError) do
CSV.parse_line('is,this "three," or four,fields', liberal_parsing: true)
end
assert_equal(["is", 'this "three', ' or four"', "fields"],
CSV.parse_line('is,this "three, or four",fields', liberal_parsing: true))
end
def test_csv_behavior_readers
%w[ unconverted_fields return_headers write_headers
skip_blanks force_quotes ].each do |behavior|
(1-1/3)