Changing the quote-character in csv parsing

ruby-talk
1

Hi,
I have a bunch of files containing lines as comma-seperated values.
Unfortunately, the character used for quoting is a single quote (') and
not the double quote ("). How can I tell the csv library (or fastercsv
or any other csv-parsing library) which character is used for quoting?
Some of the files contain fields like 'quoted, but with comma', which
are seperated into two fields at the comma:
irb(main):003:0> line = "one, 'quoted', 'quoted, but with comma'"
=> "one, 'quoted', 'quoted, but with comma'"
irb(main):006:0> CSV::parse_line('some words "some quoted text" some
more words', ' ')
=> ["some", "words", "some quoted text", "some", "more", "words"]
irb(main):001:0> require 'rubygems'
=> true
irb(main):002:0> require_gem 'fastercsv'
=> true
irb(main):004:0> line.parse_csv
=> ["one", " 'quoted'", " 'quoted", " but with comma'"]

The output should be ["one", "'quoted'", "'quoted, but with comma'"]

I already have searched the rdoc for the csv library without any success.

2

Hi,
Don't know how it is with fastercsv but csv.rb has double quote hardcoded as
I understand. I think best is to substitute all double quots in your text
with single and vice versa. You can do it with gsub this way:

line.gsub!(/'|\"/){ |c|
  if c == "'"
    "\""
  else
    "'"
  end
}

jan

Jens Auer wrote:

···

Hi,
I have a bunch of files containing lines as comma-seperated values.
Unfortunately, the character used for quoting is a single quote (') and
not the double quote ("). How can I tell the csv library (or fastercsv
or any other csv-parsing library) which character is used for quoting?
Some of the files contain fields like 'quoted, but with comma', which
are seperated into two fields at the comma:
irb(main):003:0> line = "one, 'quoted', 'quoted, but with comma'"
=> "one, 'quoted', 'quoted, but with comma'"
irb(main):006:0> CSV::parse_line('some words "some quoted text" some
more words', ' ')
=> ["some", "words", "some quoted text", "some", "more", "words"]
irb(main):001:0> require 'rubygems'
=> true
irb(main):002:0> require_gem 'fastercsv'
=> true
irb(main):004:0> line.parse_csv
=> ["one", " 'quoted'", " 'quoted", " but with comma'"]

The output should be ["one", "'quoted'", "'quoted, but with comma'"]

I already have searched the rdoc for the csv library without any success.

3

Jan Topinski wrote:

Hi,
Don't know how it is with fastercsv but csv.rb has double quote hardcoded as
I understand. I think best is to substitute all double quots in your text
with single and vice versa. You can do it with gsub this way:

line.gsub!(/'|\"/){ |c|
  if c == "'"
    "\""
  else
    "'"
  end
}

"He said, \"I don't care.\"".tr("'\"", "\"'")

4

ups me blind :wink:
jan
William James wrote:

···

Jan Topinski wrote:

Hi,
Don't know how it is with fastercsv but csv.rb has double quote hardcoded
as I understand. I think best is to substitute all double quots in your
text with single and vice versa. You can do it with gsub this way:

line.gsub!(/'|\"/){ |c|
  if c == "'"
    "\""
  else
    "'"
  end
}

"He said, \"I don't care.\"".tr("'\"", "\"'")