Quoted Printable for Crystal

The QuotedPrintable module provides for the encoding (#encode) and decoding (#decode, #decode_string) of binary data using a base64 representation.

Installation

Add this to your application's shard.yml:

dependencies:
  quoted_printable:
    github: arcage/crystal-quotedprintable

Usage

require "quoted_printable"

data = <<-HTM
<html lang=\"ja\">
<head>
  <title>日本語タイトル</title>
</head>
<body>
  <h1>見出し</h1>
  <p>本文<p>
</body>
</html>
HTM

encoded = QuotedPrintable.encode(data)
# => <html lang=3D"ja">
#    <head>
#      <title>=E6=97=A5=E6=9C=AC=E8=AA=9E=E3=82=BF=E3=82=A4=E3=83=88=E3=83=AB</t=
#    itle>
#    </head>
#    <body>
#      <h1>=E8=A6=8B=E5=87=BA=E3=81=97</h1>
#      <p>=E6=9C=AC=E6=96=87<p>
#    </body>
#    </html>

decoded = QuotedPrintable.decode_string(encoded)
#=> <html lang="ja">
#    <head>
#      <title>日本語タイトル</title>
#    </head>
#    <body>
#      <h1>見出し</h1>
#      <p>本文<p>
#    </body>
#    </html>

An argument of the #encode can be String or Enumerable(UInt8)(eg. Bytes, Array(UInt8), Tuple(UInt8)) .

When the String object is given, it will be encoded as a text. That means it will be encoded with literal representation, and all line breaks in it will be represented as a CRLF sequence.

When the Enumerable(UInt8) object is given, all bytes wil be encoded as it is.

In case of the encoding as a text, even if original data was written with LF line breaks, all line breaks in decoded data will be CRLF. To avoid this, you can specify the line break code with line_break: argument of #decode_string method.

# decode with LF line breaks
decoded = QuotedPrintable.decode_string(encoded, line_break: "\n")

To decode non-utf-8 text, you can specify the encoding: and invalid: arguments, same as those of the some methods of String.

# decode from Shift JIS text
decoded = QuotedPrintable.decode_string(encoded, encoding: "Shift_JIS", invalid: :skip)

Contributors