Encoding Data

The standard calls the data’s encoding its mode. The QR code standard defines how to encode any given piece of data. There are four possible modes. This module supports three of them: numeric, alphanumeric, and binary.

Each mode is worse at encoding the QR code’s contents. In other words, each mode will require more room in the QR code to store the data. How much data a code version can hold is dependent on what mode is used and the error correction level. For example, the binary encoding always requires more code words than the numeric encoding.

Because of this, it is generally better to allow the QRCode object to auto-select the most efficient mode for the code’s contents.

Note

The QRCode object can automatically choose the best mode based on the data to be encoded. In general, it is best to just let the object figure it out for you.

Numeric Encoding

The numeric type is the most efficient way to encode digits. Problematically, the standard make no provisions for encoding negative or fractional numbers. This encoding is better than Alphanumeric, when you only have a list of digits.

To use this encoding, simply specify a string of digits as the data. You can also use a positive integer as the code’s contents.

>>> number = pyqrcode.create(123456789012345)
>>> number2 = pyqrcode.create('0987654321')

Alphanumeric

The alphanumeric type is very limited in that it can only encode some ASCII characters. It encodes:

• Uppercase letters
• Digits 0-9
• The horizontal space
• Eight punctuation characters: $, %, *, +, -, ., /, and :

A complete list of the possible characters can be found in the pyqrcode.tables.ascii_codes dictionary. While limited, this encoding is much more efficient than using the binary encoding, in many cases. Luckily, the available characters will let you encode a URL.

>>> url = pyqrcode.create('http://uca.edu'.upper())

Kanji

The final mode allows for the encoding of Kanji characters. Denso Wave, the creators of the QR code, is a Japenese company. Hence, they made special provisions for using QR codes with Japenese text.

Only one python string encoding for Kanji characters is supported, shift-jis. The auto-detection algorithm will try to encode the given string as shift-jis. if the characters are supported, then the mode will be set to kanji. Alternatively, you can explicitly define the data’s encoding.

>>> utf8 = 'モンティ'.encode('utf-8')
>>> monty = pyqrcode.create(utf8, encoding='utf-8')
>>> python = pyqrcode.create('錦蛇')

Binary

When all else fails the data can be encoded in pure binary. This encoding does not change the data in any way. Instead its pure bytes are represented directly in the QR code. This is the least efficient way to store data in a QR code. You should only use this as a last resort.

The quotation below must be encoded in binary because of the apostrophe, exclamation point, and the new line character. Notice, that the string’s characters will not have their case changed.

>>> life = pyqrcode.create('''MR. CREOSOTE: Better get a bucket. I'm going to throw up.
    MAITRE D: Uh, Gaston! A bucket for monsieur. There you are, monsieur.''')