Important
🚧 This documentation is still under construction. Parts of the node are still in development. There may be therefore minor differences between the node itself and the documentation of the node. The documentation is also not yet complete.
A distinction must be made between the terms node and
node. In the Comfy registry there is a node registered. In this case this
is the ComfyUI-Simple_QR_Codes node. Such a node can be
installed using the ComfyUI Manager or using a terminal window and cloning
the node to the correct directory custom_nodes. In case of my
registered node, there are seven different nodes are contained which can be
used in a workflow.
The word QR Code is a registered trademark of the DENSO WAVE INCORPORATED in Japan and other countries. The trademark refers primarily to the word QR Code. However, it should be noted that the trademark must also refer to the encoding and decoding, otherwise the benefits of the QR code would be diluted. Nevertheless the created QR Code images can be used private as well as commercial free of charge.
For the QR code generation one can use the Python
module qrcode or the Python module segno. I
make use of both of them. In terms of regulations, I have to say that
I have not checked that both Python modules comply with the published
QR code standards.
All QR codes used in this documentations are created using my new node.
A fully coloured QR code can look like the one below.
The finder patterns or position markers are the three large square shapes at the corners of the QR codes. They help QR code readers to detect the position of the QR code and align the code properly for scanning.
The alignment patterns or alignment markers are the smaller square patterns inside of a QR code. They help the QR code reader to read the code accurately, especially when distorted or not perfectly aligned.
The quiet zone (tomato) is the blank margin or border around the QR code. It separates the QR code from other elements like graphics or text. This empty space ensures that the QR reader can identify the QR code, allowing for accurate scanning and preventing interference from the surrounding objects.
The timing pattern in the QR code is the alternating dark-light blocks between the finder patterns. These help the QR reader determine the grid's structure by providing a reference for positioning the data in the QR code. These patterns are essential for accurate scanning. They ensure that the QR code is read in the correct orientation and that the data blocks are properly aligned.
Format information refers to a small portion of information in a QR code that stores details about the error correction level. It is located near the finder patterns and is essential for ensuring the accurate decoding, even if the QR code is partially damaged or distorted.
Version information pertains to data that shows the code's dimensions and capability. In bigger sizes from version 7 and above, the data is encoded in the QR code's top-right and bottom-left corners. The version number tells the QR code reader scanner how many blocks or squares are in the grid. QR codes come in 40 different versions, where Version 1 is the smallest with 21x21 blocks and Version 40 is the largest with 177x177 blocks. The QR code’s version affects the amount of data it can store. Larger versions are able to hold more informations.
| Level | Name | Percent (%) |
|---|---|---|
| L | Low | 7 % |
| M | Medium | 15 % |
| Q | Quartile | 25 % |
| H | High | 30 % |
Figure 1 shows a preview of the three nodes which can be used for the creation of QR codes. In the text area one writes the text which should be encoded as QR code.
Figure 1: Nodes overview
The first node is a full implementation of the Python module segno with respect to the QR code parameter.
Figure 2: Workflow preview
So far I have created three nodes. The main goal is to create good looking QR codes that can be used in ComfyUI ComfyUI or that can be saved and used in other applications or in other contexts.
The first node, the simplest, was for testing and learning purposes. The second node can create colorful QR codes and round corners can be added. The third node can be used to integrate logos or image in the QR code.
One idea I have since discarded is the ability to invert the created images. Especially with the QR code with logo, this idea no longer makes sense.
The node is supplied with everything needed to work with each node.
Figure 3: QRCodes (Simple Color)
Figure 4: QRCodes (Logo)
Figure 5: QRCodes (Simple B&W)
A special feature of the second node is the ability to add round corners to the QR code.
The rudimentary QR code reader node is only suitable for reading black QR code on white background so far.
Using the Python package qrcode it
is not clear how to control the size of the created QR code. For the
moment the QR Code is created an then this QR code is resized. This
leads to a loss in quality.
In the course of the day I will programme a qr code scan node. Then I can check the result of the QR code creation within the same workflow.
QR Code is a registered trademark of the DENSO WAVE INCORPORATED.
[1] https://pypi.org/project/qrcode/
[2] https://pypi.org/project/segno/
[3] https://pypi.org/project/segno-pil/
[4] https://segno.readthedocs.io/en/latest/
[5] https://www.copus.io/work/a64d054abaa25f57bc1d23c59e5bec71?spaceId=zentrocdotsposts
[6] https://www.copus.io/work/4444e7bc1d45f2fca85a06d5646feb47?spaceId=zentrocdotsposts
[7] https://github.com/coreyryanhanson/ComfyQR-scanning-nodes
[8] https://registry.comfy.org/nodes
