Regression on Brother QL500 label printer introduced in 1.28.12

[christianlupus]

Hello,

I have a Brother QL500 label printer. I use Archlinux and updated recently from 1.28.11 to 1.28.12. After the update, I was no longer able to properly print on DK-11209 labels (62x29mm, landscape). See below for an image of the problem. The upper left image is printed with 1.28.12 while the lower one with 1.28.11.

To rule out a physical defect, I printed the small 23x23mm label to the right (also with 1.28.12). The printer is built such that the labels are aligned on the right edge of the carrier. That shows that the problem of white margins only appears for this specific page size.

Unfortunately, I do not have other labels at hand just now. So, I cannot test other sizes. I'll check if I find some in the storage tomorrow.

Switching back and forth between the versions enables and disabled the problem so it seems a regression. Please tell me if I can somehow support you in finding the issue. Doing a git bisect might be possible but will require some preparation on my side.

[tillkamppeter]

Could you please do the following:

Run the command

cupsctl --debug-logging

Install cups-filters 1.28.12 (failing case). Then print a job, check whether it still fails (= has wrong margins). Attach your /var/log/cups/error_log file to this bug report.

Install cups-filters 1.28.11 (succeeding case). Then print a job, check whether it still succeeds (= has correct margins). Attach your /var/log/cups/error_log file to this bug report.

Please also install your file /etc/cups/ppd/<queuename>.ppd file where your replace <queuename> by the name of the print queue of your label printer.

Please do not compress or package together the files and attach copies renamed to have a .txt file extension to get accepted here on the GitHub site.

[christianlupus]

Sorry for my late answer. I had to prepare for the tests first.

Here come the requested files:

• The QL500.ppd.txt
• The logs of the failing print logs-failing.txt
• The logs of the succeeding print logs-succeeding.txt
• The complete logs (first failing then succeeding) error_log-after-all.txt

I verified that the prints fail/succeed according to my initial post. The two selective logs are created by making a copy of /var/log/cups/error_log just after cupsctl --debug-logging and one after the print itself. Then, I checked the number of lines and used tail to cut away anything before the print. If you need a general overview, please have a look at the total logs.

[tillkamppeter]

Could you also provide your input (PDF?) file which you had sent to the printer? With which application did you create it?

[christianlupus]

Hello. I created a PDF using glabels-3. The file was generated by print-to-pdf: Ausgabe.pdf

I sent to the printer via lp -d QL500 Ausgabe.pdf.

[tillkamppeter]

Thanks, I could already reproduce the bug by creating the PDF test page for this printer with bannertopdf and use this as input file.

The problem occurs whenever a printer pulls the paper long-edge first and not the usual short-edge first, like most inkjets or lasers do. Long-edge first is common on label printers and on (usually roll-fed with cutter) large format printers. So I could reproduce your problem with your PPD and also with a PPD for the HP DesignJet T1100 (choosing the "A4.Transverse" page size there).

What happens is that the cropping of the page image is not done correctly at least with page-scaling settings auto (default), none, and fill. I have already found a fix for this and will commit it soon and then release cups-filters 1.28.13 (which will then also be the version of Ubuntu 22.04 LTS, Jammy Jellyfish)..

[christianlupus]

Wow, that was fast. Thanks for the help! I will wait for the updated version.

[tillkamppeter]

Fixed in the GIT repositories, on master with commit 70e0f47 and on 1.x with commit d4ae672.

I will release 1.28.13 with this fix soon.

Thanks a lot for this bug report.

[tillkamppeter]

1.28.13 is released now.