[Closed] Issues with Version 6

[Rescue9]

@kovidgoyal Just wanted to say thanks for squashing bug 2c4766a8598f075 quickly. Makes reading so much easier. Thx again!

[GloverEggs]

I skipped 6.0.0 and just installed 6.1.0. Every library I have (there are 39!) has the column widths messed up. Everything I'd spent hours setting up to look just so on my monitor is now gone.

Are there any settings that I can use to get back what I used to have?

[thiago.eec]

Originally Posted by GloverEggs

I skipped 6.0.0 and just installed 6.1.0. Every library I have (there are 39!) has the column widths messed up. Everything I'd spent hours setting up to look just so on my monitor is now gone.

Are there any settings that I can use to get back what I used to have?

Probably because of a QT change. See this post.

[jukin]

Upgraded to v6 then found that I would get the following when I clicked on 'Download metadata':

calibre, version 5.8.0
ERROR: Download failed: Failed to download metadata. Click Show Details to see details

Traceback (most recent call last):
File "calibre\utils\ipc\simple_worker.py", line 304, in main
File "calibre\ebooks\metadata\sources\worker.py", line 103, in single_identify
File "calibre\ebooks\metadata\sources\update.py", line 85, in patch_plugins
File "calibre\ebooks\metadata\sources\update.py", line 41, in load_plugin
File "<string>", line 24, in <module>
ImportError: cannot import name 'random_common_chrome_user_agent' from 'calibre.utils.random_ua' (C:\Program Files\Calibre2\app\bin\python-lib.bypy.frozen\calibre\utils\random_ua.pyc)

I uninstalled Calibre. Deleted everything fro APPDATA and installed v5.8.0.

Get the same error now with 5.8, which I was on before. Library is the same and so is its location. I have tried with new books as well as books already installed with full metadata.

Any ideas?

[net-buoy]

Confirming that DeDRM 10.0.3 on Calibre 6.1 is working fine on OSX
Thanks all.
https://github.com/noDRM/DeDRM_tools/discussions/54

[theducks]

Yup, it is (the same) wrong there also.
for the help in tracking this down.
FWIW the other Good system has most of the same apps installed:
Libre office Hmm good has 7.3, Bad has 7.2
Calibre
Sigil
PageEdit
Firefox
It also has a couple NOT on this system

[theducks]

Portable Installed (flash drive) on problem PC
Problem is MISSING while running Portable

I think this is Beer time. (time to forget these worries )

[BetterRed]

@theducks - maybe you're seeing the eraser icon IIRC its a pink thingme, see posts 272 and 277 and thereabouts.

BR

[theducks]

Originally Posted by BetterRed

@theducks - maybe you're seeing the eraser icon IIRC its a pink thingme, see posts 272 and 277 and thereabouts.

BR

The icon is small (and not pink). Looking in app/Resources
It is edit-clear (appears to be greyscale)

[kovidgoyal]

@retiredbiker: It's probably a timing issue if it doesnt show up in debug mode, likely IME related, if you have one enabled, try disabling it.

@DNSB and @Rescue9 you are welcome.

@jukin: You cannot be getting that error on 6.1 it would be a different one. In any case in an hour or two you will stop getting that error on 5.8 as well.

[jukin]

Originally Posted by kovidgoyal

@retiredbiker: It's probably a timing issue if it doesnt show up in debug mode, likely IME related, if you have one enabled, try disabling it.

@DNSB and @Rescue9 you are welcome.

@jukin: You cannot be getting that error on 6.1 it would be a different one. In any case in an hour or two you will stop getting that error on 5.8 as well.

Hi Kovid. I upgraded to 6.1 and everything is fine now. Thanks for your help.

[un_pogaz]

Is it Qt that does the font-weight a numeric value in Calibre 5 and string value in 6 for the comments?
Because currently the comments inherited from 5 have a slightly erroneous rendering of the font-weight [picture].

Actually, it's much nicer so it's cool.

If it's Qt, I understand, I would just be careful when Qt7 will happen.

Windows 10, Calibre 6.1

[kovidgoyal]

Qt 5 uses 600, Qt 6 uses 700. I changed calibre 6 to convert numeric ones to symbolic ones so that this kind of nonsense doesnt happen with future qt upgrades.

[Comfy.n]

Originally Posted by thiago.eec

I noticed this problem with the first beta launched months ago. I had completely forget about this until you mentioned it now. Here is my original post about this.

Apparently, QT has trouble when you use a non integer scale factor (I use 125% on my Windows 10 machine). There are multiple ways to correct this using an environment variable (see @Quoth's post on the other page). In addition to what he suggested, I can confirm that any of this variables work too:

QT_SCALE_FACTOR_ROUNDING_POLICY = Round
QT_ENABLE_HIGHDPI_SCALING = 0

Only one of the above is necessary.

Source:
https://doc.qt.io/qt-6/portingguide.html#high-dpi
https://doc.qt.io/qt-6/highdpi.html#...able-reference

Thank you so much for that input.

Using that second variable has fixed undesired Calibre/QT scaling changes on my side (Winblows Tense, 1920 x 1080 laptop screen, forcibly set to custom 125% scaling in order to bypass MS sharpness incompetence).

For those that face this issue, it might be of help to know where to insert that line: here.

[JSWolf]

Originally Posted by Comfy.n

Thank you so much for that input.

Using that second variable has fixed undesired Calibre/QT scaling changes on my side (Winblows Tense, 1920 x 1080 laptop screen, forcibly set to custom 125% scaling in order to bypass MS sharpness incompetence).

For those that face this issue, it might be of help to know where to insert that line: here.

I too have applied QT_ENABLE_HIGHDPI_SCALING = 0 and it works very well now. Calibre now displays as it did before.