qgis / QGIS

QGIS is a free, open source, cross platform (lin/win/mac) geographical information system (GIS)
https://qgis.org
GNU General Public License v2.0
10.66k stars 3.01k forks source link

Size of attribute selector is wrong in attribute table in editing mode #49574

Open dbantoine opened 2 years ago

dbantoine commented 2 years ago

What is the bug or the crash?

Hello,

I have an appearance bug on QGIS 3.26.1 on Windows. The size of the attribute selector is wrong in the attribute table in editing mode. It is not wide enough, and definitely less wide than il previous version (using 3.24.3 as a comparison). 3.26.1 : 3 26 1 3.24.3 : 3 24 3

I am using the same project and the same data in both versions. I have tested with a few differrents data. From what I can tell it is not a configuration thing in any parameter I could find. But maybe I am missing something.

I will give any more information if needed. Thanks,

Have a nice day,

Steps to reproduce the issue

Open any data attribute table Go to editing mode

Versions

Version de QGIS | 3.26.1-Buenos Aires | Révision du code | b609df9ed4d -- | -- | -- | -- Version de Qt | 5.15.3 Version de Python | 3.9.5 Version de GDAL/OGR | 3.5.1 Version de Proj | 9.0.1 Version de la base de données du registre EPSG | v10.064 (2022-05-19) Version de GEOS | 3.10.3-CAPI-1.16.1 Version de SQLite | 3.38.1 Version de PDAL | 2.3.0 Version du client PostgreSQL | unknown Version de SpatiaLite | 5.0.1 Version de QWT | 6.1.6 Version de QScintilla2 | 2.13.1 Version de l'OS | Windows 10 Version 2009   |   |   |   Extensions Python actives cadastre | 1.15.0 geometry_paster | 0.1.1 go2streetview | 8.6 google-street-view-layer-master | 0.1 mmqgis | 2021.9.10 PointConnector | 2.0 qgis-plugin-findreplace-main | 1 QuickOSM | 2.0.1 quick_map_services | 0.19.29 send2google_earth | 0.3.1 SpreadsheetLayers | 2.0.1 db_manager | 0.1.20 processing | 2.12.99 sagaprovider | 2.12.99 Version de QGIS 3.26.1-Buenos Aires Révision du code [b609df9ed4d](https://github.com/qgis/QGIS/commit/b609df9ed4d) Version de Qt 5.15.3 Version de Python 3.9.5 Version de GDAL/OGR 3.5.1 Version de Proj 9.0.1 Version de la base de données du registre EPSG v10.064 (2022-05-19) Version de GEOS 3.10.3-CAPI-1.16.1 Version de SQLite 3.38.1 Version de PDAL 2.3.0 Version du client PostgreSQL unknown Version de SpatiaLite 5.0.1 Version de QWT 6.1.6 Version de QScintilla2 2.13.1 Version de l'OS Windows 10 Version 2009 Extensions Python actives cadastre 1.15.0 geometry_paster 0.1.1 go2streetview 8.6 google-street-view-layer-master 0.1 mmqgis 2021.9.10 PointConnector 2.0 qgis-plugin-findreplace-main 1 QuickOSM 2.0.1 quick_map_services 0.19.29 send2google_earth 0.3.1 SpreadsheetLayers 2.0.1 db_manager 0.1.20 processing 2.12.99 sagaprovider 2.12.99 ### Supported QGIS version - [X] I'm running a supported QGIS version according to the roadmap. ### New profile - [ ] I tried with a new QGIS profile ### Additional context _No response_
nicogodet commented 2 years ago

3.26.1: image image

Even if you expand the attribut table window ?

dbantoine commented 2 years ago

Yes, even in full screen. The screens above are full screen. Sc_2022 08 04_15 26 But it looks like it is somewhat inconsistent. Sometimes it’s ok in the tries I just made. I am not able to find what is doing it or not. 24" monitor if that helps

gioman commented 2 years ago

I also can't confirm this. Maybe is related to the theme/desktop manager your are using?

dbantoine commented 2 years ago

I am not using anything modifying the UI either from parameters, customization, theme or plugin. Using Style = Fusion and Theme = default, which are default I think. image

I just tried to reproduce: Add one a layer (shapefile, line layer) to a fresh project: no problem even playing with the window and resizing it in every possible way. Add another layer (same type): it directly did the bug opening the attribute table.

Example of the layer used test.zip

Other test: Closed QGIS Open a fresh project Add the same layer as the first one from above: bug directly.

Re-did the same with the second layer, bug again.

I tried to resave the layer, same problem. I tried with a few different layers and same problem.

Always testing in full screen. No style on any of the layers. Only one QGIS running at the same time in all the test.

gioman commented 2 years ago

Example of the layer used test.zip

image

dbantoine commented 2 years ago

It's not related to this layer in particular. I tried with a few other ones (just dowloaded from known sources), and I have the same problem. Sc_2022 08 09_09 05 Sc_2022 08 09_09 08 Sc_2022 08 09_09 04 So I do think it has nothing to do with the layer opened.

I tried to use a new profile. And also to uninstall / download / install QGIS, the problem is still here.

nicogodet commented 2 years ago

Still we can't reproduce...

Font size in parameter ? (default is 8) image

In Windows, do you have any zoom defined ? (as you are French : Paramètres -> Systèmes -> Affichage -> Mises à l'échelle et disposition) image

A zoom != 100% can mess up with QGIS from experience.

dbantoine commented 2 years ago

Font size is 8 default. Windows zoom is 100%.

I have no clue as how to reproduce on your side.

It's a strange bug, but not very major. I get it if you think it's not worth investigate.

gioman commented 2 years ago

I get it if you think it's not worth investigate.

@dbantoine the point is that if is not possible to replicate in a consistent matter ten is hard/impossible to troubleshoot. We can leave this open with a feedback label, if no more infos about to replicate will be added in the upcoming weeks this will be automatically closed.

dbantoine commented 2 years ago

I guess it is not going to help much, but I just tried on another PC, and the bug occured right away.

Used another PC (laptop), which had and old version of QGIS installed. To start fresh: -uninstalled QGIS -deleted all profile files -deleted all qgis related files -cleaned the Windows registry -rebooted -downloaded and installed QGIS QGIS 3.26.1 -downloaded some data from official sources : https://geoservices.ign.fr/adminexpress (last version available), in Shapefile format -opened some data and the attribute table : Sc_2022 08 11_18 32 Sc_2022 08 11_18 30 Sc_2022 08 11_18 31

It's not consistent. One of the layers opened was fine the first time I opened the attribute table, and was not right the second time.

I have no clue as what more information I can give to reproduce since it's a fresh QGIS install.

dbantoine commented 2 years ago

A new piece of information : I just installed QGIS 3.26.1 on a brand new PC (Windows 11), and the error occured directly.

github-actions[bot] commented 2 years ago

The QGIS project highly values your report and would love to see it addressed. However, this issue has been left in feedback mode for the last 14 days and is being automatically marked as "stale". If you would like to continue with this issue, please provide any missing information or answer any open questions. If you could resolve the issue yourself meanwhile, please leave a note for future readers with the same problem and close the issue. In case you should have any uncertainty, please leave a comment and we will be happy to help you proceed with this issue. If there is no further activity on this issue, it will be closed in a week.

github-actions[bot] commented 2 years ago

While we hate to see this happen, this issue has been automatically closed because it has not had any activity in the last 42 days despite being marked as feedback. If this issue should be reconsidered, please follow the guidelines in the previous comment and reopen this issue. Or, if you have any further questions, there are also further support channels that can help you.

nicogodet commented 1 year ago

I run into this issue myself on 3.28.3 Can't figure out why...

nicogodet commented 1 year ago

@agiudiceandrea I hesitated to re-open as I really don't know how to reproduce...

agiudiceandrea commented 1 year ago

@nicogodet I think that since more than one user have experienced the issue, then it could be considered confirmed even if there is still no known way to reproduce it consistently.