Bug report #16657
date edit widget only shows current date whatever the real data is underneath
Status: | Closed | ||
---|---|---|---|
Priority: | High | ||
Assignee: | Denis Rouzaud | ||
Category: | Edit widget | ||
Affected QGIS version: | 2.18.9 | Regression?: | Yes |
Operating System: | Easy fix?: | No | |
Pull Request or Patch supplied: | No | Resolution: | fixed/implemented |
Crashes QGIS or corrupts data: | No | Copied to github as #: | 24557 |
Description
Hi,
it seems a regression since 2.18.3, date edit widget will only show the current date in a form, even without any default value set or being readonly.
This makes it pretty unusable.
I had hard time spotting it since another issue with qlr export #16655 let me think that only my project was corrupted.
History
#1 Updated by Nyall Dawson over 7 years ago
- Status changed from Open to Feedback
That's a datetime field - I think the issue is that you've set the format to "date" along. Does it work correctly if you change that to "date/time"?
What data provider is this? can you share the data?
#2 Updated by Regis Haubourg over 7 years ago
- File date_default_issue.gpkg added
That's a datetime field - I think the issue is that you've set the format to "date" along. Does it work correctly if you change that to "date/time"?
Nope, I tested many widget configuration changes before. Only setting it back to textEdit allowed me to see the actual data.
What data provider is this?
Original issue was detected in postgres with a view with edit triggers pointing to a table with a "timestamp with time zone" field.
can you share the data?
Oh, I joined the wrong file. Here is a gpkg file going with the date_default_widget_value_issue.qgs file
#3 Updated by Giovanni Manghi over 7 years ago
- Status changed from Feedback to Open
#4 Updated by Regis Haubourg over 7 years ago
when I cast the field in the SQL view to a date field, QGIS will handle it correctly.
#5 Updated by Denis Rouzaud about 7 years ago
using yyyy-MM-dd HH:mm:ss+02 in the format solves the diplay issue.
The field format must be set accordingly to what is in the data, so it's not really a bug here.
Although I tried yyyy-MM-dd HH:mm:ss+t (time zone) without success. So there might be an issue here.
#6 Updated by Giovanni Manghi about 7 years ago
- Status changed from Open to Feedback
#7 Updated by Denis Rouzaud about 7 years ago
- Status changed from Feedback to Open
So I guess, there is two things to improve here:
- guessing the format from the source
- fix time zone format
#8 Updated by Giovanni Manghi about 7 years ago
Denis Rouzaud wrote:
So I guess, there is two things to improve here:
- guessing the format from the source
- fix time zone format
the ticket has been tagged as a regression because apparently the guessing worked until 2.18.3
#9 Updated by Denis Rouzaud about 7 years ago
- Assignee set to Denis Rouzaud
#10 Updated by Regis Haubourg about 7 years ago
- Priority changed from High to Normal
Lowering the priority since casting to DATE format gives a workaround
#11 Updated by Giovanni Manghi about 7 years ago
- Priority changed from Normal to High
Regis Haubourg wrote:
Lowering the priority since casting to DATE format gives a workaround
still a regression, so to keep consistency with the others better keep priority "high".
#12 Updated by Luigi Pirelli about 7 years ago
error is probably only located in setting the correct format string here:
https://github.com/qgis/qgis/blob/release-2_18/src/gui/editorwidgets/qgsdatetimeeditfactory.cpp#L88
btw Qt:ISODate formate parse correctly the proposed date.
#13 Updated by Denis Rouzaud almost 7 years ago
#14 Updated by Denis Rouzaud almost 7 years ago
fixed in master and 2.18
#15 Updated by Nyall Dawson almost 7 years ago
- Status changed from Open to Closed
#16 Updated by Giovanni Manghi over 6 years ago
- Resolution set to fixed/implemented