Bug report #7229
Saving layers with records with no/empty geometries messes badly a vector
| Status: | Closed | ||
|---|---|---|---|
| Priority: | Severe/Regression | ||
| Assignee: |  Marco Hugentobler | ||
| Category: | Vectors | ||
| Affected QGIS version: | master | Regression?: | No |
| Operating System: | Easy fix?: | No | |
| Pull Request or Patch supplied: | No | Resolution: | fixed |
| Crashes QGIS or corrupts data: | Yes | Copied to github as #: | 16260 |
Description
This ticket is a follow up of #5584
In that ticket an issue that caused crashes in qgis 1.8/1.7.4 (but not 1.6) was fixed, by allowing a record with no geometry to be saved. Actually the ticket was reopened because while the data model seems to accept this case, it does not seems to make much sense and it easy to prove that can be lead to confusions and mistakes (so the requested action is to reject the record and throw a warning/message).
While making tests for the #5584 ticket I discovered another bad issue, that is well shown in the attached screencast:
when for any reason an empty geometry is created, and the vector saved, there are polygons that become duplicated and if a user continue on that line (digitizing something that results in empty geometries and saving) the mess grows up.
In the example the empty geometries are easily created by using the "avoid intersection" option and digitizing against existing polygons with topological errors in it (ex: a simple auto-intersection).
History
#1
Updated by Giovanni Manghi almost 12 years ago
- Status changed from Open to Feedback
with
208c9206ad972b96aef0fdfdebd8012ffa120fe7
this is fixed.
I just wonder if there are other ways (other than the standard digitizing tools) to obtain a record with an empty geometry. If yes, then I guess that in those cases this issue would still surface. If not then I guess this can be closed.
#2
Updated by Giovanni Manghi over 11 years ago
- Resolution set to fixed
- Status changed from Feedback to Closed