#20342 closed defect (fixed)
Validator results not cleared
Reported by: | Hb--- | Owned by: | GerdP |
---|---|---|---|
Priority: | minor | Milestone: | 21.02 |
Component: | Core validator | Version: | latest |
Keywords: | template_report | Cc: |
Description
What steps will reproduce the problem?
- Create a new layer
- Draw a new way and tag highway=primary to it.
- Click Upload data ...
- In the Suspicious data found. Upload anyway? dialog click Cancel.
- Notice the Unnamed ways warning in the Validation Results panel
- Edit the tag highway=primary and change it to waterway=stream.
- Click again Upload data ...
What is the expected result?
Validation Results panel without an entry.
What happens instead?
No change in the Validation Result panel. It still shows the outdated Unnamed ways warning.
Please provide any additional information below. Attach a screenshot if possible.
The steps described are for easily reproducing the ticket. The misleading behaviour can occur while editing and checking without uploading.
URL:https://josm.openstreetmap.de/svn/trunk Repository:UUID: 0c6e7542-c601-0410-84e7-c038aed88b3b Last:Changed Date: 2021-01-02 16:14:00 +0100 (Sat, 02 Jan 2021) Build-Date:2021-01-03 02:30:59 Revision:17432 Relative:URL: ^/trunk Identification: JOSM/1.5 (17432 en) Windows 7 64-Bit OS Build number: Windows 7 Professional (7601) Memory Usage: 1278 MB / 3604 MB (678 MB allocated, but free) Java version: 1.8.0_242-b08, AdoptOpenJDK, OpenJDK 64-Bit Server VM Look and Feel: com.sun.java.swing.plaf.windows.WindowsLookAndFeel Screen: \Display0 1280×1024 (scaling 1.00×1.00) Maximum Screen Size: 1280×1024 Best cursor sizes: 16×16→48×48, 32×32→48×48 Dataset consistency test: No problems found Plugins: + Mapillary (1.5.33-dirty) + OpenStreetCam (356) + apache-commons (35524) + apache-http (35589) + buildings_tools (35669) + ejml (35458) + geotools (35458) + imagery_offset_db (35640) + jaxb (35543) + jna (35662) + jts (35458) + mapwithai (1.7.1.1) + opendata (35640) + reverter (35640) + utilsplugin2 (35674)
Attachments (0)
Change History (11)
follow-ups: 2 3 comment:1 by , 3 years ago
follow-up: 4 comment:2 by , 3 years ago
I think it should be fixed. It's not at a random time, it's when we click upload data, so the validator is run. It makes sense to update the validator panel.
comment:3 by , 3 years ago
Replying to GerdP:
... each time when you change something ...
That enhancement was dismissed in #17101 for good.
So, if you want new validator results run Validator.
With clicking Upload data ... the second time in Step 7 another validator run is initiated. And on starting this run JOSM should clear the Validator Results panel and the Validator error layer. My additional information was not completely clear for this. I had noticed this misleading behaviour somehow in a workflow without uploading but could not remember the steps leading to it.
comment:4 by , 3 years ago
Replying to Don-vip:
I think it should be fixed. It's not at a random time, it's when we click upload data, so the validator is run. It makes sense to update the validator panel.
OK, I assume an empty error list was ignored to be able to keep the results from a normal validator run? I concentrated on the last sentence:
The misleading behaviour can occur while editing and checking without uploading.
If this is main problem see comment:1
comment:5 by , 3 years ago
Owner: | changed from | to
---|---|
Status: | new → assigned |
comment:6 by , 3 years ago
OK, at least two different problems here:
1) If all errors that were found are in the ignore list the validator tree is not rebuild (probably a regression of #16335)
2) The error layer is not updated when ValidatorTreePanel.buildTree()
is called with an empty error list because
invalidationListeners.fireEvent(Runnable::run);
isn't executed.
comment:10 by , 3 years ago
Milestone: | → 21.01 |
---|
I think this is a wontfix. It's nearly impossible to recompute the list of errors each time when you change something in the data. It would extremely slow down JOSM. So, if you want new validator results run Validator.