Modify

Opened 9 years ago

Closed 9 years ago

#12993 closed defect (wontfix)

Validator rules which only work on downloaded areas did not work on new layer

Reported by: mdk Owned by: team
Priority: normal Milestone:
Component: Core validator Version: latest
Keywords: Cc:

Description

If you try to create a sample file by copy&paste data into a new layer or by drawing samples on a new layer, the validator treat the new layer as compleatly not downloaded. This is technically correct, but it is very confusing if you try to create a sample file for a ticket.

Either a new layer should compleatly covered with the "not downloaded" hatching as long as no data is downloaded or the validator should treat a new layer as compleatly downloaded if no download has happend.

To reproduce:
1.) Download an area (not too big)
2.) Draw a waterway=stream in the downloaded area which is not connected to other waterway=*
3.) Validate -> correct warning
4.) Copy the waterway
5.) Add a new layer
6.) Paste the waterway (or draw a secon waterway=stream)
7.) Validate the new layer -> no warning
8.) Download on the new layer an area which covers the new waterway
9.) Validate the new layer -> correct warning

Attachments (0)

Change History (6)

comment:1 by Klumbumbus, 9 years ago

Replying to mdk ticket 12992:

If you copy all and past it to a new layer, validation shown no warning.

This is because of the code :in-downloaded-area. This has two effects:

  • It avoids false positives if the waterway ends out of the downloaded area
  • If there is no downloaded area in your layer then this rule is not executed. This is useful because if you e.g. download only some single waterways with overpass then you don't get this warning for every waterway because in the complete osm data these waterways are correctly connected.

Either a new layer should compleatly covered with the "not downloaded" hatching as long as no data is downloaded

That would be pretty disturbing.

or the validator should treat a new layer as compleatly downloaded if no download has happend.

That would not be good for the case explained above.

To me the current bahavior is the best solution.

Last edited 9 years ago by Klumbumbus (previous) (diff)

comment:2 by mdk, 9 years ago

But the (missing) hatching on a new layer already is only possible because there is already an exceptional case if nothing is downloaded. For the rendering a missing bounding box is treated as "all is downloaded", but for the validator this means "nothing is downloaded".

in reply to:  2 comment:3 by Klumbumbus, 9 years ago

Replying to mdk:

but for the validator this means "nothing is downloaded".

for the validator it rather means "the data in this layer was not downloaded as area, so the data is probably not complete, so don't execute the tests, which depend on complete data (or the difference between inside and outside the downloaded area)"

Last edited 9 years ago by Klumbumbus (previous) (diff)

comment:4 by mdk, 9 years ago

But on the other hand, this makes it very hard to create validator test data for a normal mapper - as I had to figure out by my self :-/

in reply to:  4 comment:5 by Klumbumbus, 9 years ago

Replying to mdk:

But on the other hand, this makes it very hard to create validator test data for a normal mapper - as I had to figure out by my self :-/

You should be happy and not sad, that you figured it out :)

I added the hint to Help/Dialog/Validator

Every effort to make a ticket more clear is appreciated. Thanks. However, for such simple cases like in #12992 it is not really required to add a sample file. If we cannot reproduce, we will ask you for a sample file or an object id.

comment:6 by Klumbumbus, 9 years ago

Resolution: wontfix
Status: newclosed

Modify Ticket

Change Properties
Set your email in Preferences
Action
as closed The owner will remain team.
as The resolution will be set.
The resolution will be deleted. Next status will be 'reopened'.

Add Comment


E-mail address and name can be saved in the Preferences .
 
Note: See TracTickets for help on using tickets.