Bug #585
gmtsimplify
| Status: | Closed | Start date: | 2014-07-03 | |
|---|---|---|---|---|
| Priority: | Normal | Due date: | ||
| Assignee: |  Paul | % Done: | 100% | |
| Category: | - | |||
| Target version: | Candidate for next bugfix release | |||
| Affected version: | 5.1.x | Platform: | Linux |
Description
Sometimes (not always) gmtsimplify connects start and end points of previously opened contours.
command : gmt gmtsimplify contours_orig.txt -T0.00001m -V -fg > contours_simpl.txt
Contours are in lat/long.
See the pictures : contours_original - original contours after grdcontour
contours_simplified - same contours after simplification.
Associated revisions
Address issue #585
History
#1
Updated by Paul almost 3 years ago
- Status changed from New to Feedback
Could you possibly attach the contours_orig.txt file so I can have a look? If possible, a subset that still shows the problem would be even better.
You say sometimes: Do you mean that this particular command is not repeatable (sometimes fails, sometimes not) or that it depends on the data set?
#2
Updated by Peeter almost 3 years ago
Compressed size of this file is 71 MB, so I can not add it to this thread. Subset may not show the problem - I can not check this right now.
So I added the file to dropbox and can share it, but I need some email address for that.
Alternatively I can upload the file to some ftp site.
Under repeatability I mean different contour files - I have not seen any problems in some smaller areas.
I have such file also available for comparison (contours_orig_ok.txt).
#3
Updated by Paul almost 3 years ago
OK, send link to
#4
Updated by Peeter almost 3 years ago
done
#5
Updated by Paul almost 3 years ago
Got them, you can remove them from Dropbox if you'd like. Late here in Hawaii, will have a look tomorrow.
#6
Updated by Paul almost 3 years ago
- Status changed from Feedback to Resolved
- Assignee changed from Peeter to Paul
- % Done changed from 0 to 100
Input had a mix of polygons and lines and gmtsimplify got confused and forced all to be closed polygons. For some reasons I was thinking of gmtconnect so it took a while to mentally realize this was a much simpler bug than I first thought. In r13321.