A bug calculating landsat angles

[gillins]

Original report by Anonymous.

---

The function sunAnglesForExtent in landsatangls.py

Line 227
Original code : (ul_long, ul_lat, ur_long, ur_lat, lr_long, lr_lat, ll_long, ll_lat) = cornerLatLong

Changed code : (ul_lat, ul_long, ur_lat, ur_long, lr_lat, lr_long, ll_lat, ll_long) = cornerLatLong

Then the output : angle map is correct.

[gillins]

Original comment by Neil Flood (Bitbucket: neilflood, GitHub: neilflood).

---

That line of code is correct. I cannot reproduce this.

If there is an error further down, we should find it, but the order of values in that tuple is correct as it is.

[bnucaowu]

from rios import fileinfo
import numpy
file = "/home/LC80420082013220LGN00_B1_toa.TIF"
imgInfo = fileinfo.ImageInfo(file)
cornerLatLong = imgInfo.getCorners(outEPSG=4326)
(ul_lat, ul_long, ur_lat, ur_long, lr_lat, lr_long, ll_lat, ll_long) = cornerLatLong # changed
pts = numpy.array([
... [ul_long, ul_lat],
... [ur_long, ur_lat],
... [ll_long, ll_lat],
... [lr_long, lr_lat]
... ])
longDeg = pts[:, 0]
latDeg = pts[:, 1]
longDeg
array([-98.41048362, -89.37133398, -97.6997143 , -89.84929715])
latDeg
array([74.60575466, 74.64178368, 72.2106416 , 72.2415592 ])
cornerLatLong
(74.60575466167826, -98.41048362112267, 74.64178367907462, -89.37133397624285, 72.24155919861137, -89.84929715401962, 72.21064160380321, -97.69971430154234)

In my case, the older values in that tuple is wrong.

[neilflood]

After some further investigation, it appears that this occurs when using GDAL-3.x. Furthermore, it seems that this is a deliberate change in the behaviour of GDAL, which everyone is now going to have to cope with.
See OSGeo/gdal#1546.

We will have to work out the best approach to dealing with this. Until then, nothing will work with GDAL-3.x.

Thank you for bringing it to our attention.

[neilflood]

With release 1.4.10 of rios, this now works correctly with GDAL 3.x