an issue with extracting westward trajectories

[HzFWQhvevm4s]

Hello, I've been trying to extract westward eddy trajectories with a longitude span over 10 degrees, but it ends up extracting those with a span between -10.0 and 0.0(-10.0 ≤ delta_lon ≤ 0.0), so I took a look at the source code and noticed that there might be something causing the issue:

py-eddy-tracker/src/py_eddy_tracker/observations/tracking.py

Lines 366 to 385 in 3a54bbb

	def extract_toward_direction(self, west=True, delta_lon=None):
	"""
	Get trajectories going in the same direction
	:param bool west: Only eastward eddies if True return westward
	:param None,float delta_lon: Only eddies with more than delta_lon span in longitude
	:return: Only eastern eddy
	:rtype: __class__
	.. minigallery:: py_eddy_tracker.TrackEddiesObservations.extract_toward_direction
	"""
	lon = self.longitude
	i0, nb = self.index_from_track, self.nb_obs_by_track
	i1 = i0 - 1 + nb
	d_lon = lon[i1] - lon[i0]
	m = d_lon < 0 if west else d_lon > 0
	if delta_lon is not None:
	m *= delta_lon < d_lon
	m = m.repeat(nb)
	return self.extract_with_mask(m)

my settings:

c.extract_toward_direction(west=True, delta_lon= -10.0)

# let lon[i1] = -40.0, lon[i0] = -20.0
d_lon = lon[i1] - lon[i0]                          
# d_lon = -20.0

m = d_lon < 0 if west else d_lon > 0    
# west = True, d_lon < 0 = True, m = True

if delta_lon is not None:                        
# True

    m *= delta_lon < d_lon                     
    # delta_lon < dlon = False, m= 0

m = m.repeat(nb)                                 
# nb of points masked

return self.extract_with_mask(m) 

an alternative approach might be:

if delta_lon is not None:
    m *= abs(delta_lon) < abs(d_lon)

Best wishes,
lolcat

[HzFWQhvevm4s]

Hello, it's been a while. Am I misunderstanding the concepts, or is it just unsolved?

[AntSimi]

There was a mistake, abs on d_lon solve problem and delta_lon must be positiv.
It's corrected on last commit.
Thanks,