not six significant decimal digits.

reading and spares some disk space: we can reduce the size of the
field from 20 to 14 characters. We do not loose precision with this
reduction because the order of magnitude of the area is 1e3 km2 or
more so the decimal part of the field was 0. Accordingly, change the
are for a null contour from - 1 to - 1e6 m2.

Bug fix in program test_set_max_speed. No need for width 20 for field "speed".

validity of an eddy depending not only on its amplitude but also its
surface. (Note that, in this version, the component valid is redundant: one
can simply test the component out_cont%n_points.)

amplitude min_amp if the difference between innermost level from
local_extrema and ssh extremum is smaller. This is not only a
performance improvement but also an improvement of results: the
algorithm sometimes finds a contour at just min_amp when before it did
not find any: the bisection algorithm converged just below
min_amp. Also, as a result, if an outermost contour is found, it is
now necessarily of sufficient amplitude.

corner_window for all eddies in a two-dimensional array.

hard-coding them.

Remove component twice of type eddy, no longer needed.

the logic with flat_extr and noise_around is no longer needed.

set_all_outerm: the value of component suff_amp is modified inside the
first batch of get_1_outerm.

>= 0 everywhere outside of procedure local_extrema.

In test Nearby_extr, use output from Get_snapshot_region_1_noise since
we now use suff_amp in extremum shapefile.

than extr_map /= i.

Preparing to remove the coding of sufficient amplitude in extr_map: do
not use anywhere the sign of extr_map, use the component suff_amp of
the type eddy instead.

In program test_successive_overlap, I would rather not use default
values for namelist variables.

Add a larger test for successive_overlap, to have eddies wih
insufficient amplitude in the region.

Stdin to Input. Two input dir provided for in tests.json: $input_dir
and $large_input_dir.

Rename everywhere variable nearby_extr to outside_points to avoid
confusion with function nearby_extr. Rename file "nearby_extr.csv" to
"outside_points.csv".

In procedure nearby_extr, include in the selection extrema with
different sign, even if they do not have sufficient amplitude.

Bug fix in read_eddy: use the value of read attribute to define
e%interpolated.

Bug fix in test_set_max_speed: forgot conversion to rad.

set_all_outerm and in get_snapshot.

define a selection of extrema. Then get their coordinates. This
prepares the way for a more complex selection: taking cylonicity into account.

procedure get_snapshot. The idea is that, in this branch, nearby_extr
will eventually include extrema depending not only on them having
sufficient amplitude but also on their cyclonicity. Both pieces of
information cannot be simply encoded in extr_map. The downside is that
we compute needlessly nearby_extr if e%radius4 == 1 in set_max_speed.

set_max_speed. Abandon the computation with pack_indices, this was
supposed to give the same result than the computation with argwhere
but be faster if not any(extr_map) < 0. I was not sure of that. I
abandon that because I want to eventually remove the coding of
sufficient amplitude in extr_map.

procedure set_all_outerm. The idea is that, in this branch,
nearby_extr will eventually include extrema depending not only on them
having sufficient amplitude but also on their cyclonicity. Both pieces
of information cannot be simply encoded in extr_map.

title lines in "edgelist_1.csv". Add python script reading
"edgelist_1.csv" into a NetworkX graph.

GPC_F and GPC must be used.

read_snapshot, assert that there is no duplicate tuple in
s%ind_extr. Since we are not sure of that, we must use a do construct
and not a forall in order to define s%extr_map.

Bug fix in program test_local_extrema: support missing values. Add
corresponding test.

Output s%ind_extr in program test_read_snapshot.

be useful for testing successive_overlap.

for upcoming procedure read_snapshot.

Bug fix in procedure rossby_number: e%speed_cont could be null.

multi-polygon. This is a useful test because, in successive_overlap,
the intersection of two polygons may give a multi-polygon.

specifying polygon in source code.