toddkarin/global-land-mask

Check whether a lat/lon point is on land for any point on earth

global-land-mask's Issues

numpy requirement missing in install_requires

The numpy requirement, although documented, is missing from the install_requires field in setup.py (which is missing in general). It would be great if we had it, so that pip installs it automatically. ☺️

Caspian Sea recorded as being a land

Hey guy's
Caspian sea seems recorded as being a land and not a sea, how could it be modified ?

lat = 40.44435
lon = 50.61982

Many thanks

Does not work on replit

On replit.com, using global land mask does not work. It shows this error when starting the project

  File "main.py", line 5, in <module>
    import image
  File "/home/runner/mapple-1/image.py", line 1, in <module>
    from global_land_mask import globe
  File "/home/runner/mapple-1/venv/lib/python3.8/site-packages/global_land_mask/__init__.py", line 3, in <module>
    from .globe import *
  File "/home/runner/mapple-1/venv/lib/python3.8/site-packages/global_land_mask/globe.py", line 26, in <module>
    _mask_fid = np.load(_mask_filename)
  File "/home/runner/mapple-1/venv/lib/python3.8/site-packages/numpy/lib/npyio.py", line 407, in load
    fid = stack.enter_context(open(os_fspath(file), "rb"))
FileNotFoundError: [Errno 2] No such file or directory: '/home/runner/.cache/pip/pool/bc/6d/73/globe_combined_mask_compressed.npz'```

Malware in the package, do not execute

In one of the first lines of the globe.py file a malicious compressed file called globe_combined_mask_compressed.npz is loaded in path.

It is responsible for loading the file into memory.

_mask_filename = os.path.join(_path,'globe_combined_mask_compressed.npz')
_mask_fid = np.load(_mask_filename)

I found it strange a coordinate calculation using a hidden library, and decided to pass the file in the VirusTotal website. This library when executed is malicious and contains malware.

HERE IS the link

Mask mismatch at poles

Wikimedia "Mercator Projection" vs Global Land Mask (green & red). Green means an antipode with land on both sides.

I believe my x,y to lat, long coordinate conversion is accurate +/- <1 degree, feel free to check it out yourself:

import numpy as np;
import math;
def XYToLatLong(x,y):
    longitude = (x/(mapWidth/360))-180;
    mercN = ((-y + (mapHeight/2))*(2*math.pi))/mapWidth;
    latRad = 2*(np.arctan(np.exp(mercN))-(math.pi/4));
    latitude = latRad/(math.pi/180);

    return latitude,longitude

Use f(lon, lat) instead of f(lat, lon) ordering of arguments

Hi,

just a feedback to suggest that it's more natural to use f(lon, lat) given that longitude (latitude) values refer to x (y) coordinates on a map,
so calling f(lon, lat) it's equivalent to f(x,y).

Nan values and masked arrays

It would be really useful going forward if commands such as:

from global_land_mask import globe
globe.is_land(lat,lon)

would take masked arrays or ignore nan values

global inland lake

is there a mask out there for global inland lakes? this would be a useful feature as well.

High memory requirement

I'm trying to run your package on a Raspberry Pi Zero and I'm running into memory issues.

Traceback (most recent call last):
  File "grid-nn.py", line 7, in <module>
    from global_land_mask import globe
  File "/home/pi/.local/lib/python3.7/site-packages/global_land_mask/__init__.py", line 3, in <module>
    from .globe import *
  File "/home/pi/.local/lib/python3.7/site-packages/global_land_mask/globe.py", line 28, in <module>
    _mask = _mask_fid['mask']
  File "/home/pi/.local/lib/python3.7/site-packages/numpy/lib/npyio.py", line 256, in __getitem__
    pickle_kwargs=self.pickle_kwargs)
  File "/home/pi/.local/lib/python3.7/site-packages/numpy/lib/format.py", line 770, in read_array
    array = numpy.ndarray(count, dtype=dtype)
numpy.core._exceptions.MemoryError: Unable to allocate 890. MiB for an array with shape (933120000,) and data type bool

Is there any way for this package to only load parts of the globe or otherwise reduce the required memory? I'm only interested in a specific "rectangle" on the globe.

Coordinates in land accused of being in the ocean

This coordinates (lat: -3.730884, lon: -38.457926) were
accused of being in the ocean
https://www.google.com/maps/place/3°43'51.2"S+38°27'28.5"W/@-3.7308889,-38.4601054,17z/data=!3m1!4b1!4m5!3m4!1s0x0:0x0!8m2!3d-3.730884!4d-38.457926?hl=pt

lat = -3.730884
lon = -38.457926
is_on_land = globe.is_land(lat, lon)
is_in_ocean = globe.is_ocean(lat, lon)
print(f"is_on_land: {is_on_land}")
print(f"is_in_ocean: {is_in_ocean}")

Results:
is_on_land: False
is_in_ocean: True