Half-sized grid near meridian return wrong coordinates

[solsticedhiver]

Just look at that example: 31TBN73
This 10x10km MGRS zone is near the prime meridian. It is not a complete full size zone and its size is divided roughly by 2.

When you run:

> mgrs.inverse('31TBN73')
[-0.035799295305842804, 47.18325044910646, 0.0910966355893934, 47.276538500931856]

The returned coordinate are the one for a full sized zone. whereas the first longitude should be 0°.

One can find this error for all zones that are near a meridian of an UTM zone where the MGRS zone are half-sized.

On this fiddle https://jsfiddle.net/a8830fcv/4/, one can see in black the correct 31TBN73 and in red the portion of the zone that is wrong, crossing over the prime meridian.

The same is true for example for 30TYT21 and so on...

[DanielJDufour]

Great catch @solsticedhiver . A new maintainer here. I'll look into this! Thanks!

[DanielJDufour]

I looked into this more and I can confirm your issue.

I downloaded the relevant and authoritative MGRS 1km grid zone file from http://earth-info.nga.mil/GandG/update/index.php?dir=coordsys&action=mgrs-1km-polyline-dloads . I loaded the file into QGIS and it does appear that the zones cut off at the prime meridian.

I should also note that earthpoint.us appears to have it wrong as well.