id	summary	reporter	owner	description	type	status	priority	milestone	component	version	resolution	keywords	cc
12186	extend projection support in josm core	bastiK	team	"We have essentially all the ingredients for a fully fledged projection library, but only a modest number of projections in our database.

It should be possible to include a lot more without too much work.

I've tried the current epsg file of the proj.4 project (5006 entries), here are some preliminary results:
* At the moment we should be able to support 4503 of 5006 entries. The rest doesn't work (right now), because:
  * unsupported base projection (""code""=""nr. of entries""): aea=31, laea=16, cea=2, cass=20, poly=5, merc=19, nzmg=1, stere=38, krovak=4, omerc=20, eqc=5
  * relies on NAD27 datum, which requires a shift database for conversion: 204 entries
  * relies on various other shift databases (`+geoidgrids=*`): 5 entries
  * `+proj=geocent` ""projection"" converts to 3D (X,Y,Z) - not useful in JOSM: 138 entries
* 4466 of 4503 potentially supported entries have been tested against proj.4 and work correctly
  * 15 of the remaining 37 entries are untested, because I couldn't extract bounds information for them (EPSG:4296, EPSG:21817, EPSG:23853, EPSG:23886, EPSG:29118, EPSG:29119, EPSG:29120, EPSG:29121, EPSG:29177, EPSG:29178, EPSG:29179, EPSG:29180, EPSG:29635, EPSG:29636, EPSG:31461)
  * for the rest (22 entries) there is a mismatch between JOSM and proj.4. (EPSG:2155, EPSG:2156, EPSG:2577, EPSG:2636, EPSG:2694, EPSG:2754, EPSG:3143, EPSG:3389, EPSG:3390, EPSG:3460, EPSG:3477, EPSG:3851, EPSG:5925, EPSG:5930, EPSG:5935, EPSG:6086, EPSG:6093, EPSG:6120, EPSG:6122, EPSG:6403, EPSG:26940, EPSG:26979) [[br]] There are two reasons for this:
     1. Problems when the central meridian is near 180 degrees longitude.
     2. The Lambert Conformal Conic projection breaks when two of the standard parallels are the same.
"	enhancement	closed	normal	16.02	Core		fixed	projection	wiktorn