[Proj] Central Meridian (lon_0) sign

[Richard didier]

Hi all,

When looking at the epsg file ($PROJ_HOME/share/proj) for the french Lambert I 
noticed that lon_0 is negative whilst the longitude is at east of Greenwich.
The following tests show that something is wrong with the negative value :

Transformation between 2 lamberts :
cs2cs -f "%.2f" +init=epsg:27561 +to +init=epsg:27572 \
<<EOF
750000.00 300000.00
EOF
gives:
750283.12       2600360.77 0.00
which is correct.

Transformation between a lambert and its base geographic system :
cs2cs +init=epsg:27561 +to +init=epsg:4807 \
<<EOF
1029705.083 272723.849
EOF
gives:
3d39'45.975"E   50dN 0.000
which is wrong, it should be :
8d20'14.025"E   50dN 0.000

Transformation between a geographic system and its lambert :
cs2cs -f "%.3f" +init=epsg:4807 +to +init=epsg:27562 \
<<EOF
 2d45'43.846"E 46d37'35.430"N
EOF
gives:
990233.155      193379.657 0.000
which is wrong, it should be :
632542.063      180804.162 0.000

I obtain the expected correct values by changing the sign of lon_0 (from 
-2.33722917 to 2.33722917).

Any idea ?
-- 
Didier RICHARD, chef de projet "Diffusion des données numériques" 
Institut Geographique National           tel: +33/0 1.43.98.83.23
2/4 Avenue Pasteur                       fax: +33/0 1.43.98.81.71
F-94165 St-Mande Cedex                e-mail: Didier.Richard at ign.fr