Fwd: Re: [Proj] PROJ 4.5.0 beta (and Krovak!)

[Gerald I. Evenden]

I did not do a "reply all" so only Frank got the original reply.

----------  Forwarded Message  ----------

Subject: Re: [Proj] PROJ 4.5.0 beta (and Krovak!)
Date: Wednesday 18 October 2006 11:38 am
From: "Gerald I. Evenden" <gerald.evenden at verizon.mail>
To: Frank Warmerdam <warmerdam at pobox.com>

On Tuesday 17 October 2006 4:41 pm, you wrote:
> Gerald I. Evenden wrote:
> > I do not particularly care what the projection is called but I would
> > appreciate some communications as to using the 'kocc' designation so I
> > could at least change my releases of libproj4 to conform with external
> > usage.

I thought I had previously apologized for my senior moment about the 'krovak'
designation and that +proj=kocc is in libproj4.

> Gerald,
>
> OK, I've gone back over this, thinking to modify PROJ.4 to support either
> name (proj=krovak or proj=kocc) similarly.  However, I've come to the
> conclusion there must be some other differences in the defaults,
> orientation or something for these two implementations of the projection.
>
> In particular, for a test point at 10E, 45N, lproj seems to fail:
>
> warmerda at gdal2200[917]% ./lproj +proj=kocc +ellps=bessel
> 10 45
> nan     -inf

Indeed, I get the same answers with the above run line.

> While proj produces results, though I don't know how correct they are:
>
> warmerda at gdal2200[1926]% ./proj +proj=krovak +ellps=bessel
> 10 45
> -1164802.15     -1546506.69
>
> I don't really know any test points for Krovak or even what options
> are required in the two implementations.
>
> For now, I'll stick with the existing Krovak code for the 4.5.0 release,
> but if there are things we can do to make the two implementations
> compatible with the aim of a seamless transition to use of libproj4
> in the future, I'd be all for that!
>
> PS. I see PJ_krovak was checked in sometime in 2002.
>
> Best regards,

I inserted the 10 45 number into an old script:

#!/bin/sh
# This is a test script to check the correct execution
# of the Krovak projection
cat <<EOF >temp
35d42'35.2147 48d7'46.2973
10 45
EOF
echo "forward projection\n"
../lproj +proj=kocc +ellps=bessel +czech temp \
     +lon_0=42d30 +lat_0=49d30 \
     +lat_t=59d42\'42.69689 +lat_1=78d30 +k_0=.9999
../lproj +proj=kocc +ellps=bessel +czech -o temp \
     +lon_0=42d30 +lat_0=49d30 \
     +lat_t=59d42\'42.69689 +lat_1=78d30 +k_0=.9999 \

| ../lproj -I +proj=kocc +ellps=bessel +czech -i \

     +lon_0=42d30 +lat_0=49d30 \
     +lat_t=59d42\'42.69689 +lat_1=78d30 +k_0=.9999 -W6
#
echo "above should agree with input of\n"
cat temp
rm temp

which produces the results:

forward projection\n
-16643462.09	2422822.35
-15043707.84	7571503.40
35d42'35.214700"E	48d07'46.297300"N
10d00'00.000000"E	45d00'00.000000"N
above should agree with input of\n
35d42'35.2147 48d7'46.2973
10 45

--
Mythology is what grownups believe,
folklore is what they tell children and religion is both.
-- Cedric Whitman (1969),

-------------------------------------------------------