Hi Piotr.
Sorry for the delay in getting back to you...
On Jan 27, 2005, at 6:39 AM, Piotr Kamiński wrote:
> Hi Don,
>
> Don Dietrick wrote:
>
>> Hi Piotr,
>>
>> Thanks for all the updates, I'll try to get them into 4.6.2. I might
>> have to wait and do a quick 4.6.3 release, I'm juggling a few things
>> right now and need to get the 4.6.2 release out.
>
> I've attached a few additional files, which has changed:
> - 'Layer*' correction of 'removeable' to 'removable' property name.
> - 'ProjectionMenu' and 'I18N' - localization
Thanks, I'm integrating them in now...
>> On Jan 25, 2005, at 11:31 AM, Piotr Kamiński wrote:
>>
>>> A few additional questions:ła
>>> 1. UTMPoint was changed. Instead of Zone letter code there is N or
>>> S now. Why such a change? I though that zone letters was some kind
>>> of standard. I assume that you decided to use ONLY S or N letter.
>>> Am I right? There is erronous comment in UTMPoint constructor -
>>> "If the letter is something else instead of 'N' or 'S', 'N' is
>>> assumed." Method checkZone throws exception when something
>>> different than N or S is specified.
>>
>>
>> It turns out that N or S are the only valid zones for UTM. All the
>> other zone letters only pertain to MGRS. I had a reference that
>> blurred this line, which is the reason for the original design.
>
> I've just searched some resources and most of them use 'old' UTM Grid
> Zone Designator, based on MGRS notation. Only one or two publication
> use notation wihout any letter (eg. 16 3423432.0 mE 4234324.0 mN).
> There was a note that specifying hemisphre is important, but it was
> not explained how to do it. I think that changes in UTMPoint might be
> little confusing. In MGRS both S and N letters are used. Now if I see
> 16S 2343242 234324 I can't be sure if 'S' stands for MGRS zone or
> south hemishere. I think that GPS devices also use MGRS notation.
>
> Could you point me to references which exactly defines how UTM
> coordinates should be presented.
UTM coordinates have a zone number, along with a northing value and an
easting value. There also needs to be an hemisphere notation, N or S.
MGRS has the same zone number, a letter, plus a single coordinate that
combines the 100k subzone plus northing and easting.
>> Thanks for picking up on the comment, it's a result of hesitation on
>> my part, I was trying to decide whether it was better to throw an
>> exception or quietly interpret incorrect input. I decided to throw
>> an exception, but forgot to update the comment.
>>
> Exception is better, but what do you think about using two schemes of
> zone numbering - new one with hemisphere and old one with MGRS zone
> letter. You could add second constructor to support this.
The zone letter isn't used for UTM translations. I guess you're
saying that you'd like for the class to interpret a MGRS zone letter
for a N/S indication, but that doesn't seem right? 'N' and above is
north, 'M' and below is south for MGRS. How about if I add a static
method to MGRSPoint.MGRSZoneToUTMZone(char) that does the conversion,
you can call that in-line for the UTMPoint.
>>> 2. What about changing LatLonPoint from float to double type. I had
>>> asked about this some time ago but haven't found any change in
>>> source code. I saw reports on OpenMap's list about successful
>>> transformation to double. Maybe just such change is enough? Our
>>> changes in ProjMath (insufficient precision of float type)
>>> testifies against current, mostly float based, implementation of
>>> calculation algorithms.
>>
>>
>> It's not going to happen for the 4.6.2 release, but I've hammered out
>> a design in a different project (which uses OpenMap, too) that
>> preserves backward compatibility which providing a choice between
>> floats and doubles. The API change to the projection libraries will
>> be significant, (but still backward compatible) so I think a 4.7
>> version change will be proper.
>
>
> When we could expect 4.7 release ?
It really depends on funding, I'm thinking end of Spring.
- Don
--
[To unsubscribe to this list send an email to "majdart_at_bbn.com"
with the following text in the BODY of the message "unsubscribe openmap-users"]
Received on Tue Feb 01 2005 - 18:38:24 EST