> Jody Garnett ha scritto:
>>>>> b) alternatively, add a flag to control the DMS/degree behavior.
>>>> I would prefer to add a flag; even if the default value is true.
>>> Cool. Any reasoning behind the preference?
>> Yeah; we are making an assumption - I would like the ability for users to 
>> make the assumption the other way if they need to.
> 
> Errr... except that if the user does not set the flag he
> won't get the CRS to decode at all. Did you read the
> first part of the mail?

Yes I did.

Indeed if the user was in some kind of "strict" mode I would expect them not to 
be able to decode that CRS at all.

> If you try to get EPSG:2150 today you get an exception, period,
> in other words, it's not usable at all for any purpose.

Agreed; and that would be a good thing would it not?

> So a meaningful description for the flag would be:
> - when true, the codes do work and we assume decimal degree
>  instead of DMS
> - when false, the codes using DMS don't work, but they may
>  start working if we introduce the necessary transformation
>  code in the future
> 

Correct; I would phrase it as.
When true we assume that DMS entries are understood to be represented as 
decimal degrees; this assumption is shared by many applications but may not be 
correct for your dataset.
When false we carefully do not assume DMS entries are decimal deegrees; this 
limits the number of codes supported

This flag would even still be useful when/if we do support DMS (as chances are 
a lot of data is actually recodded in decimal deegrees).

Andrea this is just my preference (to have a flag) I would also settle for a 
warning in the javadocs.

> The flag is at least more honest than just flipping though,
> today you get an exception and you know you cannot work with
> those codes, if we just flip the situation without a flag
> people that have storage in the format recommended by EPSG
> (see back my first mail) will start seeing odd errors because
> the first digits after the decimal separator have a different
> meaning in DMS. I don't think it's of any likeliness that anybody
> uses that format (never stumbled into any GIS using it), but
> just in case, the flag would be a sort of documentation
> about the assumptions we made.

Agreed.
Jody
------------------------------------------------------------------------------
ThinkGeek and WIRED's GeekDad team up for the Ultimate 
GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the 
lucky parental unit.  See the prize list and enter to win: 
http://p.sf.net/sfu/thinkgeek-promo
_______________________________________________
Geotools-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geotools-devel

Reply via email to