[udig-devel] Guidance on how to implement Attributes fora FeatureReader

[Dan Eslinger]

Cory,

The first pass is an all/none but I like the idea to do first pass cursory 
look at a number of records to see what is present... As far as design, 
another approach would be to parse the first X kilobytes of the file and see 
which NMEA messages are in progress, and instead of displaying the data 
(which may be misleading given that the receiver may not be navigating when 
data collection starts (ie in someones garage or otherwise), the first 
records could have all zero or blank values, we could instead display the 
records that may be availible based on which records are in the dataset. 
Thoughts?

I was looking for best-practice and not in fact "how", so if anyone has some 
guidance in this area it will save code re-work in the future. Thanks,

- Dan

> Dan Eslinger wrote:
>
>> ...the plug-in I have developed is able to read GPS NMEA sentences and
>> return latitude and longitude ordinate pairs for access and display.
>>
>> What I would like some guidance on now is how to extend my class which
>> inherits from FeatureReader to add the attribute data that I am
>> reading in now. Any examples or pointers that you can provide or point
>> me to would be very valuable.
>>
>> If you are curious about my project, progress or the current design,
>> the project is in SVN at the following URL:
>> http://svn.geotools.org/udig/community/dan
>
> Dan, how were you planning on handling attribute import? An all/none
> selection or more finely grained?  On the UI side, it be cool to do
> something like the excel import wizard and show the first 5 features or
> so, at which point you could label and select attributes for inclusion.
> Presumably you're looking for more of a "best practice" idea for adding
> the attributes to the datastore, rather than "how"... ?
>
> Has anyone out there delved into this area in the past?
>
> Cheers,
> Cory.
>