Fwd: [Biogeosdi] Re: Search on the GBIF REST services
Javier de la Torre
jatorre at gmail.com
Tue Jun 26 23:00:20 CEST 2007
Begin forwarded message:
> From: Donald Hobern <dhobern at gbif.org>
> Date: 26 de junio de 2007 22:47:00 GMT+02:00
> To: Javier de la Torre <jatorre at gmail.com>
> Cc: Tim Sutton <tim at linfiniti.com>, biogeosdi at lists.tdwg.org
> Subject: Re: [Biogeosdi] Re: Search on the GBIF REST services
>
> The tests we perform correspond to the following bits:
>
> /**
> * Set if latitude appears to have the wrong sign
> */
> public static final int
> GEOSPATIAL_PRESUMED_NEGATED_LATITUDE = 0x01;
> /**
> * Set if longitude appears to have the wrong sign
> */
> public static final int
> GEOSPATIAL_PRESUMED_NEGATED_LONGITUDE = 0x02;
> /**
> * Set if latitude and longitude appears to have been switched
> */
> public static final int
> GEOSPATIAL_PRESUMED_INVERTED_COORDINATES = 0x04;
> /**
> * Set if coordinates are (0,0)
> */
> public static final int
> GEOSPATIAL_ZERO_COORDINATES = 0x08;
> /**
> * Set if coordinates not in range -90 to 90 and -180 to 180
> */
> public static final int
> GEOSPATIAL_COORDINATES_OUT_OF_RANGE = 0x10;
> /**
> * Set if coordinates do not match country name (covers
> cases not handled above)
> */
> public static final int
> GEOSPATIAL_COUNTRY_COORDINATE_MISMATCH = 0x20;
>
> Donald
>
> Javier de la Torre wrote:
>> Buff... that is more than enough!!
>> I suppose you also drop latitude=longitude=0 right?
>>
>> I think we have to change our report on this topic jeje.
>>
>> Cheers.
>>
>> On 26/06/2007, at 22:13, Donald Hobern wrote:
>>
>>> Right now the dodginess is only detected for records where we
>>> interpret the Country name as a particular country and also have
>>> coordinates and the coordinates do not intersect the country. In
>>> many cases we actually also determine that the latitude is
>>> negated, the longitude is negated and/or the latitude and
>>> longitude are transposed.
>>>
>>> Donald
>>>
>>> Tim Sutton wrote:
>>>> Hi
>>>>
>>>> Great stuff that we can now exclude dodgy / non existant
>>>> occurrences -
>>>> we could probably make th e second (coordinate issues) option user
>>>> settable (or at the least a setting in our global config file).
>>>>
>>>> Regards
>>>>
>>>> Tim
>>>>
>>>> 2007/6/26, Javier de la Torre <jatorre at gmail.com>:
>>>>> Great,
>>>>>
>>>>> I will then uncomment the code :)
>>>>>
>>>>> Javier.
>>>>>
>>>>> On 26/06/2007, at 17:24, Donald Hobern wrote:
>>>>>
>>>>> > Thanks, Javier - found the problem - will get the fix deployed
>>>>> > tomorrow.
>>>>> >
>>>>> > Best wishes,
>>>>> >
>>>>> > Donald
>>>>> >
>>>>> > Javier de la Torre wrote:
>>>>> >> Great Donald,
>>>>> >>
>>>>> >> We have now some comments to remove on the report about the
>>>>> GBIF
>>>>> >> REST service ;)
>>>>> >>
>>>>> >> Still I have an issue when using the &coordinateissues=false
>>>>> >> parameter
>>>>> >>
>>>>> >> This URL returns 195 records:
>>>>> >> http://data.gbif.org/ws/rest/occurrence/list?
>>>>> >> format=brief&scientificname=Puma+concolor
>>>>> >>
>>>>> >> This one none:
>>>>> >> http://data.gbif.org/ws/rest/occurrence/list?
>>>>> >> format=brief&coordinateissues=false&scientificname=Puma
>>>>> +concolor
>>>>> >>
>>>>> >>
>>>>> >> And I suppose not all coordinates are wrong no?
>>>>> >>
>>>>> >> Cheers.
>>>>> >>
>>>>> >>
>>>>> >> On 26/06/2007, at 14:29, Donald Hobern wrote:
>>>>> >>
>>>>> >>> Dear Javier,
>>>>> >>>
>>>>> >>> The structures are now stable. Sorry not to have copied
>>>>> you on
>>>>> >>> my emails on what was changing. Adding
>>>>> "&coordinatestatus=true"
>>>>> >>> should limit responses to those with coordinates. You can
>>>>> also
>>>>> >>> use "&coordinateissues=false" to exclude records where we have
>>>>> >>> detected that the coordinates do not match the supplied
>>>>> country
>>>>> >>> name.
>>>>> >>>
>>>>> >>> Donald
>>>>> >>>
>>>>> >>> Javier de la Torre wrote:
>>>>> >>>> Hey Donald,
>>>>> >>>>
>>>>> >>>> I am fixing some stuff on the prototype we did in Brasil
>>>>> and I
>>>>> >>>> had to change my code to reflect the new format that the rest
>>>>> >>>> services are sending now. Is this more or less stable now?
>>>>> >>>>
>>>>> >>>> And another question, would be possible to ask the service to
>>>>> >>>> return only occurrences with coordinates. For us is not of
>>>>> much
>>>>> >>>> use occurrences without coordinates.
>>>>> >>>>
>>>>> >>>> Thats the query I am trying
>>>>> >>>>
>>>>> >>>> http://newportal.gbif.org/ws/rest/occurrence/list?
>>>>> >>>>
>>>>> format=brief&minLatitude=-90&maxLatitude=90&scientificname=Puma
>>>>> >>>> +concolor
>>>>> >>>>
>>>>> >>>> Thanks in advance.
>>>>> >>>>
>>>>> >>>> Javier.
>>>>> >>>>
>>>>> >>>>
>>>>> >>>>
>>>>> >>>
>>>>> >>> --
>>>>> >>> ------------------------------------------------------------
>>>>> >>> Donald Hobern (dhobern at gbif.org)
>>>>> >>> Deputy Director for Informatics Global Biodiversity
>>>>> Information
>>>>> >>> Facility Secretariat Universitetsparken 15, DK-2100
>>>>> Copenhagen,
>>>>> >>> Denmark
>>>>> >>> Tel: +45-35321483 Mobile: +45-28751483 Fax: +45-35321480
>>>>> >>> ------------------------------------------------------------
>>>>> >>>
>>>>> >>
>>>>> >>
>>>>> >>
>>>>> >
>>>>> > --
>>>>> > ------------------------------------------------------------
>>>>> > Donald Hobern (dhobern at gbif.org)
>>>>> > Deputy Director for Informatics Global Biodiversity Information
>>>>> > Facility Secretariat Universitetsparken 15, DK-2100 Copenhagen,
>>>>> > Denmark
>>>>> > Tel: +45-35321483 Mobile: +45-28751483 Fax: +45-35321480
>>>>> > ------------------------------------------------------------
>>>>> >
>>>>>
>>>>> _______________________________________________
>>>>> biogeosdi mailing list
>>>>> biogeosdi at lists.tdwg.org
>>>>> http://lists.tdwg.org/mailman/listinfo/biogeosdi
>>>>>
>>>>
>>>>
>>
>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.tdwg.org/pipermail/tdwg-content/attachments/20070626/7702970c/attachment.html
More information about the tdwg-content
mailing list