GBIF API improvements requests #32
Description
GBIF actual query example :
https://api.gbif.org/v1/occurrence/search?
limit=300&
maxPage=10&
geometry=POLYGON+%28%288.123085024113518+46.127596816081585%2C+8.094327892683024+46.91184965940513%2C+7.838609326480127+47.676941759022455%2C+7.356410459939478+48.392123204105445%2C+6.658747051521441+49.027352044634334%2C+5.768324639335576+49.55479065881764%2C+4.7197936397149896+49.95042917457144%2C+3.558717869411293+50.195686154307204%2C+2.3390720000000003+50.27878409331426%2C+1.1194261305887072+50.195686154307204%2C+-0.04164963971498929+49.95042917457144%2C+-1.0901806393355742+49.55479065881764%2C+-1.9806030515214403+49.027352044634334%2C+-2.6782664599394783+48.392123204105445%2C+-3.1604653264801272+47.676941759022455%2C+-3.416183892683025+46.91184965940513%2C+-3.4449410241135188+46.127596816081585%2C+-3.2551630297982697+45.35435490732322%2C+-2.862439778766704+44.62066626278488%2C+-2.28789863310355+43.95261392702393%2C+-1.5568134689118351+43.3731803009173%2C+-0.6974783728131996+42.90175815862464%2C+0.2596774389590625+42.55378293579951%2C+1.2827816236584904+42.34046362420427%2C+2.339071999999999+42.26859790668574%2C+3.3953623763415073+42.34046362420427%2C+4.418466561040936+42.55378293579951%2C+5.375622372813197+42.90175815862464%2C+6.234957468911833+43.3731803009173%2C+6.966042633103547+43.95261392702393%2C+7.5405837787667025+44.62066626278488%2C+7.933307029798267+45.35435490732322%2C+8.123085024113518+46.127596816081585%29%29&
dateMin=null&
dateMax=null&
class=null&
offset=1800
It returns a lot of unused fields/columns that weighs down the downloaded results, while we only use a few columns.
It would be great to have an option in GBIF API to select and limit returned fields of the API.
In our case, we would just need the taxonKey, scientific and verncaular names, observation date, taxon class.
It would also be useful to be able to order results, as an example by dates, so we would be certain to return most recent observations when we limit results to 3000 occurrences.
We would also need to be able to get vernacular names by language in Occurrence search API. Combined to previous point, we could choose only the language required.
Also actually we get a list of observations and then we have to agregate it locally to generate a list of species.
It would be a lot better to directly get a list of species instead of occurrences list that we then have to agregate on our side.
Because actually we do big requests to API, with big results to download, big calculation to do on our side, and need to limit results so partially false.
This one seems to exists (https://www.gbif.org/api/occurrence/breakdown?dimension=month&fillEnums=true&limit=12&locale=fr&offset=0&secondDimension=&taxon_key=2704179) but is not public and potentially deprecated : gbif/tech-docs#24.
It would also be useful to have dataset names in Occurrences API, and not only their UUID, to list Dataset names easily, such as detailed in #33 (comment)