[API-users] How to query genus names with the GBIF API

Fri Sep 8 22:34:38 CEST 2017

Dear Markus,

Thank you for the suggestions. I tried the Scientific name search under 
Occurence search. It is really promising and worth considering to go 
that direction. Why isn't it implemented at species search?

József

2017-09-07 13:36 keltezéssel, Markus Döring írta:
> Give it a try then. Unless you have a huge user base this should not 
> be overloading GBIF.
> We have decided to rather use a single field for the entire name and 
> do an autocomplete on that.
> Maybe this approach also works for you, then you could also just use 
> the suggest as it is.
> See the name filter here (type into "search" under Scientific Name):
> https://www.gbif.org/occurrence/search
>
> Best,
> Markus
>
>
>> On 7. Sep 2017, at 12:12, Szlamka József <szlajozs at gmail.com 
>> <mailto:szlajozs at gmail.com>> wrote:
>>
>> Dear Markus,
>>
>> My idea is that based on the validation services provided by GBIF to 
>> implement a search form with similar functionality (autocomplete, 
>> etc.) as the "Search for scientific names" form at Catalogue of life:
>> http://www.catalogueoflife.org/col/search/scientific
>>
>> What is your opinion, is this way with the GBIF suggest api plus 
>> client side filtering a proper way? Considering the possible number 
>> of unnecessary result records (everything where kingdomKey<>6) and 
>> the number of unnecessary fields in each record (everything except 
>> the GENUS field), wouldn't that be an unacceptable overload at GBIF 
>> side and performance / bandwidth issues at the requestor webapp side?
>>
>> Cheers:
>> József
>>
>>
>> 2017-09-07 11:47 keltezéssel, Markus Döring írta:
>>> this is unfortunately not possible.
>>> You will have to filter them clientside checking for records with 
>>> kingdomKey=6 or kingdom="Plantae"
>>>
>>> Markus
>>>
>>>
>>>> On 7. Sep 2017, at 11:22, Szlamka József <szlajozs at gmail.com 
>>>> <mailto:szlajozs at gmail.com>> wrote:
>>>>
>>>>
>>>> Dear Markus,
>>>>
>>>> Thank you very much. Do you have any idea, using the "suggest" API 
>>>> would it be possible somehow to filter the results for the Plantae 
>>>> kingdom only?
>>>>
>>>> József
>>>>
>>>>
>>>> 2017-09-07 10:17 keltezéssel, Markus Döring írta:
>>>>> Hi József,
>>>>>
>>>>> the GBIF API hardly anywhere accepts wildcards. Apart from fuzzy 
>>>>> matching entire words the only place you can use is the suggest 
>>>>> API which should help you with your problem.
>>>>>
>>>>> It is a kind of autocomplete and you have to match the beginning 
>>>>> of a word. It wont match in the middle or end. It does match on 
>>>>> all words or tokens of a string, so for a species name it will 
>>>>> match the genus part, the epithet or even the authorship. Results 
>>>>> are prioritized with author based matches coming last and higher 
>>>>> ranks being more important.
>>>>>
>>>>> Here is an example:
>>>>> http://api.gbif.org/v1/species/suggest?datasetkey=d7dddbf4-2cf0-4f39-9b2a-bb099caae36c&rank=GENUS&q=Ado
>>>>>
>>>>> The suggest API only allows you to retrieve 100 records though, 
>>>>> but you can page through results up to 100.000 records max:
>>>>> http://api.gbif.org/v1/species/suggest?datasetkey=d7dddbf4-2cf0-4f39-9b2a-bb099caae36c&rank=GENUS&q=Ado&limit=100
>>>>> -> returns first 100 genera starting with Ado
>>>>>
>>>>> http://api.gbif.org/v1/species/suggest?datasetkey=d7dddbf4-2cf0-4f39-9b2a-bb099caae36c&rank=GENUS&q=Ado&limit=100&offset=100
>>>>> -> returns 49 more genera starting with Ado
>>>>>
>>>>> http://api.gbif.org/v1/species/suggest?datasetkey=d7dddbf4-2cf0-4f39-9b2a-bb099caae36c&rank=GENUS&q=Ado&limit=100&offset=200
>>>>> -> returns 0 genera as expected
>>>>>
>>>>>
>>>>> Cheers,
>>>>> Markus
>>>>>
>>>>>
>>>>>
>>>>>> On 7. Sep 2017, at 08:36, Szlamka József <szlajozs at gmail.com 
>>>>>> <mailto:szlajozs at gmail.com>> wrote:
>>>>>>
>>>>>> Dear Scott,
>>>>>>
>>>>>> Many thanks for your answer. I also added the "&highertaxonKey=6" 
>>>>>> param to narrow the results for plant genera only.
>>>>>>
>>>>>> To be more precise what I exactly wish to achieve is as follows:
>>>>>>
>>>>>> 1./
>>>>>> I only want to search within the "Plantae" kingdom.
>>>>>>
>>>>>> 2./
>>>>>> Say, I look for the GENUS name "Adonis" but I only remember that 
>>>>>> the name starts with "Ado".
>>>>>>
>>>>>> 3./
>>>>>> So my ideal API call would run a wildcard search within PLANT 
>>>>>> GENERA names for "Ado*" and would result in a list of ALL 
>>>>>> (without limit) plant genera names starting with "Ado*".
>>>>>>
>>>>>> I played with the "/species/match", "/species/search" and 
>>>>>> "/species/suggest" sub-apis and with the "q" and "name" params. 
>>>>>> They don't accept wildcards and don't give the expected results 
>>>>>> for name chunks without wildcards.
>>>>>>
>>>>>> Now my only hope is that I left out some constellations or 
>>>>>> misused something.
>>>>>>
>>>>>> József
>>>>>>
>>>>>>
>>>>>> 2017-09-06 23:21 keltezéssel, Scott Chamberlain írta:
>>>>>>> Have you tried something like
>>>>>>>
>>>>>>> http://api.gbif.org/v1/species/search?datasetkey=d7dddbf4-2cf0-4f39-9b2a-bb099caae36c&rank=GENUS
>>>>>>>
>>>>>>> where datasetkey=d7dddbf4-2cf0-4f39-9b2a-bb099caae36c is for the 
>>>>>>> GBIF backbone taxonomy
>>>>>>>
>>>>>>> Scott
>>>>>>>
>>>>>>> On Wed, Sep 6, 2017 at 2:10 PM Szlamka József 
>>>>>>> <szlajozs at gmail.com <mailto:szlajozs at gmail.com>> wrote:
>>>>>>>
>>>>>>>     Hello to all list members,
>>>>>>>
>>>>>>>     I'm new on this list and also new to the GBIF API. I'd like
>>>>>>>     to get a
>>>>>>>     list of all the genus names occuring in the GBIF database.
>>>>>>>     After several
>>>>>>>     hours of trying with the API, and no success, I found this
>>>>>>>     mailing list.
>>>>>>>     I hope that API guru members on the list can show me the
>>>>>>>     right way or
>>>>>>>     tell if this is not possible at all.
>>>>>>>
>>>>>>>     Thank you in advance.
>>>>>>>
>>>>>>>     _______________________________________________
>>>>>>>     API-users mailing list
>>>>>>>     API-users at lists.gbif.org <mailto:API-users at lists.gbif.org>
>>>>>>>     https://lists.gbif.org/mailman/listinfo/api-users
>>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> API-users mailing list
>>>>>> API-users at lists.gbif.org <mailto:API-users at lists.gbif.org>
>>>>>> https://lists.gbif.org/mailman/listinfo/api-users
>>>>>
>>>>
>>>
>>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.gbif.org/pipermail/api-users/attachments/20170908/5cf6b60f/attachment-0001.html>