BREAKING CHANGE: occurrences.search, occurrences.count, occurrences.count_datasets, and maps.map now default to using Catalogue of Life (COL) taxonomy instead of GBIF Backbone Taxonomy. Set checklistKey=None to use GBIF backbone.

added support for alphanumeric taxonomy keys (e.g., kingdomKey, phylumKey, etc.) in occurrences.search to support COL’s alphanumeric identifiers

added automatic detection of numeric taxonomy keys in occurrences.search: when numeric (integer) keys are provided without an explicit checklistKey, the function now automatically switches to GBIF Backbone taxonomy and issues a deprecation warning, encouraging migration to COL alphanumeric keys

preserve pygbif.search as an alias for occurrences.search and expose literature, collection, and institution search functions through their module namespaces

added species.gbif_to_col() function to convert GBIF Backbone numeric taxonomy keys to COL Extended Release alphanumeric keys, facilitating migration from GBIF Backbone to COL

species.name_lookup(), species.name_suggest(), and species.name_usage() now issue DeprecationWarning because they primarily use the outdated GBIF Backbone Taxonomy; use the datasetKey parameter to query another checklist

added support for checklistKey parameter in occurrences.download to specify taxonomy for download content. When provided, uses a specific checklist from ChecklistBank instead of the GBIF Backbone Taxonomy. See https://www.gbif.org/developer/occurrence#download for more information.

added support for occurrences.download_citation 158

added support for occurrences.download_sql 157

added support for occurrences.download_describe 142

added support for literature.search 149

fixed a bug in building the documentation with readthedocs 138, 129

tests now run against live GBIF APIs 101, 128

updated caching.py since the remove_expired_responses method is deprecated. 126

added support for predicates: isNull, isNotNull, in and not 92, 102 and 103

added support for nested queries/dictionaries 104

deprecated the add_predicate function and added add_pred_dict to accomodate for newly supported predicates to ensure that the arguments that are sent are added in the payload function 108

added support for multiple download formats 105

updated operators and look-up tables 107

included documentation on newly supported predicates and dictionaries 106

update to fix requesting GBIF downloads

minor documentation updates 95 and 99

update to fix broken dependencies 93

minor documentation updates

Fix for occurrences.download when giving geometry as a string rather than using add_geometry; predicates were being split on whitespace, which doesn’t work for WKT 81 84

Moved to using the logging module instead of print() for giving information on occurrence download methods 78

Clarify that occurrences.count for length 1 inputs only; see occurrences.search for > 1 value 75 77

Improved documentation for species.name_usage method, mostly for the language parameter 68

Gains download method download_cancel for cancelling/deleting a download request 59

occurrences.search now supports recordedByID and identifiedByID search parameters 62

clean up the Contributing file, thanks @niconoe 64

clean up internal imports in the library, thanks @niconoe 65

fix usage of is and ==, was using them inappropriately sometimes (via https://realpython.com/python-is-identity-vs-equality/), 69

remove redundant parameter in a doc string, thanks @faroit 71

make a test for internal fxn gbif_GET_write more general to avoid errors if GBIF changes content type response header slightly 72

changed base url to https for all requests; was already https for maps and downloads in previous versions

occurrences, species, and registry modules gain docstrings with brief summary of each method

pygbif gains ability to cache http requests. caching is off by default. See ?pygbif.caching for all the details 52 56 via @nleguillarme

made note in docs that if you are trying to get the same behavior as the GBIF website for name searching, species.name_backbone is likely what you want 55 thanks @qgroom

for parameters that expect a bool, convert them to lowercase strings internally before doing HTTP requests

pygbif is Python 3 only now 19

Gains maps module with maps.map method for working with the GBIF maps API 41 49

Gains new module utils with one method wkt_rewind 46 thanks @aubreymoore for the inspiration

Fixed bug in registry.installations: typo in one of the parameters identifierTyp instead of identifierType 48 thanks @data-biodiversity-aq

Link to GitHub issues from Changelog 🎉

Fix a occurrence download test 47

Much more thorough docs 25

Download methods much improved 16 27 thanks @jlegind @stijnvanhoey @peterdesmet !

MULTIPOLYGON now supported in geometry parameter 35

Fixed docs for occurrences.get, and occurrences.get_verbatim, occurrences.get_fragment and demo that used occurrence keys that no longer exist in GBIF 39

Added organizations method to registry module 12

Added remainder of datasets methods: registry.dataset_search (including faceting support 37) and registry.dataset_suggest, for the /dataset/search and /dataset/suggest routes, respectively 40

Added remainder of species methods: species.name_lookup (including faceting support 38) and species.name_usage, for the /species/search and /species routes, respectively 18

Added more tests to cover new methods

Changed species.name_suggest to give back data stucture as received from GBIF. We used to parse out the classification data, but for simplicity and speed, that is left up to the user now.

start parameter in species.name_suggest, occurrences.download_list, registry.organizations, registry.nodes, registry.networks, and registry.installations, changed to offset to match GBIF API and match usage throughout remainder of pygbif

Added many new occurrence.search parameters, including repatriated, kingdomKey, phylumKey, classKey, orderKey, familyKey, genusKey, subgenusKey, establishmentMeans, facet, facetMincount, facetMultiselect, and support for facet paging via **kwargs 30 34

Fixes to **kwargs in occurrence.search so that facet parameters can be parsed correctly and requests GET request options are collected correctly 36

Added spellCheck parameter to occurrence.search that goes along with the q parameter to optionally spell check full text searches 31

Added variable types throughout docs

Changed default limit value to 300 for occurrences.search method

tox now included, via @xrotwang 20

Added more registry methods 11

Started occurrence download methods 16

Added more names methods 18

All requests now send user-agent headers with requests and pygbif versions 13

Bug fix for occurrences.download_get 23

Fixed bad example for occurrences.get 22

Fixed wheel to be universal for 2 and 3 10

Improved documentation a lot, autodoc methods now

Fixed distribution for pypi

First release