Пример #1
0
 def pull(self):
     """
     Pull all observations from iNaturalist.
     If we don't have an observation with that iNaturalist ID, create one (check the vespawatch-evidence field to
     determine whether we should create a Nest or an Individual)
     If we do have an observation with that iNaturalist ID, update it.
     """
     self.w("\n3. Pull all observations from iNaturalist")
     observations = get_all_observations(
         params={'project_id': settings.VESPAWATCH_PROJECT_ID})
     pulled_inat_ids = []
     for inat_observation_data in observations:
         pulled_inat_ids.append(inat_observation_data['id'])
         local_obs = get_local_observation_with_inaturalist_id(
             inat_observation_data['id'])
         if local_obs is None:
             # This is a new one. Check vespawatch-evidence and create a nest or individual
             self.w(
                 f"iNaturalist observation with ID #{inat_observation_data['id']} is not yet known, we'll create it locally...",
                 ending='')
             create_observation_from_inat_data(
                 inat_observation_data
             )  #TODO: check all the required fields are set
             self.w("OK")
         else:
             # We already have an observation for this id. Update it
             self.w(
                 f'updating observation {type(local_obs).__name__} {local_obs.pk}'
             )
             local_obs.update_from_inat_data(inat_observation_data)
     return pulled_inat_ids
Пример #2
0
 def pull(self):
     """
     Pull all observations from iNaturalist.
     If we don't have an observation with that iNaturalist ID, create one (check the vespawatch-evidence field to
     determine whether we should create a Nest or an Individual)
     If we do have an observation with that iNaturalist ID, update it.
     """
     self.w(
         "\n3. Pull all observations from iNaturalist (based on the project)"
     )
     observations = get_all_observations(
         params={'project_id': settings.VESPAWATCH_PROJECT_ID})
     # The most recent observations are the more important one.
     # Let's start with those in case the sync process is interrupted (that unfortunately happens)
     observations = sorted(observations,
                           key=lambda x: x['id'],
                           reverse=True)
     pulled_inat_ids = []
     for inat_observation_data in observations:
         pulled_inat_ids.append(inat_observation_data['id'])
         local_obs = get_local_observation_with_inaturalist_id(
             inat_observation_data['id'])
         if local_obs is None:
             # This is a new one. Check vespawatch-evidence and create a nest or individual
             self.w(
                 f"iNaturalist observation with ID #{inat_observation_data['id']} is not yet known, we'll create it locally...",
                 ending='')
             create_observation_from_inat_data(
                 inat_observation_data
             )  #TODO: check all the required fields are set
             self.w("OK")
         else:
             # We already have an observation for this id. Update it
             self.w(
                 f'updating observation {type(local_obs).__name__} {local_obs.pk}'
             )
             local_obs.update_from_inat_data(inat_observation_data)
     return pulled_inat_ids
Пример #3
0
def test_get_all_observations(sleep, requests_mock):
    # Make response appear as though there are more pages
    page_1 = load_sample_data("get_observations_node_page1.json")
    page_1["total_results"] = PER_PAGE_RESULTS + 1
    page_2 = load_sample_data("get_observations_node_page2.json")

    requests_mock.get(
        urljoin(INAT_NODE_API_BASE_URL, "observations"),
        [
            {
                "json": page_1,
                "status_code": 200
            },
            {
                "json": page_2,
                "status_code": 200
            },
        ],
    )

    observations = get_all_observations(id=[57754375, 57707611])

    assert type(observations) is list
    assert len(observations) == 2
Пример #4
0
    def handle(self, *args, **options):
        self.w(
            f"Will import observations from project #{settings.VESPAWATCH_PROJECT_ID}..."
        )

        observations = get_all_observations(params=PULL_CRITERIA)

        total_count = len(observations)
        success_count, from_us_count, from_inat_total_count, from_inat_nest_count = 0, 0, 0, 0

        self.w(
            "Step 1: Will remove from our database all observations that originates from iNaturalist (they'll be recreated right after)...",
            ending="")
        Individual.from_inat_objects.all().delete()
        Nest.from_inat_objects.all().delete()
        self.w(self.style.SUCCESS("OK"))

        self.w(
            f"Step 2: Will parse the {total_count} matching observations received from iNaturalist."
        )
        for inat_observation_data in observations:
            inat_id = inat_observation_data['id']

            self.w(f"Now processing iNaturalist observation #{inat_id}...",
                   ending="")

            # TODO: it happened once during developpement that inat search (via Node API) returned...
            # TODO: ...a recently deleted occurrence, hence an ObservationNotFound raised in...
            # TODO: ...inat_observation_comes_from_vespawatch()
            if inat_observation_comes_from_vespawatch(
                    inat_observation_data['id']):
                from_us_count = from_us_count + 1
                self.w("Observation initially comes from Vespa-Watch. ",
                       ending="")
                # The only thing we do is updating the identification, if needed.
                try:
                    r = update_loc_obs_taxon_according_to_inat(
                        inat_observation_data)

                    if r == 'no_community_id':
                        self.w(
                            "There's no community ID, so we keep what we have. ",
                            ending="")
                    elif r == 'matching_community_id':
                        self.w(
                            "The community ID agree with us, so we keep what we have. ",
                            ending="")
                    elif r == 'updated':
                        self.w("Database updated to match the community ID! ",
                               ending="")
                    success_count = success_count + 1
                    self.w(" ")
                except ObjectDoesNotExist:
                    self.w(
                        self.style.ERROR(
                            "Error: can't find a local observation!!! "))
                except TaxonMatchError:
                    self.w(
                        self.style.WARNING(
                            "Error: We don't understand the community taxon id, so we ignore it "
                        ))
            else:
                from_inat_total_count = from_inat_total_count + 1
                self.w(
                    "Observation initially comes from a regular iNaturalist user. ",
                    ending="")
                # All those observations have been dropped before: recreate
                self.w("Creating a local observation for it. ", ending="")
                try:
                    r = create_observation_from_inat_data(
                        inat_observation_data)
                    if r.__class__ == Nest:
                        from_inat_nest_count = from_inat_nest_count + 1

                    self.w(self.style.SUCCESS("OK"))
                    success_count = success_count + 1
                except TaxonMatchError:
                    if inat_observation_data['taxon']['rank'] == 'genus':
                        self.w(
                            self.style.WARNING(
                                "Observation at the Genus level, skipping."))
                    else:
                        self.w(
                            self.style.ERROR(
                                "Error: cannot match taxon, skipping: " +
                                str(inat_observation_data)))
                except ParseDateError:
                    self.w(
                        self.style.ERROR(
                            "Error: cannot parse date, skipping: " +
                            str(inat_observation_data)))

        self.w("DONE. Stats:")
        self.w(f"{total_count} observations processed (total).")
        self.w(
            f"{success_count} were successful, {total_count - success_count} had errors."
        )
        self.w(
            f"{from_us_count} were from the Vespa-Watch app, {from_inat_total_count} were regular iNaturalist "
            f"observations, including {from_inat_nest_count} nest(s).")