debug-search-transform.py

Modified call_indexer() for debugging

    def call_indexer(self, entity, reindex=False, document=None, target_index=None):
        logger.info(f"Start executing call_indexer() on uuid: {entity['uuid']}, entity_type: {entity['entity_type']}")

        try:
            if document is None:
                try:
                    document = self.generate_doc(entity, 'json')
                except Exception as e:
                    logger.exception(e)
                    logger.error(f"Failed to execute generate_doc() on {entity['entity_type']} {entity['uuid']} during executing call_indexer()")

                    # Raise the exception so the caller can handle it properly
                    raise Exception(e)

            if target_index:
                self.indexer.index(entity['uuid'], document, target_index, reindex)
            elif entity['entity_type'] == 'Upload':
                target_index = self.INDICES['indices'][self.DEFAULT_INDEX_WITHOUT_PREFIX]['private']

                self.indexer.index(entity['uuid'], document, target_index, reindex)
            else:
                # write entity into indices
                for index in self.indices.keys():
                    public_index = self.INDICES['indices'][index]['public']
                    private_index = self.INDICES['indices'][index]['private']

                    # check to see if the index has a transformer, default to None if not found
                    transformer = self.TRANSFORMERS.get(index, None)

                    if self.is_public(entity):
                        try:
                            public_doc = self.generate_public_doc(entity)

                            if transformer is not None:
                                public_transformed = transformer.transform(json.loads(public_doc), self.transformation_resources)

                                
                                test_json={
                                "created_by_user_displayname": "Amanda Knoten",
                                "created_by_user_email": "aknoten@wustl.edu",
                                "created_timestamp": 1642630383146,
                                "data_access_level": "public",
                                "description": "cryosections on dry ice",
                                "entity_type": "Sample",
                                "group_name": "University of California San Diego TMC",
                                "group_uuid": "03b3d854-ed44-11e8-8bce-0e368f3075e8",
                                "hubmap_id": "HBM477.XBRN.629",
                                "lab_tissue_sample_id": "HBM564_L_D3",
                                "last_modified_timestamp": 1642630383146,
                                "protocol_url": "dx.doi.org/10.17504/protocols.io.568g9hw",
                                "sample_category": "section",
                                "submission_id": "CALT0001-HT-2-3-2",
                                "uuid": "e18f88add6de4b5abc0c07d148076390"
                                }
                                test_public_transformed = transformer.transform(test_json, self.transformation_resources)
                                logger.info("========test_public_transformed=========")
                                logger.info(test_public_transformed)
                                logger.info("========test_public_transformed=========")


                                public_transformed_doc = json.dumps(public_transformed)
                                target_doc = public_transformed_doc
                            else:
                                target_doc = public_doc

                            #self.indexer.index(entity['uuid'], target_doc, public_index, reindex)
                            logger.info(f"Finished executing indexer.index({entity['uuid']}, 'target_doc', {public_index}, {reindex}) for entity_type: {entity['entity_type']}")
                        except Exception:
                            msg = f"Exception encountered during executing generate_public_doc() inside call_indexer() for uuid: {entity['uuid']}, entity_type: {entity['entity_type']}"
                            # Log the full stack trace, prepend a line with our message
                            logger.exception(msg)

                    # add it to private
                    if transformer is not None:
                        private_transformed = transformer.transform(json.loads(document), self.transformation_resources)
                        target_doc = json.dumps(private_transformed)
                    else:
                        target_doc = document

                    #self.indexer.index(entity['uuid'], target_doc, private_index, reindex)
                    logger.info(f"Finished executing indexer.index({entity['uuid']}, 'target_doc', {private_index}, {reindex}) for entity_type: {entity['entity_type']}")

            logger.info(f"Finished executing call_indexer() on uuid: {entity['uuid']}, entity_type: {entity['entity_type']}")
        except Exception as e:
            logger.exception(e)
            msg = f"Exception encountered during executing call_indexer() for uuid: {entity['uuid']}, entity_type: {entity['entity_type']}"
            # Log the full stack trace, prepend a line with our message
            logger.error(msg)

Calling transformer.transform() directly on this test json returns two more fields mapper_metadata and mapped_consortium compared to the portal result:

The portal transform adds extra mapped fields on top of the entity-api /documents/<id> result