final cleanup code

src/hub/dataload/sources/chembl/parser.py

@@ -27,14 +27,6 @@ def parse_data(data):
                     },
                 }
                 yield output
-        # if uniprot_accessions:
-        #     output = {
-        #         "chembl": {
-        #             "chembl_target": item["target_chembl_id"],
-        #             "uniprot_accession": uniprot_accessions,
-        #         },
-        #     }
-        #     yield output
 
 
 def load_data(target_filepaths):

src/hub/dataload/sources/chembl/upload.py

@@ -1,45 +1,18 @@
-import copy
 import glob
 import os
 
 import biothings.hub.dataload.storage as storage
 import biothings.hub.dataload.uploader as uploader
-from biothings.utils.dataload import merge_root_keys
 
 from hub.datatransform.keylookup import MyGeneKeyLookup
 
 from .parser import load_data
 
 
-class ChemblMergerStorage(storage.RootKeyMergerStorage):
-    """
-    Just like MergerStorage, this storage deals with duplicated error
-    by appending key's content to existing document. Keys in existing
-    document will be converted to a list as needed.
-
-    Note:
-      - root keys must have the same type in each documents
-      - inner structures aren't merged together, the merge happend
-        at root key level
-    """
-
-    @classmethod
-    def merge_func(klass, doc1, doc2, **kwargs):
-        # caller popped it from doc1, let's take from doc2
-        _id = doc2["_id"]
-        # exclude_id will remove _if from doc2, that's why we kept it from before
-        # also, copy doc2 ref as the merged doc will be stored in
-        # a bulk op object, since doc2 is modified in place, this could impact
-        # the bulk op and procude empty $set error from mongo
-        doc = merge_root_keys(doc1, copy.copy(doc2), exclude=["_id", "xrefs"])
-        doc["_id"] = _id
-        return doc
-
-
 class ChemblUploader(uploader.BaseSourceUploader):
     name = "chembl"
 
-    storage_class = ChemblMergerStorage
+    storage_class = storage.RootKeyMergerStorage
     TARGET_FILENAME_PATTERN = "target.*.json"
 
     keylookup = MyGeneKeyLookup(