hdx-python-api 6.7.8__py3-none-any.whl → 6.7.9__py3-none-any.whl

hdx/api/_version.py

@@ -18,7 +18,7 @@ version_tuple: tuple[int | str, ...]
 commit_id: str | None
 __commit_id__: str | None
 
-__version__ = version = '6.7.8'
-__version_tuple__ = version_tuple = (6, 7, 8)
+__version__ = version = '6.7.9'
+__version_tuple__ = version_tuple = (6, 7, 9)
 
 __commit_id__ = commit_id = None

hdx/data/dataset.py

@@ -667,21 +667,90 @@ class Dataset(HDXObject):
             if "batch_mode" not in kwargs:
                 kwargs["batch_mode"] = "DONT_GROUP"
 
+    def _merge_list_fields(self, dataset_data_to_update: dict) -> list[str]:
+        """Merges each non-resource list in dataset_data_to_update with the
+        existing list of the same key (see _merge_list) in place.
+
+        Args:
+            dataset_data_to_update: Dataset data to be updated
+
+        Returns:
+            Keys whose merged list differs from the existing one
+        """
+        changed_keys = []
+        for key, value in dataset_data_to_update.items():
+            if (
+                key == "resources"
+                or not isinstance(value, list)
+                or key not in self._old_data
+            ):
+                continue
+            orig_list = self._old_data[key]
+            if not isinstance(orig_list, list):
+                orig_list = []
+            merged_list = self._merge_list(orig_list, value)
+            dataset_data_to_update[key] = merged_list
+            if merged_list != orig_list:
+                changed_keys.append(key)
+        return changed_keys
+
+    @staticmethod
+    def _merge_list(orig_list: list, new_list: list) -> list:
+        """Merges each dict in new_list that has an id or name with the dict in
+        orig_list with the same id (or failing that, name) so that fields not
+        given in new_list are kept. Other elements of new_list are kept as is.
+        Elements of orig_list that are not matched are dropped.
+
+        Args:
+            orig_list: Existing list
+            new_list: Updated list
+
+        Returns:
+            Merged list
+        """
+        merged_list = []
+        for new_value in new_list:
+            if not isinstance(new_value, dict):
+                merged_list.append(new_value)
+                continue
+            match = {}
+            for field in ("id", "name"):
+                field_value = new_value.get(field)
+                if not field_value:
+                    continue
+                match = next(
+                    (
+                        orig_value
+                        for orig_value in orig_list
+                        if isinstance(orig_value, dict)
+                        and orig_value.get(field) == field_value
+                    ),
+                    {},
+                )
+                if match:
+                    break
+            merged_list.append(match | new_value)
+        return merged_list
+
     def _revise_filter(
         self,
-        dataset_data_to_update: dict,
         keys_to_delete: Sequence[str],
         resources_to_delete: Sequence[int],
-    ):
+        changed_list_keys: Sequence[str],
+    ) -> list[str]:
         """Returns the revise filter parameter adding to it any keys in the
-        dataset metadata and resources that are specified to be deleted. Also compare
-        lists in original and updated metadata to see if any have had elements
-        removed in which case these should be added to the filter
+        dataset metadata and resources that are specified to be deleted. Keys
+        of changed lists are also added so that package_revise replaces those
+        lists rather than merging them by position (which keeps the existing
+        elements' ids eg. for groups).
 
         Args:
-            dataset_data_to_update: Dataset data to be updated
             keys_to_delete: List of top level metadata keys to delete
             resources_to_delete: List of indexes of resources to delete
+            changed_list_keys: Keys of lists that differ from the existing ones
+
+        Returns:
+            Revise filter
         """
         revise_filter = []
         for key in keys_to_delete:
@@ -689,46 +758,9 @@ class Dataset(HDXObject):
         if not self.is_requestable():
             for resource_index in resources_to_delete:
                 revise_filter.append(f"-resources__{resource_index}")
-        for key, value in dataset_data_to_update.items():
-            if not isinstance(value, list):
-                continue
-            if key not in self._old_data:
-                continue
-            orig_list = self._old_data[key]
-            elements_to_remove = []
-            for i, orig_value in enumerate(orig_list):
-                if isinstance(orig_value, dict) and any(
-                    x in orig_value for x in ("id", "name")
-                ):
-                    # Where lists have an id or name, we use one of those to
-                    # match elements and work out what needs to be deleted
-                    # ie. any elements in the original list that don't exist
-                    # in the updated list
-                    el_id = orig_value.get("id")
-                    el_name = orig_value.get("name")
-                    present = False
-                    for value_dict in value:
-                        if el_id and "id" in value_dict and value_dict["id"] == el_id:
-                            present = True
-                            break
-                        if (
-                            el_name
-                            and "name" in value_dict
-                            and value_dict["name"] == el_name
-                        ):
-                            present = True
-                            break
-                    if not present:
-                        elements_to_remove.append(i)
-                elif orig_value not in value:
-                    # Otherwise we do a simple exact match of list elements
-                    # and delete any elements that are in the original list but
-                    # not in the updated list
-                    elements_to_remove.append(i)
-            for element_index in reversed(elements_to_remove):
-                del orig_list[element_index]
-                if element_index == len(orig_list):
-                    revise_filter.append(f"-{key}__{element_index}")
+        for key in changed_list_keys:
+            if key not in keys_to_delete:
+                revise_filter.append(f"-{key}")
         return revise_filter
 
     def _revise_files_to_upload_resource_deletions(
@@ -806,8 +838,9 @@ class Dataset(HDXObject):
         ):  # Whether or not CKAN should perform validation steps (checking fields present)
             dataset_data_to_update["skip_validation"] = kwargs["skip_validation"]
         dataset_data_to_update["state"] = "active"
+        changed_list_keys = self._merge_list_fields(dataset_data_to_update)
         revise_filter = self._revise_filter(
-            dataset_data_to_update, keys_to_delete, resources_to_delete
+            keys_to_delete, resources_to_delete, changed_list_keys
         )
         files_to_upload = self._revise_files_to_upload_resource_deletions(
             resources_to_update,
@@ -1048,8 +1081,8 @@ class Dataset(HDXObject):
             **kwargs,
         )
         keep_crisis_tags = kwargs.get("keep_crisis_tags", True)
-        if keep_crisis_tags:
-            old_tags = self._old_data.get("tags", [])
+        if keep_crisis_tags and "tags" in self._old_data:
+            old_tags = self._old_data["tags"]
             for tag in self.data.get("tags", []):
                 tag_name = tag["name"]
                 if tag_name[:7] != "crisis-":
@@ -1061,7 +1094,6 @@ class Dataset(HDXObject):
                         break
                 if not found:
                     old_tags.append(tag)
-            self._old_data["tags"] = old_tags
         self._prepare_hdx_call(self._old_data, kwargs)
         revise_call = self._revise_dataset(
             allow_no_resources,

{hdx_python_api-6.7.8.dist-info → hdx_python_api-6.7.9.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: hdx-python-api
-Version: 6.7.8
+Version: 6.7.9
 Summary: HDX Python API for interacting with the Humanitarian Data Exchange
 Project-URL: Homepage, https://github.com/OCHA-DAP/hdx-python-api
 Author: Michael Rans
@@ -27,8 +27,8 @@ Requires-Python: >=3.11
 Requires-Dist: ckanapi>=4.11
 Requires-Dist: defopt>=7.0.0
 Requires-Dist: email-validator
-Requires-Dist: hdx-python-country>=4.2.6
-Requires-Dist: hdx-python-utilities>=4.2.6
+Requires-Dist: hdx-python-country>=4.3.1
+Requires-Dist: hdx-python-utilities>=4.2.7
 Requires-Dist: makefun
 Requires-Dist: requests
 Provides-Extra: docs

{hdx_python_api-6.7.8.dist-info → hdx_python_api-6.7.9.dist-info}/RECORD

@@ -1,5 +1,5 @@
 hdx/api/__init__.py,sha256=11ozJKiUsqDCZ3_mcAHhGYUyGK_Unl54djVSBBExFB4,59
-hdx/api/_version.py,sha256=mfER4mocGyfEDcrh-EPSwFbbBfh6B41aVSJ98S98M9U,520
+hdx/api/_version.py,sha256=mqZBEZ3sW4hFzYhXwQAxPeU7oI9pRVlb2fo16Eapuxc,520
 hdx/api/configuration.py,sha256=40DTsYVCe1tQ5IVQPFdhW83CemzkEXMjNBk6WVeUkj8,23802
 hdx/api/hdx_base_configuration.yaml,sha256=jj2t5b7NnI499rHVwOjj3Q5BAjE8i7NDGVKnWZ5P5-4,2222
 hdx/api/locations.py,sha256=bFl_cKz_AEQzkE_xWOkrRlUQ0GT2LMPSk-G_LsbV-Ns,4099
@@ -13,7 +13,7 @@ hdx/api/utilities/hdx_stats.py,sha256=JlEcF9LN-OjcVeh6C0te-nAlcO7W-JIpD5JJWTiWxs
 hdx/api/utilities/url_utils.py,sha256=Pb4r-BAqIHYmeoB_i7QjutDp6nOAUh_OHvCtiNLda3Y,4534
 hdx/data/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 hdx/data/activity.py,sha256=YNa7hAgsMP96EIUX7UMInxAXnJ8TynE1KpecRrv8Vso,9200
-hdx/data/dataset.py,sha256=ze9fSvPGpg0zAS9yTuTHpqhsGdny8CSqsHmrsfccvrY,105723
+hdx/data/dataset.py,sha256=wbo2ShjXjelZcLm4I8V9wx-RDf_lXf2FsUjH-74VvWU,106379
 hdx/data/hdxobject.py,sha256=SCrnmOhBtP-6-1EdKzjgV4u0aclTzDhA4QghsaQ0qN4,28379
 hdx/data/location.py,sha256=eEdHZGEYKa9qRDyYqGTwAh22TB4XBkcTu47Mz39fmEs,11171
 hdx/data/organization.py,sha256=T2lITJXEYDN5t8dYkfcQTe-x0joULTtgp3Vn9JeVh-k,13133
@@ -27,7 +27,7 @@ hdx/facades/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 hdx/facades/infer_arguments.py,sha256=3Z5vPPkni0Cgg9cUb7wDF_8TxeX1TiiffcnGqyahtiY,4015
 hdx/facades/keyword_arguments.py,sha256=aVapHCfqBO6H7xxBgTdK9ISdgrBFsNypPDWihfQ3edU,1125
 hdx/facades/simple.py,sha256=QWsBkaHG_UM6X3Zu2u3LWVOQzeR24q-EUClrfsL_mkM,1064
-hdx_python_api-6.7.8.dist-info/METADATA,sha256=A3h4HM4h438MTxH4dnLkqWPrK9Dyw0w59rqkd2aDNOQ,4895
-hdx_python_api-6.7.8.dist-info/WHEEL,sha256=2iZrJkDIkeDh0k9SikKMqFuRP-ua6mb4Ir1AsER4BEw,87
-hdx_python_api-6.7.8.dist-info/licenses/LICENSE,sha256=wc-4GpMn-ODs-U_bTe1YCiPVgvcjzrpYOx2wPuyAeII,1079
-hdx_python_api-6.7.8.dist-info/RECORD,,
+hdx_python_api-6.7.9.dist-info/METADATA,sha256=640fkpkJHrkSRiYLy1UUP9hJ-jxQgEWMz-GIB8U5Dr0,4895
+hdx_python_api-6.7.9.dist-info/WHEEL,sha256=2iZrJkDIkeDh0k9SikKMqFuRP-ua6mb4Ir1AsER4BEw,87
+hdx_python_api-6.7.9.dist-info/licenses/LICENSE,sha256=wc-4GpMn-ODs-U_bTe1YCiPVgvcjzrpYOx2wPuyAeII,1079
+hdx_python_api-6.7.9.dist-info/RECORD,,