First version of renaming group paths and relabeling the Logger on group relabel.

Author: GeigerJ2

src/aiida/cmdline/commands/cmd_group.py

@@ -319,12 +319,19 @@ def _dry_run_callback(pks):
 @with_dbenv()
 def group_relabel(group, label):
     """Change the label of a group."""
+    # TODO: Add a message here that if one has the profile mirrored, they should also run the command `verdi profile
+    # TODO: mirror relabel-group ` to update the mirrored profile.
     try:
         group.label = label
     except UniquenessError as exception:
         echo.echo_critical(str(exception))
     else:
         echo.echo_success(f"Label changed to '{label}'")
+        msg = (
+            'Note that if you are mirroring your profile data to disk, to reflect the relabeling of the group, '
+            'run the command: `verdi profile mirror --update-groups.'
+        )
+        echo.echo_report(msg)
 
 
 @verdi_group.command('description')

src/aiida/cmdline/commands/cmd_profile.py

@@ -303,6 +303,14 @@ def profile_delete(force, delete_data, profiles):
     show_default=True,
     help='If a top-level process calls sub-processes, create a designated directory only for the top-level process.',
 )
+# TODO: Implement this...
+# TODO: Possibly
+@click.option(
+    '--update-groups/--no-update-groups',
+    default=False,
+    show_default=True,
+    help='Update directories if nodes have been added to other groups, or organized differently in terms of groups.',
+)
 @options.INCLUDE_INPUTS()
 @options.INCLUDE_OUTPUTS()
 @options.INCLUDE_ATTRIBUTES()
@@ -319,6 +327,7 @@ def profile_mirror(
     organize_by_groups,
     symlink_duplicates,
     delete_missing,
+    update_groups,
     only_top_level_calcs,
     only_top_level_workflows,
     include_inputs,
@@ -343,6 +352,11 @@ def profile_mirror(
     # ? Does it even make sense to provide both options, as they are mutually exclusive?
     incremental = not overwrite
 
+    if not organize_by_groups and update_groups:
+        # Add check outside in cmd_profile?
+        msg = '`--update-groups` selected, even though `--organize-by-groups` is set to False.'
+        echo.echo_critical(msg)
+
     if path is None:
         path = Path.cwd() / f'{profile.name}-mirror'
 
@@ -421,6 +435,9 @@ def profile_mirror(
     num_processes_to_dump = len(profile_dumper.processes_to_dump)
     num_processes_to_delete = len(profile_dumper.processes_to_delete)
 
+    # num_groups_to_dump = len(profile_dumper.groups_to_dump)
+    num_groups_to_delete = len(profile_dumper.groups_to_delete)
+
     if dry_run:
         dry_run_message = (
             f'Dry run for mirroring of profile `{profile.name}`. '
@@ -438,23 +455,35 @@ def profile_mirror(
         # TODO: Maybe add y/n confirmation here?
         echo.echo_report(msg)
 
-    if num_processes_to_dump == 0:
-        msg = 'No processes to dump.'
-        echo.echo_success(msg)
-    else:
-        profile_dumper.dump_processes()
-        msg = f'Dumped {num_processes_to_dump} new nodes.'
-        echo.echo_success(msg)
+    if dump_processes:
+        if num_processes_to_dump == 0:
+            msg = 'No processes to dump.'
+            echo.echo_success(msg)
+        else:
+            profile_dumper.dump_processes()
+            msg = f'Dumped {num_processes_to_dump} new nodes.'
+            echo.echo_success(msg)
 
     if delete_missing:
-        # breakpoint()
         if num_processes_to_delete == 0:
             echo.echo_success('No processes to delete.')
         else:
             profile_dumper.delete_processes()
-
             echo.echo_success(f'Deleted {num_processes_to_delete} node directories.')
 
+        if num_groups_to_delete == 0:
+            echo.echo_success('No groups to delete.')
+        else:
+            profile_dumper.delete_groups()
+            echo.echo_success(f'Deleted {num_groups_to_delete} group directories.')
+
+    if update_groups:
+        relabeled_paths = profile_dumper.update_groups()
+
+        msg = 'Mirrored directories and '
+        echo.echo_success(msg)
+        print(relabeled_paths)
+
     # Append the current dump time to dumping safeguard file
     with safeguard_file_path.open('a') as fhandle:
         fhandle.write(f'Last profile mirror time: {last_dump_time.isoformat()}\n')

src/aiida/tools/dumping/logger.py

@@ -15,6 +15,9 @@
 
 from aiida.common.exceptions import NotExistent
 
+# TODO: Possibly mirror hierarchy of mirrored directory inside json file
+# TODO: Currently, json file has only top-level "groups", "workflows", and "calculations"
+
 
 @dataclass
 class DumpLog:
@@ -94,10 +97,12 @@ class DumpLogStoreCollection:
 
     calculations: DumpLogStore
     workflows: DumpLogStore
+    groups: DumpLogStore
+    # data: DumpLogStore
 
 
 class DumpLogger:
-    """Main logger class using dataclasses for better structure."""
+    """Main dumping logger singleton."""
 
     DUMP_LOG_FILE: str = '.dump_log.json'
     _instance: 'DumpLogger | None' = None  # Class-level singleton instance
@@ -116,7 +121,8 @@ def __init__(
         dump_parent_path: Path | None = None,
         calculations: DumpLogStore | None = None,
         workflows: DumpLogStore | None = None,
-        # counter: int = 0,
+        groups: DumpLogStore | None = None,
+        # data: DumpLogStore | None = None,
     ) -> None:
         # Ensure __init__ is only called once
         if hasattr(self, '_initialized') and self._initialized:
@@ -125,7 +131,8 @@ def __init__(
         self.dump_parent_path = dump_parent_path or Path.cwd()
         self.calculations = calculations or DumpLogStore()
         self.workflows = workflows or DumpLogStore()
-        # self.counter = counter
+        self.groups = groups or DumpLogStore()
+        # self.dat = data or DumpLogStore()
 
         # Mark the object as initialized
         self._initialized = True
@@ -144,20 +151,26 @@ def del_entry(self, store: DumpLogStore, uuid: str) -> bool:
     @property
     def log(self) -> DumpLogStoreCollection:
         """Retrieve the current state of the log as a dataclass."""
-        return DumpLogStoreCollection(calculations=self.calculations, workflows=self.workflows)
+        return DumpLogStoreCollection(calculations=self.calculations, workflows=self.workflows, groups=self.groups)
 
     def save_log(self) -> None:
         """Save the log to a JSON file."""
 
         def serialize_logs(container: DumpLogStore) -> dict:
             serialized = {}
             for uuid, entry in container.entries.items():
-                serialized[uuid] = {'path': str(entry.path), 'time': entry.time.isoformat()}
+                serialized[uuid] = {
+                    'path': str(entry.path),
+                    'time': entry.time.isoformat(),
+                    'links': [str(link) for link in entry.links],
+                }
             return serialized
 
         log_dict = {
             'calculations': serialize_logs(self.calculations),
             'workflows': serialize_logs(self.workflows),
+            'groups': serialize_logs(self.groups),
+            # 'data': serialize_logs(self.data),
         }
 
         with self.log_file_path.open('w', encoding='utf-8') as f:
@@ -185,12 +198,20 @@ def deserialize_logs(category_data: dict) -> DumpLogStore:
                 container = DumpLogStore()
                 for uuid, entry in category_data.items():
                     container.add_entry(
-                        uuid, DumpLog(path=Path(entry['path']), time=datetime.fromisoformat(entry['time']))
+                        uuid,
+                        DumpLog(
+                            path=Path(entry['path']),
+                            time=datetime.fromisoformat(entry['time']),
+                            links=[Path(p) for p in entry['links']],
+                        ),
                     )
+
                 return container
 
             instance.calculations = deserialize_logs(data['calculations'])
             instance.workflows = deserialize_logs(data['workflows'])
+            instance.groups = deserialize_logs(data['groups'])
+            # instance.data = deserialize_logs(data['data'])
 
         except (json.JSONDecodeError, OSError):
             raise
@@ -206,7 +227,7 @@ def get_store_by_uuid(self, uuid: str) -> DumpLogStore:
             if uuid in store.entries:
                 return store
 
-        msg = f"No corresponding `DumpLogStore` found for UUID: `{uuid}`."
+        msg = f'No corresponding `DumpLogStore` found for UUID: `{uuid}`.'
         raise NotExistent(msg)
 
     def get_path_by_uuid(self, uuid: str) -> Path | None:
@@ -215,13 +236,17 @@ def get_path_by_uuid(self, uuid: str) -> Path | None:
 
         try:
             current_store = self.get_store_by_uuid(uuid=uuid)
-            path = current_store.entries[uuid].path
-            return path
         except NotExistent as exc:
             raise NotExistent(exc.args[0]) from exc
+        try:
+            path = current_store.entries[uuid].path
+            return path
         except KeyError as exc:
-            msg = f"UUID: `{uuid}` not contained in store `{current_store}`."
+            msg = f'UUID: `{uuid}` not contained in store `{current_store}`.'
             raise KeyError(msg) from exc
         except:
             # For debugging
+            import ipdb
+
+            ipdb.set_trace()
             raise