Reduce 'noise' in create changeset from changing order of items? #40

Open
opened 2026-05-26 14:24:51 +00:00 by adina · 3 comments
Owner

If I create a changeset, I often see changes that are simply a different order of authors (though I didn't touch them). Not meaningful per se, but it kind of adds noise to the curation process. Is there a way to sort authors in a fixed way by any chance?

Here is an example:

diff --git a/records/forgejo-user-adina/xyzrins:datasets-_6ac48b8c--c971--4c9f--be9c--be6e8480b724 b/records/forgejo-user-adina/xyzrins:datasets-_6ac48b8c--c971--4c9f--be9c--be6e8480b724
index ba4ef79..120d25e 100644
--- a/records/forgejo-user-adina/xyzrins:datasets-_6ac48b8c--c971--4c9f--be9c--be6e8480b724
+++ b/records/forgejo-user-adina/xyzrins:datasets-_6ac48b8c--c971--4c9f--be9c--be6e8480b724
@@ -2,19 +2,22 @@
   "attributed_to": [
     {
       "schema_type": "dlthings:Attribution",
-      "object": "xyzrins:persons/michael-hanke",
+      "object": "xyzrins:persons/christian-haeusler",
       "roles": [
         "marcrel:cre"
       ]
     },
     {
       "schema_type": "dlthings:Attribution",
-      "object": "xyzrins:persons/christian-haeusler",
+      "object": "xyzrins:persons/michael-hanke",
       "roles": [
         "marcrel:cre"
       ]
     }
   ],
+  "about": [
+    "xyzrins:topics/naturalistic-stimulation"
+  ],
   "description": "Movie stimulus annotations for the research cut of the \"Forrest Gump\" movie used in the studyforrest.org project. Annotations are collected from various contributors and publications.\n\nAnnotations are typically provided as plain text tables, using a tab-separated-value markup with a header row. Table usually contain a onset and a duration column to indicate the timing of an event. All other columns contain variables that describe properties of an event.",
   "display_label": "Studyforrest: Annotations",
   "attributes": [
If I create a changeset, I often see changes that are simply a different order of authors (though I didn't touch them). Not meaningful per se, but it kind of adds noise to the curation process. Is there a way to sort authors in a fixed way by any chance? Here is an example: ```diff diff --git a/records/forgejo-user-adina/xyzrins:datasets-_6ac48b8c--c971--4c9f--be9c--be6e8480b724 b/records/forgejo-user-adina/xyzrins:datasets-_6ac48b8c--c971--4c9f--be9c--be6e8480b724 index ba4ef79..120d25e 100644 --- a/records/forgejo-user-adina/xyzrins:datasets-_6ac48b8c--c971--4c9f--be9c--be6e8480b724 +++ b/records/forgejo-user-adina/xyzrins:datasets-_6ac48b8c--c971--4c9f--be9c--be6e8480b724 @@ -2,19 +2,22 @@ "attributed_to": [ { "schema_type": "dlthings:Attribution", - "object": "xyzrins:persons/michael-hanke", + "object": "xyzrins:persons/christian-haeusler", "roles": [ "marcrel:cre" ] }, { "schema_type": "dlthings:Attribution", - "object": "xyzrins:persons/christian-haeusler", + "object": "xyzrins:persons/michael-hanke", "roles": [ "marcrel:cre" ] } ], + "about": [ + "xyzrins:topics/naturalistic-stimulation" + ], "description": "Movie stimulus annotations for the research cut of the \"Forrest Gump\" movie used in the studyforrest.org project. Annotations are collected from various contributors and publications.\n\nAnnotations are typically provided as plain text tables, using a tab-separated-value markup with a header row. Table usually contain a onset and a duration column to indicate the timing of an event. All other columns contain variables that describe properties of an event.", "display_label": "Studyforrest: Annotations", "attributes": [ ```
Owner

Looking into it

Looking into it
Owner

I believe that the reordering might have happened on the client side, e.g., in shacl-vue. I have not found any code path or example in dump-things-server or dtc where list-content would be reordered (which is what it looks like in your example: the attributed_to-list is sorted and the list-entry with "object"-value "xyzrins:persons/christian-haeusler" is sorted before the list-entry with "object"-value "xyzrins:persons/michael-hanke").

I believe that the reordering might have happened on the client side, e.g., in shacl-vue. I have not found any code path or example in `dump-things-server` or `dtc` where list-content would be reordered (which is what it looks like in your example: the `attributed_to`-list is sorted and the list-entry with `"object"`-value `"xyzrins:persons/christian-haeusler"` is sorted before the list-entry with `"object"`-value `"xyzrins:persons/michael-hanke"`).
Owner

@jsheunis: is it possible, that sorting of the attributed_to-list is performed in shacl-vue?

@jsheunis: is it possible, that sorting of the `attributed_to`-list is performed in `shacl-vue`?
Sign in to join this conversation.
No labels
No milestone
No project
No assignees
2 participants
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
orinoco/dump-things-pyclient#40
No description provided.