(migration): Resolve isupdate correctly

specifyweb/workbench/migrations/0008_alter_spdataset_isupdate.py

@@ -0,0 +1,18 @@
+# Generated by Django 3.2.15 on 2024-08-30 15:10
+
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ('workbench', '0007_spdataset_parent'),
+    ]
+
+    operations = [
+        migrations.AlterField(
+            model_name='spdataset',
+            name='isupdate',
+            field=models.BooleanField(default=False, null=True),
+        ),
+    ]

specifyweb/workbench/models.py

@@ -7,9 +7,16 @@
 from django.utils import timezone
 
 from specifyweb.specify.func import Func
-from specifyweb.specify.models import Collection, Specifyuser, Agent, datamodel, custom_save
+from specifyweb.specify.models import (
+    Collection,
+    Specifyuser,
+    Agent,
+    datamodel,
+    custom_save,
+)
 from specifyweb.specify.api import uri_for_model
 
+
 class Dataset(models.Model):
     # All these attributes are meta-data.
     name = models.CharField(max_length=256)
@@ -29,88 +36,134 @@ class Dataset(models.Model):
     # Misc meta-data.
     timestampcreated = models.DateTimeField(default=timezone.now)
     timestampmodified = models.DateTimeField(auto_now=True)
-    createdbyagent = models.ForeignKey(Agent, null=True, on_delete=models.SET_NULL, related_name="+")
-    modifiedbyagent = models.ForeignKey(Agent, null=True, on_delete=models.SET_NULL, related_name="+")
-
-    base_meta_fields = ["name", "uploaderstatus", "timestampcreated", "timestampmodified"]
-    object_response_fields = [*base_meta_fields, "id", "remarks", "importedfilename", "uploadresult", "uploadplan"]
+    createdbyagent = models.ForeignKey(
+        Agent, null=True, on_delete=models.SET_NULL, related_name="+"
+    )
+    modifiedbyagent = models.ForeignKey(
+        Agent, null=True, on_delete=models.SET_NULL, related_name="+"
+    )
+
+    base_meta_fields = [
+        "name",
+        "uploaderstatus",
+        "timestampcreated",
+        "timestampmodified",
+    ]
+    object_response_fields = [
+        *base_meta_fields,
+        "id",
+        "remarks",
+        "importedfilename",
+        "uploadresult",
+        "uploadplan",
+    ]
 
     @classmethod
     def get_meta_fields(cls, request, extra_meta_fields=None, extra_filters=None):
-        attrs = [*cls.base_meta_fields, *(extra_meta_fields if extra_meta_fields is not None else [])]
+        attrs = [
+            *cls.base_meta_fields,
+            *(extra_meta_fields if extra_meta_fields is not None else []),
+        ]
 
-        dss  = cls.objects.filter(
+        dss = cls.objects.filter(
             specifyuser=request.specify_user,
             collection=request.specify_collection,
             **(extra_filters if extra_filters is not None else {})
         ).only(*attrs)
-        return [{'id': ds.id, **{attr: getattr(ds, attr) for attr in attrs}, 'uploadplan': json.loads(ds.uploadplan) if ds.uploadplan else None} for ds in dss]
+        return [
+            {
+                "id": ds.id,
+                **{attr: getattr(ds, attr) for attr in attrs},
+                "uploadplan": json.loads(ds.uploadplan) if ds.uploadplan else None,
+            }
+            for ds in dss
+        ]
 
     # raise_404: Whether to raise 404 or return http 404.
     # lock_object: Whether to run a "select for update" or "select"
     @classmethod
     def validate_dataset_request(cls, raise_404: bool, lock_object: bool):
         def decorator(func):
             def inner(request, **kwargs):
-                ds_id = kwargs.get('ds_id', None)
+                ds_id = kwargs.get("ds_id", None)
                 if ds_id is None:
-                    raise Exception('ds_id not a key in the request. '
-                                    'Probably because correct group name is not used url regexp')
+                    raise Exception(
+                        "ds_id not a key in the request. "
+                        "Probably because correct group name is not used url regexp"
+                    )
                 try:
-                    ds = cls.objects.select_for_update().get(id=ds_id) \
-                        if lock_object else cls.objects.get(id=ds_id)
+                    ds = (
+                        cls.objects.select_for_update().get(id=ds_id)
+                        if lock_object
+                        else cls.objects.get(id=ds_id)
+                    )
                 except ObjectDoesNotExist as e:
                     if raise_404:
                         raise Http404(e)
                     return http.HttpResponseNotFound()
 
                 if ds.specifyuser != request.specify_user:
                     return http.HttpResponseForbidden()
-                new_args = {key: kwargs[key] for key in kwargs if key != 'ds_id'}
+                new_args = {key: kwargs[key] for key in kwargs if key != "ds_id"}
                 return func(request, ds, **new_args)
+
             return inner
-        return decorator
 
+        return decorator
 
     def get_dataset_as_dict(self):
         ds_dict = {key: getattr(self, key) for key in self.object_response_fields}
-        ds_dict.update({
-            "rows": self.data,
-            "uploadplan": Func.maybe(self.uploadplan, json.loads),
-            "createdbyagent": uri_for_model('agent', self.createdbyagent_id) if self.createdbyagent_id is not None else None,
-            "modifiedbyagent": uri_for_model('agent', self.modifiedbyagent_id) if self.modifiedbyagent_id is not None else None
-        })
+        ds_dict.update(
+            {
+                "rows": self.data,
+                "uploadplan": Func.maybe(self.uploadplan, json.loads),
+                "createdbyagent": (
+                    uri_for_model("agent", self.createdbyagent_id)
+                    if self.createdbyagent_id is not None
+                    else None
+                ),
+                "modifiedbyagent": (
+                    uri_for_model("agent", self.modifiedbyagent_id)
+                    if self.modifiedbyagent_id is not None
+                    else None
+                ),
+            }
+        )
         return ds_dict
 
     class Meta:
         abstract = True
 
 
 class Spdataset(Dataset):
-    specify_model = datamodel.get_table('spdataset')
+    specify_model = datamodel.get_table("spdataset")
 
     columns = models.JSONField()
     visualorder = models.JSONField(null=True)
     rowresults = models.TextField(null=True)
 
-    isupdate = models.BooleanField(default=False)
+    isupdate = models.BooleanField(default=False, null=True)
 
     # very complicated. Essentially, each batch-edit dataset gets backed by another dataset (for rollbacks).
     # This should be a one-to-one field, imagine the mess otherwise.
-    parent = models.OneToOneField('Spdataset', related_name='backer', null=True, on_delete=models.CASCADE)
+    parent = models.OneToOneField(
+        "Spdataset", related_name="backer", null=True, on_delete=models.CASCADE
+    )
 
     class Meta:
-        db_table = 'spdataset'
+        db_table = "spdataset"
 
     def get_dataset_as_dict(self):
         ds_dict = super().get_dataset_as_dict()
-        ds_dict.update({
-            "columns": self.columns,
-            "visualorder": self.visualorder,
-            "rowresults": self.rowresults and json.loads(self.rowresults),
-            "isupdate": self.isupdate
-        })
+        ds_dict.update(
+            {
+                "columns": self.columns,
+                "visualorder": self.visualorder,
+                "rowresults": self.rowresults and json.loads(self.rowresults),
+                "isupdate": self.isupdate == True,
+            }
+        )
         return ds_dict
 
     def was_uploaded(self) -> bool:
-        return self.uploadresult and self.uploadresult['success']
+        return self.uploadresult and self.uploadresult["success"]

specifyweb/workbench/views.py

@@ -660,7 +660,7 @@ def rows(request, ds) -> http.HttpResponse:
             )
 
         rows = regularize_rows(
-            len(ds.columns), json.load(request), skip_empty=(not ds.isupdate)
+            len(ds.columns), json.load(request), skip_empty=(ds.isupdate != True)
         )
 
         ds.data = rows
@@ -993,7 +993,7 @@ def validate_row(request, ds_id: str) -> http.HttpResponse:
     bt, upload_plan = uploader.get_ds_upload_plan(collection, ds)
     row = json.loads(request.body)
     ncols = len(ds.columns)
-    rows = regularize_rows(ncols, [row], skip_empty=(not ds.isupdate))
+    rows = regularize_rows(ncols, [row], skip_empty=(ds.isupdate != True))
     if not rows:
         return http.JsonResponse(None, safe=False)
     row = rows[0]