feat: ROOT-1: Add offset to storage link model to support storage link to task within file of tasks on cloud storage (#7534)

Co-authored-by: Jo Booth <jo.m.booth@gmail.com>
Co-authored-by: matt-bernstein <matt-bernstein@users.noreply.github.com>

Author: matt-bernstein

label_studio/io_storages/base_models.py

@@ -689,11 +689,17 @@ class ImportStorageLink(models.Model):
 
     task = models.OneToOneField('tasks.Task', on_delete=models.CASCADE, related_name='%(app_label)s_%(class)s')
     key = models.TextField(_('key'), null=False, help_text='External link key')
+
+    # This field is set to True on creation and never updated; it should not be relied upon.
     object_exists = models.BooleanField(
         _('object exists'), help_text='Whether object under external link still exists', default=True
     )
+
     created_at = models.DateTimeField(_('created at'), auto_now_add=True, help_text='Creation time')
 
+    row_group = models.IntegerField(null=True, blank=True, help_text='Parquet row group')
+    row_index = models.IntegerField(null=True, blank=True, help_text='Parquet row index, or JSON[L] object index')
+
     @classmethod
     def exists(cls, key, storage):
         return cls.objects.filter(key=key, storage=storage.id).exists()

label_studio/io_storages/migrations/0019_azureblobimportstoragelink_row_group_and_more.py

@@ -0,0 +1,93 @@
+# Generated by Django 5.1.9 on 2025-05-14 17:23
+
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ("io_storages", "0018_alter_azureblobexportstorage_project_and_more"),
+    ]
+
+    operations = [
+        migrations.AddField(
+            model_name="azureblobimportstoragelink",
+            name="row_group",
+            field=models.IntegerField(
+                blank=True, help_text="Parquet row group", null=True
+            ),
+        ),
+        migrations.AddField(
+            model_name="azureblobimportstoragelink",
+            name="row_index",
+            field=models.IntegerField(
+                blank=True,
+                help_text="Parquet row index, or JSON[L] object index",
+                null=True,
+            ),
+        ),
+        migrations.AddField(
+            model_name="gcsimportstoragelink",
+            name="row_group",
+            field=models.IntegerField(
+                blank=True, help_text="Parquet row group", null=True
+            ),
+        ),
+        migrations.AddField(
+            model_name="gcsimportstoragelink",
+            name="row_index",
+            field=models.IntegerField(
+                blank=True,
+                help_text="Parquet row index, or JSON[L] object index",
+                null=True,
+            ),
+        ),
+        migrations.AddField(
+            model_name="localfilesimportstoragelink",
+            name="row_group",
+            field=models.IntegerField(
+                blank=True, help_text="Parquet row group", null=True
+            ),
+        ),
+        migrations.AddField(
+            model_name="localfilesimportstoragelink",
+            name="row_index",
+            field=models.IntegerField(
+                blank=True,
+                help_text="Parquet row index, or JSON[L] object index",
+                null=True,
+            ),
+        ),
+        migrations.AddField(
+            model_name="redisimportstoragelink",
+            name="row_group",
+            field=models.IntegerField(
+                blank=True, help_text="Parquet row group", null=True
+            ),
+        ),
+        migrations.AddField(
+            model_name="redisimportstoragelink",
+            name="row_index",
+            field=models.IntegerField(
+                blank=True,
+                help_text="Parquet row index, or JSON[L] object index",
+                null=True,
+            ),
+        ),
+        migrations.AddField(
+            model_name="s3importstoragelink",
+            name="row_group",
+            field=models.IntegerField(
+                blank=True, help_text="Parquet row group", null=True
+            ),
+        ),
+        migrations.AddField(
+            model_name="s3importstoragelink",
+            name="row_index",
+            field=models.IntegerField(
+                blank=True,
+                help_text="Parquet row index, or JSON[L] object index",
+                null=True,
+            ),
+        ),
+    ]