Add dataset keywords model

gateway/sds_gateway/api_methods/admin.py

@@ -21,10 +21,19 @@ class CaptureAdmin(admin.ModelAdmin):  # pyright: ignore[reportMissingTypeArgume
 
 @admin.register(models.Dataset)
 class DatasetAdmin(admin.ModelAdmin):  # pyright: ignore[reportMissingTypeArgument]
-    list_display = ("name", "doi")
-    search_fields = ("name", "doi")
+    list_display = ("name", "doi", "get_keywords", "status", "owner")
+    search_fields = ("name", "doi", "keywords__name", "owner__email")
+    list_filter = ("status", "keywords")
     ordering = ("-updated_at",)
 
+    @admin.display(description="Keywords")
+    def get_keywords(self, obj):
+        """Display comma-separated list of keywords."""
+        keywords = obj.keywords.filter(is_deleted=False)
+        if keywords.exists():
+            return ", ".join([kw.name for kw in keywords[:5]])
+        return "-"
+
 
 @admin.register(models.TemporaryZipFile)
 class TemporaryZipFileAdmin(admin.ModelAdmin):  # pyright: ignore[reportMissingTypeArgument]
@@ -85,3 +94,16 @@ class ShareGroupAdmin(admin.ModelAdmin):  # pyright: ignore[reportMissingTypeArg
     list_display = ("name", "owner")
     search_fields = ("name", "owner")
     ordering = ("-updated_at",)
+
+
+@admin.register(models.Keyword)
+class KeywordAdmin(admin.ModelAdmin):  # pyright: ignore[reportMissingTypeArgument]
+    list_display = ("name", "get_datasets", "created_at")
+    search_fields = ("name", "datasets__name")
+    list_filter = ("datasets",)
+    ordering = ("name",)
+
+    @admin.display(description="Datasets")
+    def get_datasets(self, obj):
+        """Display comma-separated list of dataset names."""
+        return ", ".join([dataset.name for dataset in obj.datasets.all()[:3]])

gateway/sds_gateway/api_methods/migrations/max_migration.txt

@@ -1 +1 @@
-0017_convert_dataset_authors_to_object_format
+0021_remove_old_dataset_keywords_table

gateway/sds_gateway/api_methods/models.py

@@ -18,6 +18,7 @@
 from django.db.models.signals import post_save
 from django.db.models.signals import pre_delete
 from django.dispatch import receiver
+from django.template.defaultfilters import slugify
 
 from .utils.opensearch_client import get_opensearch_client
 
@@ -27,6 +28,24 @@
 log = logging.getLogger(__name__)
 
 
+class KeywordNameField(models.CharField):
+    """
+    Custom field that auto-slugifies keyword names for consistency.
+    Enforces lowercase, converts spaces to hyphens, removes non-printable chars.
+    """
+
+    def get_prep_value(self, value):
+        """Convert the value to a slug before saving to database."""
+        value = super().get_prep_value(value)
+        if value is None:
+            return value
+        if not isinstance(value, str):
+            msg = "Name attribute must be a string"
+            raise TypeError(msg)
+        # Slugify the value (lowercase, hyphens instead of spaces, etc.)
+        return slugify(value)
+
+
 class CaptureType(StrEnum):
     """The type of radiofrequency capture."""
 
@@ -571,14 +590,48 @@ def debug_opensearch_response(self) -> dict[str, Any] | None:
             return None
 
 
+class Keyword(BaseModel):
+    """
+    Model for user-entered keywords that can be associated with datasets.
+    Keywords can be associated with multiple datasets via ManyToManyField.
+
+    The name field is auto-slugified and serves as the primary key,
+    enforcing uniqueness and immutability after creation.
+    """
+
+    # Override the uuid primary key from BaseModel
+    uuid = models.UUIDField(default=uuid.uuid4, editable=False)
+
+    name = KeywordNameField(
+        max_length=255,
+        primary_key=True,
+        help_text=(
+            "The keyword slug (auto-slugified, e.g., 'RF Spectrum' → 'rf-spectrum')"
+        ),
+    )
+    datasets = models.ManyToManyField(
+        "Dataset",
+        related_name="keywords",
+        help_text="The datasets this keyword is associated with",
+    )
+
+    class Meta:  # pyright: ignore[reportIncompatibleVariableOverride]
+        ordering = ["name"]
+        verbose_name = "Keyword"
+        verbose_name_plural = "Keywords"
+
+    def __str__(self) -> str:
+        return self.name
+
+
 class Dataset(BaseModel):
     """
     Model for datasets defined and created through the API.
 
     Schema Definition: https://github.com/spectrumx/schema-definitions/blob/master/definitions/sds/abstractions/dataset/README.md
     """
 
-    list_fields = ["authors", "keywords", "institutions"]
+    list_fields = ["authors", "institutions"]
 
     STATUS_CHOICES = [
         (DatasetStatus.DRAFT, "Draft"),
@@ -604,7 +657,6 @@ class Dataset(BaseModel):
     doi = models.CharField(max_length=255, blank=True)
     authors = models.TextField(blank=True)
     license = models.CharField(max_length=255, blank=True)
-    keywords = models.TextField(blank=True)
     institutions = models.TextField(blank=True)
     release_date = models.DateTimeField(blank=True, null=True)
     repository = models.URLField(blank=True)

gateway/sds_gateway/api_methods/serializers/dataset_serializers.py

@@ -10,6 +10,7 @@
 
 class DatasetGetSerializer(serializers.ModelSerializer[Dataset]):
     authors = serializers.SerializerMethodField()
+    keywords = serializers.SerializerMethodField()
     created_at = serializers.DateTimeField(format="%m/%d/%Y %H:%M:%S", read_only=True)
     is_shared_with_me = serializers.SerializerMethodField()
     is_owner = serializers.SerializerMethodField()
@@ -24,6 +25,10 @@ def get_authors(self, obj):
         """Return the full authors list using the model's get_authors_display method."""
         return obj.get_authors_display()
 
+    def get_keywords(self, obj):
+        """Return a list of keyword names for the dataset."""
+        return [kw.name for kw in obj.keywords.filter(is_deleted=False)]
+
     def get_is_shared_with_me(self, obj):
         """Check if the dataset is shared with the current user."""
         request = self.context.get("request")

gateway/sds_gateway/api_methods/tests/factories.py

@@ -20,6 +20,7 @@
 from sds_gateway.api_methods.models import Dataset
 from sds_gateway.api_methods.models import File
 from sds_gateway.api_methods.models import ItemType
+from sds_gateway.api_methods.models import Keyword
 from sds_gateway.api_methods.models import UserSharePermission
 from sds_gateway.users.tests.factories import UserFactory
 
@@ -42,7 +43,6 @@ class DatasetFactory(DjangoModelFactory):
         doi: Random UUID4 for DOI identifier
         authors: Fixed list of test authors ["John Doe", "Jane Smith"]
         license: Fixed value "MIT"
-        keywords: Fixed list ["RF", "capture", "analysis"]
         institutions: Fixed list ["Example University"]
         release_date: Random datetime
         repository: Random URL
@@ -55,8 +55,14 @@ class DatasetFactory(DjangoModelFactory):
         is_deleted: Fixed value False
         is_public: Fixed value False
 
+    Post-generation Hooks:
+        keywords: Creates Keyword instances after Dataset creation.
+                 Pass keyword names as a list:
+                 DatasetFactory(keywords=["RF", "capture"])
+                 Pass None to skip keyword creation: DatasetFactory(keywords=None)
+
     Example:
-        # Create a basic dataset
+        # Create a basic dataset with default keywords
         dataset = DatasetFactory()
 
         # Create a dataset with custom owner
@@ -65,6 +71,12 @@ class DatasetFactory(DjangoModelFactory):
 
         # Create a public dataset
         dataset = DatasetFactory(is_public=True)
+
+        # Create a dataset with custom keywords
+        dataset = DatasetFactory(keywords=["custom", "keywords"])
+
+        # Create a dataset without keywords
+        dataset = DatasetFactory(keywords=None)
     """
 
     uuid = Faker("uuid4")
@@ -74,7 +86,6 @@ class DatasetFactory(DjangoModelFactory):
     doi = Faker("uuid4")
     authors = ["John Doe", "Jane Smith"]
     license = "MIT"
-    keywords = ["RF", "capture", "analysis"]
     institutions = ["Example University"]
     release_date = Faker("date_time")
     repository = Faker("url")
@@ -87,6 +98,46 @@ class DatasetFactory(DjangoModelFactory):
     is_deleted = False
     is_public = False
 
+    @post_generation
+    def keywords(self, create, extracted, **kwargs):
+        """Create Keyword instances for the dataset after creation.
+
+        This post-generation hook creates Keyword objects associated with this
+        dataset. By default, it creates three keywords: "RF", "capture", "analysis".
+
+        Args:
+            create: Boolean indicating if the object is being created
+            extracted: Value passed to the factory for keywords parameter.
+                      If a list is provided, those keyword names will be used.
+                      If None is provided, no keywords will be created.
+                      If not provided, default keywords will be created.
+            **kwargs: Additional keyword arguments
+
+        Example:
+            # Default keywords
+            dataset = DatasetFactory()  # Creates "RF", "capture", "analysis"
+
+            # Custom keywords
+            dataset = DatasetFactory(keywords=["custom", "test"])
+
+            # No keywords
+            dataset = DatasetFactory(keywords=None)
+        """
+        if not create:
+            return
+
+        # If extracted is None, skip keyword creation entirely
+        if extracted is None:
+            return
+
+        # Use extracted keywords or default keywords
+        keyword_names = extracted or ["RF", "capture", "analysis"]
+
+        # Create Keyword instances
+        for keyword_name in keyword_names:
+            keyword, _created = Keyword.objects.get_or_create(name=keyword_name)
+            keyword.datasets.add(self)
+
     class Meta:
         model = Dataset
 

gateway/sds_gateway/static/js/actions/DetailsActionManager.js

@@ -203,6 +203,9 @@ class DetailsActionManager {
 		// Update authors
 		this.updateAuthors(modal, datasetData.authors || []);
 
+		// Update keywords
+		this.updateKeywords(modal, datasetData.keywords || []);
+
 		// Update statistics
 		if (statistics) {
 			this.updateElementText(
@@ -442,6 +445,33 @@ class DetailsActionManager {
 		authorsContainer.textContent = authorsText;
 	}
 
+	/**
+	 * Update keywords section
+	 * @param {Element} modal - Modal element
+	 * @param {Array} keywords - Keywords array
+	 */
+	updateKeywords(modal, keywords) {
+		const keywordsContainer = modal.querySelector(".dataset-details-keywords");
+		if (!keywordsContainer) return;
+
+		// Clear existing content
+		keywordsContainer.innerHTML = "";
+
+		if (!keywords || keywords.length === 0) {
+			keywordsContainer.innerHTML =
+				'<span class="text-muted">No keywords</span>';
+			return;
+		}
+
+		// Create badges for each keyword
+		for (const keyword of keywords) {
+			const badge = document.createElement("span");
+			badge.className = "badge bg-secondary me-2 mb-2";
+			badge.textContent = keyword;
+			keywordsContainer.appendChild(badge);
+		}
+	}
+
 	/**
 	 * Update captures table
 	 * @param {Element} modal - Modal element

gateway/sds_gateway/static/js/deprecated/captureGroupingComponents.js

@@ -128,6 +128,7 @@ class FormHandler {
 
 	show(container, showClass = "display-block") {
 		container.classList.remove("display-none");
+		container.classList.remove("d-none"); // ensure Bootstrap's d-none is removed
 		container.classList.add(showClass);
 	}
 
@@ -178,6 +179,9 @@ class FormHandler {
 				document.querySelector("#step4 .dataset-description").textContent =
 					document.getElementById("id_description").value ||
 					"No description provided.";
+				const keywordsInput = document.getElementById("id_keywords");
+				document.querySelector("#step4 .dataset-keywords").textContent =
+					keywordsInput?.value.trim() || "No keywords";
 
 				// Update captures table
 				const capturesTableBody = document.querySelector(
@@ -466,6 +470,7 @@ class FormHandler {
 					method: "POST",
 					body: formData,
 					headers: {
+						"X-Requested-With": "XMLHttpRequest",
 						"X-CSRFToken": document.querySelector("[name=csrfmiddlewaretoken]")
 							.value,
 					},

gateway/sds_gateway/static/js/deprecated/datasetDetailsModal.js

@@ -123,6 +123,22 @@ class DatasetDetailsModal {
 		document.querySelector(".dataset-details-description").textContent =
 			dataset.description || "No description available";
 
+		// Populate keywords
+		const keywordsContainer = document.querySelector(
+			".dataset-details-keywords",
+		);
+		if (dataset.keywords && dataset.keywords.length > 0) {
+			keywordsContainer.innerHTML = dataset.keywords
+				.map(
+					(keyword) =>
+						`<span class="badge bg-secondary me-1 mb-1">${this.escapeHtml(keyword)}</span>`,
+				)
+				.join("");
+		} else {
+			keywordsContainer.innerHTML =
+				'<span class="text-muted">No keywords</span>';
+		}
+
 		// Format status with badge using database values
 		const statusElement = document.querySelector(".dataset-details-status");
 		if (dataset.status === "draft") {