fix(dataset): make api fields optional and add guards to prevent null fields

src/common/pipes/include-validation.pipe.ts

@@ -29,7 +29,9 @@ export class IncludeValidationPipe implements PipeTransform<
       ? inValue
       : isJsonString(inValue)
         ? JSON.parse(inValue ?? "{}").include
-        : Array(inValue);
+        : inValue.includes(",")
+          ? IncludeValidationPipe.splitCsv(inValue)
+          : Array(inValue);
 
     includeValueParsed?.map((field) => {
       let relationField = field;
@@ -46,4 +48,15 @@ export class IncludeValidationPipe implements PipeTransform<
 
     return inValue;
   }
+
+  /**
+   * Split a comma-separated string into individual relation values.
+   * Handles CSV-style serialization used by some OpenAPI client generators.
+   */
+  static splitCsv(inValue: string): string[] {
+    return inValue
+      .split(",")
+      .map((s) => s.trim())
+      .filter(Boolean);
+  }
 }

src/common/utils.ts

@@ -1340,6 +1340,13 @@ export function decodeMetadataKeyStrings(keys: string[]): string[] {
   return keys.map((key) => decodeURIComponentExtended(key));
 }
 
+export function filterNullFromArray<T>(
+  value: (T | null | undefined)[] | undefined,
+): T[] {
+  if (!value) return [];
+  return value.filter((item): item is T => item !== null && item !== undefined);
+}
+
 export function parseDate(dateString?: string): Date | undefined {
   if (!dateString) return undefined;
   const parsedDate = new Date(dateString);

src/datasets/datasets-public.v4.controller.ts

@@ -303,6 +303,8 @@ export class DatasetsPublicV4Controller {
     type: String,
     required: false,
     isArray: true,
+    style: "form",
+    explode: true,
   })
   async findByIdPublic(
     @Param("pid") id: string,

src/datasets/datasets.v4.controller.ts

@@ -713,6 +713,8 @@ export class DatasetsV4Controller {
     type: String,
     required: false,
     isArray: true,
+    style: "form",
+    explode: true,
   })
   async findById(
     @Req() request: Request,

src/datasets/dto/output-dataset.dto.ts

@@ -1,8 +1,17 @@
-import { ApiProperty, PartialType } from "@nestjs/swagger";
+import { ApiProperty, PartialType, getSchemaPath } from "@nestjs/swagger";
 import { CreateDatasetDto } from "./create-dataset.dto";
-import { IsDateString, IsString } from "class-validator";
-import { decodeScientificMetadataKeys } from "src/common/utils";
-import { Transform } from "class-transformer";
+import { IsArray, IsDateString, IsOptional, IsString } from "class-validator";
+import {
+  decodeScientificMetadataKeys,
+  filterNullFromArray,
+} from "src/common/utils";
+import { Transform, Type } from "class-transformer";
+import { OutputOrigDatablockDto } from "src/origdatablocks/dto/output-origdatablock.dto";
+import { Datablock } from "src/datablocks/schemas/datablock.schema";
+import { OutputAttachmentV4Dto } from "src/attachments/dto/output-attachment.v4.dto";
+import { Instrument } from "src/instruments/schemas/instrument.schema";
+import { ProposalClass } from "src/proposals/schemas/proposal.schema";
+import { OutputSampleDto } from "src/samples/dto/output-sample.dto";
 
 export class OutputDatasetDto extends CreateDatasetDto {
   @ApiProperty({
@@ -51,12 +60,13 @@ export class OutputDatasetDto extends CreateDatasetDto {
 
   @ApiProperty({
     type: String,
-    required: true,
+    required: false,
     description:
       "Version of the API used when the dataset was created or last updated. API version is defined in code for each release. Managed by the system.",
   })
+  @IsOptional()
   @IsString()
-  version: string;
+  version?: string;
 
   @ApiProperty({
     type: Object,
@@ -66,6 +76,106 @@ export class OutputDatasetDto extends CreateDatasetDto {
   })
   @Transform(({ value }) => decodeScientificMetadataKeys(value))
   declare scientificMetadata?: Record<string, unknown>;
+
+  @Transform(({ value }) => filterNullFromArray<string>(value))
+  declare keywords?: string[];
+
+  @Transform(({ value }) => filterNullFromArray<string>(value))
+  declare sharedWith?: string[];
+
+  @Transform(({ value }) => filterNullFromArray<string>(value))
+  declare proposalIds?: string[];
+
+  @Transform(({ value }) => filterNullFromArray<string>(value))
+  declare sampleIds?: string[];
+
+  @Transform(({ value }) => filterNullFromArray<string>(value))
+  declare instrumentIds?: string[];
+
+  @Transform(({ value }) => filterNullFromArray<string>(value))
+  declare inputDatasets?: string[];
+
+  @Transform(({ value }) => filterNullFromArray<string>(value))
+  declare usedSoftware?: string[];
+
+  @Transform(({ value }) => filterNullFromArray<string>(value))
+  declare principalInvestigators?: string[];
+
+  // ---------------------------------------------------------------------------
+  // Includable relation fields — populated via ?include query parameter
+  // ---------------------------------------------------------------------------
+
+  @ApiProperty({
+    type: "array",
+    items: { $ref: getSchemaPath(OutputOrigDatablockDto) },
+    required: false,
+    description:
+      "Containers that list all files and their attributes which make up a dataset. Included when ?include=origdatablocks is used.",
+  })
+  @IsOptional()
+  @IsArray()
+  @Type(() => OutputOrigDatablockDto)
+  origdatablocks?: OutputOrigDatablockDto[];
+
+  @ApiProperty({
+    type: "array",
+    items: { $ref: getSchemaPath(Datablock) },
+    required: false,
+    description:
+      "Archived file blocks with checksums. Included when ?include=datablocks is used.",
+  })
+  @IsOptional()
+  @IsArray()
+  @Type(() => Datablock)
+  datablocks?: Datablock[];
+
+  @ApiProperty({
+    type: "array",
+    items: { $ref: getSchemaPath(OutputAttachmentV4Dto) },
+    required: false,
+    description:
+      "Small attachments such as preview images. Included when ?include=attachments is used.",
+  })
+  @IsOptional()
+  @IsArray()
+  @Type(() => OutputAttachmentV4Dto)
+  attachments?: OutputAttachmentV4Dto[];
+
+  @ApiProperty({
+    type: "array",
+    items: { $ref: getSchemaPath(Instrument) },
+    required: false,
+    description:
+      "Instruments associated with the dataset. Included when ?include=instruments is used.",
+  })
+  @IsOptional()
+  @IsArray()
+  @Type(() => Instrument)
+  instruments?: Instrument[];
+
+  @ApiProperty({
+    type: "array",
+    items: { $ref: getSchemaPath(ProposalClass) },
+    required: false,
+    description:
+      "Proposals associated with the dataset. Included when ?include=proposals is used.",
+  })
+  @IsOptional()
+  @IsArray()
+  @Type(() => ProposalClass)
+  proposals?: ProposalClass[];
+
+  @ApiProperty({
+    type: "array",
+    items: { $ref: getSchemaPath(OutputSampleDto) },
+    required: false,
+    description:
+      "Samples associated with the dataset. Included when ?include=samples is used.",
+  })
+  @IsOptional()
+  @IsArray()
+  @Type(() => OutputSampleDto)
+  samples?: OutputSampleDto[];
 }
 
 export class PartialOutputDatasetDto extends PartialType(OutputDatasetDto) {}

src/datasets/schemas/dataset.schema.ts

@@ -215,8 +215,8 @@ export class DatasetClass extends OwnableClass {
     description:
       "Array of tags associated with the meaning or contents of this dataset. Values should ideally come from defined vocabularies, taxonomies, ontologies or knowledge graphs.",
   })
-  @Prop({ type: [String], required: false })
-  keywords: string[];
+  @Prop({ type: [String], required: false, default: [] })
+  keywords: string[] = [];
 
   @ApiProperty({
     type: String,
@@ -366,7 +366,7 @@ export class DatasetClass extends OwnableClass {
     description:
       "First and last name of principal investigator(s). Multiple PIs can be provided as separate strings in the array. This field is required if the dataset is a Raw dataset.",
   })
-  @Prop({ type: [String], required: false })
+  @Prop({ type: [String], required: false, default: [] })
   principalInvestigators?: string[];
 
   @ApiProperty({
@@ -420,7 +420,7 @@ export class DatasetClass extends OwnableClass {
     description:
       "The ID of the proposal to which the dataset belongs to and it has been acquired under.",
   })
-  @Prop({ type: [String], ref: "Proposal", required: false })
+  @Prop({ type: [String], ref: "Proposal", required: false, default: [] })
   proposalIds?: string[];
 
   @ApiProperty({
@@ -429,7 +429,7 @@ export class DatasetClass extends OwnableClass {
     description:
       "Single ID or array of IDS of the samples used when collecting the data.",
   })
-  @Prop({ type: [String], ref: "Sample", required: false })
+  @Prop({ type: [String], ref: "Sample", required: false, default: [] })
   sampleIds?: string[];
 
   @ApiProperty({
@@ -438,7 +438,7 @@ export class DatasetClass extends OwnableClass {
     description:
       "Id of the instrument or array of IDS of the instruments where the data contained in this dataset was created/acquired.",
   })
-  @Prop({ type: [String], ref: "Instrument", required: false })
+  @Prop({ type: [String], ref: "Instrument", required: false, default: [] })
   instrumentIds?: string[];
 
   @ApiProperty({
@@ -447,7 +447,7 @@ export class DatasetClass extends OwnableClass {
     description:
       "Array of input dataset identifiers used in producing the derived dataset. Ideally these are the global identifier to existing datasets inside this or federated data catalogs.",
   })
-  @Prop({ type: [String], required: false })
+  @Prop({ type: [String], required: false, default: [] })
   inputDatasets?: string[];
 
   @ApiProperty({
@@ -456,7 +456,7 @@ export class DatasetClass extends OwnableClass {
     description:
       "A list of links to software repositories which uniquely identifies the pieces of software, including versions, used for yielding the derived data.",
   })
-  @Prop({ type: [String], required: false })
+  @Prop({ type: [String], required: false, default: [] })
   usedSoftware?: string[];
 
   @ApiProperty({

src/datasets/types/dataset-lookup.ts

@@ -70,7 +70,7 @@ export const DATASET_LOOKUP_FIELDS: Record<
             $expr: {
               $anyElementTrue: {
                 $map: {
-                  input: "$relationships",
+                  input: { $ifNull: ["$relationships", []] },
                   as: "relationship",
                   in: {
                     $and: [

src/origdatablocks/dto/output-origdatablock.dto.ts

@@ -1,6 +1,6 @@
 import { ApiProperty, PartialType } from "@nestjs/swagger";
 import { CreateOrigDatablockDto } from "./create-origdatablock.dto";
-import { IsDateString, IsString } from "class-validator";
+import { IsDateString, IsOptional, IsString } from "class-validator";
 
 export class OutputOrigDatablockDto extends CreateOrigDatablockDto {
   @ApiProperty({
@@ -41,12 +41,13 @@ export class OutputOrigDatablockDto extends CreateOrigDatablockDto {
 
   @ApiProperty({
     type: String,
-    required: true,
+    required: false,
     description:
       "Version of the API used when the origdatablock was created or last updated. API version is defined in code for each release. Managed by the system.",
   })
+  @IsOptional()
   @IsString()
-  version: string;
+  version?: string;
 }
 
 export class PartialOutputOrigDatablockDto extends PartialType(

src/origdatablocks/origdatablocks.v4.controller.ts

@@ -748,6 +748,8 @@ export class OrigDatablocksV4Controller {
     type: String,
     required: false,
     isArray: true,
+    style: "form",
+    explode: true,
   })
   async findById(
     @Req() request: Request,

src/samples/dto/output-sample.dto.ts

@@ -1,5 +1,5 @@
 import { ApiProperty } from "@nestjs/swagger";
-import { IsDateString, IsString } from "class-validator";
+import { IsDateString, IsOptional, IsString } from "class-validator";
 import { CreateSampleDto } from "./create-sample.dto";
 import { Transform } from "class-transformer";
 import { decodeScientificMetadataKeys } from "src/common/utils";
@@ -47,6 +47,7 @@ export class OutputSampleDto extends CreateSampleDto {
     description:
       "Version of the API used when the dataset was created or last updated. API version is defined in code for each release. Managed by the system.",
   })
+  @IsOptional()
   @IsString()
   version?: string;