The only settings you can change for task outputs is the Name and Description. All other settings are read-only. By default, the Service Editor copies the name and description of the parameter from the tool's item description.
The information displayed in the Service Editor for output parameters depends on the type of output. There are two basic types of output: datasets and scalars.
- Datasets are features, tables, files, rasters, folders, geodatabases, and so on.
- Scalars are numbers, Booleans, strings, and so on—anything that is not a dataset.
The information displayed for output dataset depends on the type of output dataset, as described below.
Features and tables
For output features and tables, the attribute schema (fields and their type) is displayed. For features, the schema includes the geometry type of the output. The list of fields is the same as the output dataset or layer in the result. The schema cannot be modified in the Service Editor. To change the schema, you need to rerun the tool using different inputs in order to create a new result with a different output that you then publish.
Output schema is dynamic
The output schema shown in the Service Editor is read from the output dataset or layer in the result you are publishing. It's not uncommon that this schema might change depending on what inputs are provided—that is, it is dynamic (can change) versus static (never changes). For example, the Generate Near Table tool will optionally add the fields NEAR_X, NEAR_Y, and NEAR_ANGLE to the output features based on the values of its Location and Angle parameters. The task you are publishing may be similar to the Generate Near Table tool in that the output schema and may vary depending on other parameter values set by the client. Other tools, and tasks created from these tools, may output different geometry types (point, line, or polygon) depending on other parameter values set by the client.
When your task executes, the schema sent to the client is whatever is produced by the execution of the task. This schema may be different than what is displayed in the Service Editor.
Rasters and files
Output raster datasets will show a Data type of Raster dataset, and output files will show a Data type of File. There is no schema associated with either of these types.
Nontransferable output datasets and layers
If the output of your task is a dataset other than a feature class, table, raster, or file, then the dataset cannot be transported back to the client. The Data type of the output will be String, as illustrated below. If the output of your task is a layer that references a nontransferable dataset, it will also be converted to a string.
When your task executes, the name of the dataset or layer will be returned to the client.
In most cases, where you have a non-transportable output dataset, you'll use a result map service so the client can view the result as a map. No dataset is transported to the client, only a map of the data.
See Input modes and parameter data types for more information about non-transportable datasets.
A scalar data type contains anything that isn't a dataset. The following data types are fully supported by all clients and can be sent and received across the Internet without conversion: Boolean, Date, Double, Linear Unit, Long, and String. Any other scalar data type is converted to a string and is referred to as unsupported or non-transportable.