The Input Table parameter value can be a feature layer, a table view, or a raster layer with an attribute table. If a data path is used, the layer will be created with the join. The join will always reside in the layer, not with the data.
To make a permanent join, either use the Join Field tool or use the joined layer as input to one of the following tools: Copy Features, Copy Rows, Export Features, or Export Table. When saving the results to a new feature class or table, the Maintain fully qualified field names environment can be used to control whether the joined output field names will be qualified with the name of the table the field came from. Field aliases are persisted from the layer to the output, except when the output is a shapefile.
Use the Make Query Layer, Create Database View, or Make Aggregation Query Layer tools to optimize join performance and for more capabilities when you want to join enterprise geodatabase or SQLite database data.
If the input is a feature class or dataset path, this tool will create and return a new layer with the result of the tool applied.
When a one-to-many join is produced by the join, the result of the join can be viewed in the attribute table, where a warning message will indicate if the table has duplicate object IDs. Because many geoprocessing tools do not support data with duplicate object IDs and processing such data can produce unexpected results, it is recommended that you first copy the joined layer to a new feature class using the Export Features tool. Then use the new feature class as input to other geoprocessing tools.
Optionally, set the Join Operation parameter to Join one to first to prevent duplicate object IDs.
The Join Operation parameter has three states to adjust the cardinality. The default is no value and will allow the data source to attempt a one-to-many join. The Join one to many option will work only on specific data sources that have an Object ID field. The Join one to first option will use the first match in the table, which may result in different outputs if the Object ID field is changed or the workspace the table is copied to changes. One-to-first joins are not case sensitive; one-to-many joins are case sensitive. If the joined table does have an Object ID, the first match will be the record with the lowest Object ID.
The following tables include possible outcomes of performing a join with various inputs.
The first table shows a one-to-many join. Keeping only matching records will have no effect, as all records have matches.
| Input table | Join table | Result |
|---|
Input field | Type | Join field | Value | Input field | Type | Join field | Value |
1 | A | 1 | 100 | 1 | A | 1 | 100 |
2 | B | 2 | 200 | 2 | B | 2 | 200 |
| 1 | 300 | 1 | A | 1 | 300 |
2 | 400 | 2 | B | 2 | 400 |
Add Join example: One-to-many join when each table has an Object ID field
The second table uses a join table with no Object ID field; only a one-to-first join is possible. A one-to-first join is also only possible if each table is from a different workspace. A one-to-first join is not a case-sensitive match.
| Input table | Join table | Result |
|---|
Input field | Type | Join field | Value | Input field | Type | Join field | Value |
1 | A | 1 | 100 | 1 | A | 1 | 100 |
2 | B | 2 | 200 | 2 | B | 2 | 200 |
| 3 | 300 | |
4 | 400 |
Add Join example: One-to-first join when either table does not have an Object ID field
In the last table, the input table has more records than the join table. Keeping all records will keep all of the matching records plus the records from the input table that did not match.
| Input table | Join table | Result |
|---|
Input field | Type | Join field | Value | Input field | Type | Join field | Value |
1 | A | 1 | 100 | 1 | A | 1 | 100 |
2 | B | 2 | 200 | 2 | B | 2 | 200 |
3 | C | 1 | 300 | 1 | A | 1 | 300 |
4 | D | 2 | 400 | 2 | B | 2 | 400 |
| 3 | C | <Null> | <Null> |
4 | D | <Null> | <Null> |
Add Join example: One-to-many join when each table has an Object ID field and the Keep all input records parameter is checked
The input table must have an Object ID field to perform a one-to-many join and be in the same workspace.
Records from the join table can be matched to more than one record if the join table has an Object ID field; otherwise, a one-to-first join will be performed.
When joining tables, the default option is to keep all records. If a record in the input table doesn't have a match in the join table, that record is given null values for all the fields being appended into the input table from the join table.
| Input table | Join table | Result |
|---|
Input field | Type | Join field | Value | Input field | Type | Join field | Value |
1 | A | 1 | 100 | 1 | A | 1 | 100 |
2 | B | 2 | 200 | 2 | B | 2 | 200 |
3 | C | 1 | 300 | 1 | A | 1 | 300 |
4 | D | 2 | 400 | 2 | B | 2 | 400 |
| 3 | C | <Null> | <Null> |
4 | D | <Null> | <Null> |
When the Keep all input records parameter is unchecked, if a record in the input table doesn't have a match in the join table, that record is removed from the resultant output. If the input table is the attribute table of a layer, features that don't have data joined to them are not shown on the map.
| Input table | Join table | Result |
|---|
Input field | Type | Join field | Value | Input field | Type | Join field | Value |
1 | A | 1 | 100 | 1 | A | 1 | 100 |
2 | B | 2 | 200 | 2 | B | 2 | 200 |
| 3 | 300 | |
4 | 400 |
The Calculate Field tool will update the first record encountered when using a one-to-many layer and skip the remainder duplicate records. When editing the joined layer field values manually in the table view, the last edit made remains..
Field properties, such as aliases, visibility, and number formatting, are maintained when a join is added or removed.
The join persists only for the duration of the layer. A layer can be retained by saving the ArcGIS Pro session or by saving it to a layer file using the Save Layer To File tool.
To see the results of a join created in a script tool, the tool must include the layer as a derived output parameter. Similarly, the Updated Input Layer or Table View parameter must be set as a derived output parameter in a model tool to see the joined results.
In the resulting table, fields will be prefixed with the input's name and a period (.), and all fields from the join table will be prefixed with the join table name and a period as the default.
For example, joining landuse, which has fields A and B, to lookup_tab, which has fields C and D, will result in a layer or table view with landuse.A, landuse.B, lookup_tab.C, and lookup_tab.D fields.
A layer must have unique field names. If both the input and join tables have the same name and are in different workspaces, a join cannot be performed without creating a poorly defined layer.
Indexing the input field and join field can improve performance. If the Index join fields parameter is checked, an attribute index will be added to both fields. Alternatively, each joining field can be indexed with the Add Attribute Index tool.
If the join results are unexpected or incomplete, review whether the input field and join field are indexed. If the fields are not indexed, try adding an index. If the fields are indexed, try deleting and re-adding the index to correct any problems with the index. Alternatively, check the Rebuild Join Field Indexes parameter to remove existing indexes and rebuild them.
-
If the input layer or table view's fields were modified (renamed or hidden) using the Field Info parameter in the Make Feature Layer or Make Table View tool, the field modifications will not be included in the output joined layer or table view.
The definition query of the join table will be applied to the input layer or table view by adding a new active query. The previous query is preserved and set to inactive so that the query can be disabled from the joined table if needed. The definition query can be removed using the Remove Join tool.
If the join table has a definition query, the Keep all input records parameter will have no effect. Manually updating the definition query by appending or OBJECTID is null can fix this if appropriate.
The Validate Join tool can be used to validate a join between two layers or tables to determine if the layers or
tables have valid field names and Object ID fields, if the join
produces matching records, if the join is a one-to-one or one-to-many join, and
other properties of the join.
A button to validate the join is available on the tool dialog box for ease of use.
This tool ignores selections on the input and join tables. The Join Field tool supports selections. To only join with a selected subset, create a selection layer and use it as input to this tool. Join layer properties are copied when you create a selection layer.
