A row identifier can be established for any Socrata dataset. A row identifier is a specific column of a dataset that contains values that uniquely identify each row in the dataset. In other words, the value in this column cannot be the same for any two rows in the dataset, it must be different for each row. A common column to use is an ID column with some kind of number or code that identifies that row of data. Row identifiers are required for using the Upsert function within DataSync.
An example of a Row Identifier is the 'Reference Number' column on Edmonton's 311 dataset: https://data.edmonton.ca/Indicators/311-Explorer/ukww-xkmj
How to set a row identifier
Step 1: The row identifier is set through the "Edit Metadata" page, which can be accessed one of two ways.
Step 2: From the Primer page, select "Edit Metadata" in the About section of the page. If viewing the "Grid View", select the About pane on the dataset which will open up a box on the screen. From here select the "Edit Metadata" button.
Step 3: Once on the edit metadata page, Scroll to the subheader called “API Endpoint”, in this section you can select the row identifier from the drop down menu that contains all the columns within the dataset.
Step 4: After selecting the appropriate column. Click “Save” at the bottom.