Describe the problem/error/question
Our current HRIS system doesn’t integrate into Okta, so I’m trying to build a flow that periodically gets employee details to update the corresponding user in Okta.
The problem, however, is that the HRIS system uses employees’ legal name, whereas users have opted for a preferred name in Okta. For example:
- Some users have a surname preposition like in the HRIS system, but opt to drop that in their user account (eg. Baron von Trappe | Baron Trappe)
- Hispanic users often have multiple surnames which is present in the HRIS system, but they opt to keep just the first surname in their user account (eg. Juan Fernando Hernandez | Juan Fernando)
- Many users have a formal legal given name, but opt for a shortened version (eg. William Smith | Will Smith)
As the Merge node only allows me to specify the field names to match, but not manipulate it with an expression to conduct the match, I want to resort to a Code node to handle the matching.
I’ve used the following code in the node:
const hrisUsers = $('GET HRIS Employees').all().map(employee => {
return {
json: employee.json.attributes
};
});
const oktaUsers = $('GET Okta Users').all().map(user => {
return {
json: user.json.profile
};
});
const surnamePrepositions = ['da', 'de', 'del', 'der', 'di', 'van', 'von', 'ver'];
const firstNameMap = {
"Andrew": ["Andy"],
"Christopher": ["Chris"],
"David": ["Dave"],
"Donald": ["Don"],
"Edward": ["Ed", "Eddy"],
"Gregory": ["Greg"],
"Jacob": ["Jake"],
"Jonathan": ["Jon", "John"],
"Matthew": ["Matt"],
"Michael": ["Mike"],
"Peter": ["Pete"],
"Richard": ["Rich", "Rick"],
"Robert": ["Rob"],
"Samuel": ["Sam"],
"Stuart": ["Stu"],
"Thomas": ["Tom", "Tommy"],
"Timothy": ["Tim"],
"William": ["Will", "Bill"],
};
function normalizeLastName(lastName) {
if(lastName != null){
let parts = lastName.split(' '); // Split the last name into parts
// Filter out surname prepositions
parts = parts.filter(part => !surnamePrepositions.includes(part.toLowerCase()));
// Return the first remaining part, or an empty string if none
return parts.length > 0 ? parts[0] : '';
}
}
function formalizeFirstName(firstName) {
for (const [key, value] of Object.entries(firstNameMap)) {
if (value.includes(firstName)) {
return key;
}
}
return firstName;
}
// Create a new array to hold the merged users
const matchedUsers = [];
const unmatchedUsers = [];
// Iterate over hrisUsers
for (const hrisUser of hrisUsers) {
const firstName = formalizeFirstName(hrisUser.json.first_name);
const lastName = normalizeLastName(hrisUser.json.last_name);
// Find matching okta user
const matchedOktaUser = oktaUsers.find(oktaUser =>
formalizeFirstName(oktaUser.json.firstName) === firstName && normalizeLastName(oktaUser.json.lastName) === lastName
);
// If a match is found, merge the user data
if (matchedOktaUser) {
matchedUsers.push({
hrisData: hrisUser.json,
oktaData: matchedOktaUser.json
});
} else {
// If no match is found, push to unmatchedUsers
unmatchedUsers.push(hrisUser.json);
}
}
// Return the merged users
return matchedUsers;
What is the error message (if any)?
While the Code works if I am using pinned data in the GET HRIS Employees and GET Okta Users HTTP requests, it fails if I execute the workflow with those two nodes in separate branches, as the Code node executes after one of the HTTP requests completes but the other branch has not initiated, leading to the following error: Referenced node is unexecuted [line 2]
If I try to circumvent this by having both HTTP requests occur within the same branch, then the multiple items returned by the first node causes the subsequent node to run that many multiple times as a batch. Toggling Execute Once doesn’t appear to do anything.
A pointer in the right direction would be greatly appreciated, whether on getting separate branches working, or how to prevent multiple runs from being initiated by a previous node’s results.
Please share your workflow
Share the output returned by the last node
Referenced node is unexecuted [line 2]
Information on your n8n setup
- n8n version: 1.63.4
- Database (default: SQLite): default
- n8n EXECUTIONS_PROCESS setting (default: own, main): default
- Running n8n via (Docker, npm, n8n cloud, desktop app): npm
- Operating system: Ubuntu 24.04