How can I inject folder IDs into file items in my Google Drive workflow?

Below is one way to solve the problem by ensuring that each file has the folder ID (or any other relevant folder data) attached to it. The idea is to run the file search for each folder separately so that the folder context isn’t lost when multiple items combine into a single output. Here are two approaches you can consider:

---

Approach 1: Split the Workflow per Folder

1. Split the Folders into Batches:
   Use a SplitInBatches (or equivalent “Split”) node so that you process each folder individually. This way, each run carries the folder’s data (like its ID) along with the file search.

2. Run File Search per Folder:
   Within the batch loop, use the Google Drive node configured with the Search operation using the current folder’s ID. Because the search is executed per folder, the returned file items will be in the context of that folder.

3. Inject the Folder ID:
   Immediately after the search node, add a Set or Function node to add the folder ID (taken from the current input item) to every file item. For example, using a Function node:

   JSCopied
   // For each file item, add the folderId from the incoming data.
   return items.map(fileItem => {
     fileItem.json.folderId = $json.folderId; // Assumes the folder id is available as `folderId`
     return fileItem;
   });

4. Combine the Results:
   Once the loop is done, use a Merge node (with "Append" mode) to join all the individual file sets. This way, every file item retains its source folder ID.

---

Approach 2: Merge Manually Using a Function Node

If you already have two outputs (one from the List Folders and another from the file search that collectively returns files across folders), you can manually merge the context:

1. Collect Folder Context First:
   Save the folder information (for example, in a variable or by passing it along on the “List Folders” output) before running the file search.

2. Run Your File Search Node:
   Run the node that searches files but note that the output loses the folder context because the node returns results independently.

3. Inject the Folder ID Using a Function Node:
   Use a Function node where you iterate through each file item and inject the folder ID based on matching criteria (like a reference to some parent property). In the Function node you might do something like:

   JSCopied
   // Example: assume you stored folder info somewhere accessible (e.g. in workflow static data)
   const folderContext = getWorkflowStaticData('global').folderById; 
   // If each file item has a property indicating which folder it came from (or if you can match the file using another heuristic)  
   return items.map(item => {
     // Here, you might determine the folder lookup key. For demonstration, assume file contains folderKey.
     const folderId = folderContext[item.json.folderKey] || null;
     item.json.folderId = folderId;
     return item;
   });

   Note:
   For this to work reliably, your execution needs to preserve the mapping between folders and file search queries. It’s often simpler to use the first approach so that there is a direct one-to-one relationship making the injection trivial.

---

Key Takeaways

• Process Per Folder:
  Execute file searches on a per-folder basis (using a SplitInBatches node) so that each file search is naturally paired with its folder context.

• Use a Set/Function Node Immediately After the Search:
  After searching, immediately inject the folder ID from the split batch’s data. This avoids issues with expressions like {{ $('List Folders').item.json.name }} losing the matching relation.

• Merge Results After Injection:
  Once every file item includes its folder information, merge the results using a Merge node (set to Append mode) so that the workflow can continue processing files with context intact.

Using one of the above approaches should help you avoid losing the folder context while processing file items from Google Drive within your n8n workflow.