Hey and welcome to video number seven of the advanced course for n8n. This video will cover handling files, which is quite an advanced feature of n8n workflow building.

When dealing with files, the node output will have a new option called binary view, alongside JSON table and schema. When reading images or documents, the binary data will let you preview or download those files directly onto your computer.

All of this will help with, well, handling files. The binary output will give you additional information about the file name, directory, file type, size, etc. When dealing only with files, the table JSON and schema views will be empty.

Here you see an example of a binary view. I made an HTTP request to retrieve a file, and here I have different information about that file, name, directory, extension, size, etc. By clicking on The View, I'm able to preview this file, but I could also download it onto my machine, so I can view it directly.

Here, for an SVG format, we can obviously preview the image when dealing with files. So what are the key nodes that we're going to be able to use when handling files?

All the nodes related to handling files are in the file category. Some of the useful nodes will help you with compressing or decompressing files, using the zip format you're likely familiar with, and will allow you to read or write files from the disk.

This can be particularly useful if you're self hosting, and then or converting JSON objects into or from the file format.

To illustrate how we can use files in n8n, I have a HTTP request node here that is making a get HTTP request to n8n logo, and here we can see the expected response format is a file.

Going to put it in a field called logo. When I execute here, the first thing we're going to see is the binary view which lets us uh preview but also download the image, and the table and JSON and schema views are going to be completely empty.

If uh, back and behind this node I drag another HTTP request, this time I'm just um, same process, just looking for a different kind of file and I execute, here we can see that binary items um behave a bit differently than regular items and append uh each other.

So in the previous execution we only had logo and now we have logo and workflow image.

Now that we have these two files, um if I want to for example store them on my disk or if I want to compress them to be able to send them, I can use the compress files.

Obviously the operation is going to be compressed but we can use the same node to decompress files and I can just add here the names logo and workflow, set the output format as a zip, give it a name compress.

zip for example, and when I execute the node we go from two files down into a single file that I can download to um test.

Now that we've looked at how uh we can get multiple files with an HTTP request and compress them into a single zip file, we can look at the opposite example and request a zip file, decompressing the data and splitting it up.

The first thing we need to do obviously is get the uh file.

So here we have the same as last time, a get HTTP request that goes and gets a zip file on a server.

By executing here, once again we see table and JSON are empty and we have a binary file that we can again download here.

Using the uh decompress node here, we can um extract all of the different files that are in data. Uh here we can see that we have three different files: file zero, file one, and file two, that are an Unsplash image, a second Unsplash image, and a third one.

Again, all of the information about the file can be found here.

Here, apparently, we have three JPEGs. The next step here is going to be to split this out. As you can see, we only have one item.

So if, in um, the next part of the workflow, we're going to want to deal with these different images separately, we're going to have to split out this single item into different binary items.

For this, uh, we can use the code node. Um, this is something that you can find in the templates that will allow you to simply take these binary files and split them out.

So when I execute the node here, we see that we have this time three different binary items, um, all named data, and that correspond to the three files that we saw earlier, this time split out.

Here we can again see we have a single item here, and then we have three items here.

So following on, we can now build our workflow, um, looping through all of these different items.

Thanks for listening to the seventh video of the n8n Advanced course, where we covered handling files. In the next video, we'll cover some Enterprise features that'll take your workflow building to the next level.

n8n Advanced Course (7/8) - Handling files

n8n Advanced Course (7/8) - Handling files

Scripts