IPFS Launchpad: IPFS, 30 Jan 2023

Previous Meeting

Next Meeting

⏯

youtube image

►

From YouTube: IPLD Pathing

Description

This tutorial goes over how to use the IPFS command line API to experience IPLD pathing features which allow you navigate a directed acyclic graph (DAG). We will be navigating the DAG of the ipld.io website that is hosted on IPFS.

This content is created as a part of Protocol Labs' Launchpad program. You can get resources and written instructions in our learning platform at https://curriculum.pl-launchpad.io

See all links, commands, and instructions in the tutorial at https://curriculum.pl-launchpad.io/tutorials/ipld/ipld-pathing/
Join Protocol Labs Launchpad: https://pl-launchpad.io/

Chapters:
0:00 Introduction
00:47 Install IPFS
02:21 Install jq
03:20 Find the CID for ipld.io website
03:53 Inspect the Root
05:12 Load a Page With IPFS Pathing
06:00 Dive Deep into a DAG
07:356Load a Page With IPLD Pathing
08:45 Switching Codecs
10:25 Recap

A

A

Welcome back to this tutorial series on ipld in this video we're going to be working with the ipld pathing tools to explore the ipld. I o website in the command line.

A

If this is possible, because the ipld website is hosted on ipfs thanks to a service called Fleek which can display a static website whose code is hosted on GitHub and we will be using the command line today, and so, if you're not comfortable with using it, then just come along for the right and when you have time I highly recommend you go through the ipfs and ipld modules on our site at curriculum.pl-launchpad.io.

A

First thing: we're going to do is to install the golang implementation of ipfs on our local computer and, if we haven't already, we will head over to docs.ipfs.tech slash, install slash command line, we'll scroll down to find our operating system.

A

I am using wsl2 on Windows running Ubuntu 20, so the Linux commands will work for me. Now we can open up. Our command line can copy and paste the first command to download the Linux binary.

A

Double check: if we have it perfect, we need to unzip the binary.

A

Now we need to CD into the cueble folder that was just created.

A

Now we can run the install command.

A

Now we can test to see if everything was installed correctly.

A

Perfect now the last thing we need to do is initialize our ipfs node on our computer.

A

This will give us our PID so that we can be uniquely identified on the ipfs network.

A

The next command we will need is called JQ, which is a command line tool to format Json. In a way that is easier for us to see. So we will head over to instead All In, dot, GitHub dot, IO Dash, JQ, slash downloads.

A

Here we can see that JQ is available on Linux Mac and windows. Mac is as easy as doing Brew install JQ and for Linux, which I will run as well. I can run sudo apt, install JQ.

A

As I mentioned earlier, this website is hosted on ipfs, so that means that this website has a root CID, which we can find.

A

So, to do that, we first need to start up our ipfs node by typing out ipfs Daemon perfect and continue will open up a new tab or a new terminal, after which we will type ipfs resolve Dash ipns.ipld.io.

A

And, as you can see here is the root CID for the IPL Dio website, foreign.

A

Within our command line, using the ipld API to do that, we will type out ipfs, dag, get the root CID and then we're going to use it pipe and pipe it to JQ. So we can see here a couple. Things first, is that JQ nicely formatted the Json for us, allowing us to see clearly the ipld native codec of the DAC.

A

We can see that each Merkel, dag and ipld will have a data field and a links field where the data field will hold the actual bytes of a block and a link will hold the named hashes for the children block, and so you might ask well how does fleek know how to display this as a website instead of just hosting a directory?

A

Well, that is due to the fact that Fleek runs its own ipfs node and, with that ipfs will look in a directory to find a file called index.html, and so, if we are to just scroll down a little a little bit here, we would see that that file does in fact exist in the root directory.

A

And now that ipfs knows that the vowel exists, it pulls in the raw bytes and displays it in the Bowser, and we can do this similar thing too, by typing out, ipfs, dag get and then we'll use the same CID and then we'll do Dash, index.html and here we'll pipe it to JQ again, and so what we are seeing here are the raw bytes of the index.html file. This is just encoded in base64 and I will show you how to get the decoded version of this.

A

But first we need to understand the difference between ipfs pathing and IPL d-pathing. Currently, we are using ipfs pathing to access the raw by Tool file and we can use this to dive deeper into a dag but I, don't know what are any of the children blocks to this node, but I can find out by looking at the links field. So we can see here that there are no links in the links field, meaning There Are, No Children blocks, meaning this is a leaf Block in the Dag.

A

But if we go back to the root node, we can see it has children nodes in the links field and we can access any block like a normal file system by just typing the name of the link in the path. So, for example, I can go into the docs name, see the children for the docs block. We have Advanced Data layouts codecs data model and we can go even further by going into codex.

A

And so we got a little bit further into the dag, and this is neat because we have in essence navigated to this page in the doc site.

A

But not only that we have navigated a couple directories into the ipfs equivalent of this ipld repo I could go even further to find out what's inside known by doing by adding that to the path you can see here that these files in here in this subdirectory are equivalent to the blocks that we have in our Merkel Dag.

A

So, as I mentioned, this is an example of ipfs pathing, but we want to move away from this to use IPL deep pathing in this tutorial. The reason for this is because ipld is a modular component and can be used away from ipfs to be used with any other block, storage or content address system.

A

For that to happen, ipld needs to know of the data field and the links Field built into a native ipld codec like dagpb, which is shown here so, for example, to get back to the index.html file in the root directory.

A

We can type out ipfs Dag get ipld Dash the root CID Dash links, Dash, seven.

A

So you can see here we got back just the block that contains the index.html file.

A

And so the next step will be to dive into the index.html file to look at the raw bytes. We would do links.7 slash now. Here's where we can decode this using ipld, pathing and IPL the options to get the look at the raw data. We will do that by getting back going back to our original command um and adding in dash dash output, Dash codec equals Raw, but before we can press enter, we need to include the data field here at the end, so that the option works properly and boom.

A

We can see that we got back the raw data of the index.html block. The reason why I used 7 in this command here was because, according to the ipld dag, the index.html file is the seventh link down in the root directory. So we can I. We can see that by going to the root directory typing this to JQ, let's get rid of the raw option, because it will not work in this scenario and we can count from the top down until we find index.html.

A

We've got one two, three, four, five, six, seven eight so I said eight, but it's actually the seventh one down, because IPL D counts start from zero.

A

So, as you have noticed with the use of the ipld name at the beginning of the path and the use of the ipfs dag get commands, it lets our ipfs node know how to interpret the links, data and the hash fields to Traverse a path in a Merkle, dag, and now, with this you have learned the basics of ipld. Pathing I will leave you with the exercise of accessing deeper blocks in the graph with the ipld pathing options. We just learned.

A

One handy tool will be the dag Explorer you can find built into the ipfs desktop app to navigate a dag in a neat user interface. So, if I take this CID for the roots, the ID of our website we hit inspect.

A

We could also it'll take us down to this tab and we can see here a neat representation of the dag itself. If we click into any one of these links, say the docs link, for example, it'll show us the exact same thing. We found on our command line uh showing the next subsequent links to this block, and so, if you want to learn the different ways to do this exercise, the answer will be in the write-up of this tutorial, which you can find at curriculum.pla launchpad.io. Thank you, foreign.