HTMX + Flask: Modern Python Web Apps, Hold the JavaScript Transcripts
Chapter: Exploring the examples on htmx.org
Lecture: Example: Lazy loading

Login or purchase this course to watch this video and the rest of the course contents.
0:00 The next example that we're gonna explore. It's not one that actually appears in our final application, but it's extremely useful.
0:07 Let me lay out the scenario for you. Imagine I've got a page, trying to render it for the user, right? They make a request, we want to show it to him,
0:15 but maybe it's doing some kind of report, or there is some part of the page that's actually really,
0:19 really slow, and I want to get most of the page shown to them and then
0:23 have this other slower section, like a report or a graph or something along those
0:27 lines. I want that to be generated and delivered to the user as soon as
0:32 possible, but without blocking the original page load, and we also want some kind of indicator
0:36 for what's happening. That would fall under the "Lazy Loading" section. Check this out, that's pretty simple.
0:42 So what we do is we say, this section of the page, this whole section, that's one of those slow pieces.
0:48 What I want to do is I want you htmx, to go to the server and call /graph and when it's done, put it right here.
0:56 But while it's loading, I want you to show this indicator with those bars sort
1:01 of spinning around here. And the trigger is going to be that the page has loaded.
1:05 So just the page was shown and then fill out these missing pieces that are slow. In order for this to work,
1:12 you're going to need a little bit of CSS somewhere, included in the page to basically
1:17 have some of the behaviors that are indicated here happening and down here we have our
1:22 graph and this is the part that was slow, but it doesn't take that long, it's like half a second or a second.
1:28 So what we didn't see is this actually loading because by the time I got down here it was done. But if, notice if I load this page, it's instant,
1:35 but there's like a one second processing of something else. If I scroll down, I can refresh here and show that to you.
1:42 So page is loaded. But now it's thinking on the graph request, input came back or the response came back. Boom,
1:49 here's our report and let's go and look at this. So our initial state was that the page was loaded and this section was here.
1:57 Right, this is what we talked about. But then as the document loaded, was complete,
2:03 I said right, now we've got to fill out the slow pieces, and it went and did a request for GET /graph,
2:08 no parameters. The response was, here is your generated image, of your report. How cool is that? So I'm
2:16 sure there are many websites that there's a little part of it or a little
2:19 piece of information that's slow and the rest is really pretty quick, and yet the user
2:24 experiences, I go there and I click, and I wait, and I wait, and I wait and finally the page loads.
2:30 This would be so much better. Show most of the page, and calculate the little final things that you need using one of these lazy loads.


Talk Python's Mastodon Michael Kennedy's Mastodon