Spatial support in Tableau Prep

0:00Hey, it's Tim here and in today's video I'm

0:03telling you about Tableau Prep in the

0:05browser

0:05and the ability to support spatial files.

0:08So this is actually amazing, two features

0:10in one. Not only have they added Tableau

0:12Prep to the browser, check out my previous

0:14video

0:14on that, but they've also brought out this

0:16new feature where you can work with spatial

0:18files in Tableau Prep. Let me show you how

0:21that works. So I've already got a flow here

0:23that I worked on in my previous example,

0:25check out that video in the description

0:27below and

0:27also in the pop-up. If I go ahead and hit

0:30the plus icon here and I select spatial

0:32file,

0:33you'll see that I get this sort of window

0:35to basically connect to a file. Now

0:37remember

0:37I'm working with Tableau Prep in the

0:39browser, so it's actually going to ask me

0:41to upload

0:41a file. So I'm going to go to my 2020

0:44sample file folder, go to spatial and you

0:46'll see

0:47that I've actually got two types of spatial

0:49files here. I've got zipped files and I've

0:51got unzipped files. Now what you've had to

0:53do in the past is unzip the files, but

0:55because

0:56you're uploading this Tableau server needs

0:58everything in one go. So actually you need

1:00to select the zipped version of this where

1:02everything is zipped together. Just to show

1:04you what that looks like, this is the unz

1:06ipped version for the state file that I

1:08actually

1:08downloaded from the US government website.

1:11If I go back and I just grab the single zip

1:14file that contains everything in it. So go

1:16ahead and open that and you can see that it

1:18actually uploads the file and it creates a

1:21spatial file as a data source. Now what I

1:23want to do here is actually join the

1:26geometry data in this file to my superstore

1:28output.

1:29So if I go ahead and click on this clean

1:31five step here, you'll see that I can

1:33actually

1:33see the individual columns that are held

1:36inside of that shape file. And the key

1:37thing I'm

1:38interested in here is the spatial file here

1:40. So if I just, let me just try and grab

1:43this

1:43and move it up and actually just go to the

1:45second tab here. So I just see a Tableau

1:47view.

1:47I can go across here and you can see that I

1:49've got this geometry window here on the

1:51right hand side. I can't see those polygons

1:53, but we'll have a look at that in a second,

1:55but they are there. So what we'll do is we

1:57'll actually just bring this in. Now it

2:00makes

2:00the most sense to connect it at the very

2:02end where I've done all my data prep. So I

2:04'm going

2:04to drag clean five here onto clean four,

2:07click join. And what I want to do is

2:09essentially

2:09join this data. Now you can see it's

2:11created a new branch. That's fine. I can

2:13fix that

2:13later on. And if I actually bring this back

2:16up, you can see that it's got state and it

2:18's

2:18connected it to state ends. That's not

2:20quite what I wanted. I actually need to

2:22connect

2:22it to the name. And I think that's the best

2:25match that we're going to have. So we can

2:27see here that the state and the name are

2:30probably the best match. Let that finish

2:32running here.

2:33And we're going to see that it's not a

2:36perfect match actually. So if I just scroll

2:39down here,

2:40let's see what's going on here. So

2:41Connecticut is there and Connecticut is

2:43there. So what's

2:45going on here is that it looks like my data

2:47set on the left hand side doesn't have

2:50these

2:50seven states. And that's okay because I

2:52sort of trust my superstore data sets, but

2:55the

2:55majority of records are actually there and

2:57everything works. It's just going to be

2:58seven

2:58records that don't quite match. Let's go

3:01ahead and do a left join to that. You can

3:04see here

3:04those seven states just aren't in my

3:06original data set. So all 12,420 rows are

3:09going to

3:10come through. I never actually noticed that

3:12those states were missing. So if you're in

3:14Puerto Rico, United States, Virgin Islands,

3:17Hawaii, Guam, Commonwealth of North America

3:20,

3:20American Samoa, Alaska, have a word with

3:23Tableau. Now let's go ahead and look at

3:25what's going

3:26on here. So we've done the join. Obviously

3:28I want to try and fix this. So let's remove

3:30that and just bring that here. We can

3:32actually add it there and then just sort of

3:34add the

3:34connection like that. Sometimes I always

3:37wonder what's the fastest way of doing

3:38something

3:38in Tableau prep. It's a fairly new tool. So

3:41I'm still always trying to find the best

3:43way

3:43to do it. But I also like to make my flow

3:45nice and tidy and I like to keep the main

3:47part of the flow in the middle and any data

3:49that I'm bringing in on either side. So

3:51that's

3:52pretty much it. That's the flow in a

3:54nutshell. Now because I've already got a

3:55published data

3:56source here called prep in the browser,

3:58when I run this, it's obviously just going

4:00to add

4:00the additional columns into the data set.

4:03So now when I run this, of course, this is

4:05going to publish the changes first so that

4:07this is actually saved. And then once that

4:09's

4:09published, you'll get the little green icon

4:11and you can by all means go ahead and run

4:13the flow. And of course it's queued the

4:16flow and then you're going to be able to

4:17see the

4:18process of that flow. If I go to view flow,

4:20you'll see that it's actually queued. You

4:23can see that it's running in fact as we

4:24speak. If I click on running, it just shows

4:26you what's

4:27going on here, the run history. So you can

4:29see that it's in progress. We're going to

4:31let that finish. And if I just refresh this

4:34page, we should see a completion status.

4:37Yeah,

4:37that's completed in seven seconds. So you

4:39can see the duration wasn't really affected

4:41by the fact that I was using spatial files

4:43because the first time I ran it with no

4:44spatial

4:45files, it was seven seconds, five seconds,

4:47and now seven seconds again. So you're

4:48getting

4:49some really fast runtimes here as well. And

4:51now of course, if I go back to the explore

4:53tab and I go back to my data source, I look

4:56at prep in the browser. It's obviously

4:58going

4:58to give me the ass data interface initially

5:00here, but what I can actually do is see if

5:02those geospatial columns made it into my

5:05data set. So now you can see that I have

5:07this geometry

5:08field. If I actually go and create a new

5:10workbook using this data source, what we

5:12can actually

5:13do is we can double click the geometry

5:16field here, and we are going to get the

5:19spatial

5:19files that we blended just by uploading the

5:23files to a tableau. So isn't that amazing?

5:27We've got spatial data here working native

5:30ly inside of tableau. This is amazing. We

5:33did

5:34this all in the browser, and now we're

5:35building the workbook in the browser and we

5:37're able

5:38to work with it. So that's some of the

5:39power of what's happening in the browser. I

5:41think

5:41pretty much next year, everything will be

5:43in the browser. I'm not sure that's a great

5:45thing because it might break some people's

5:47workflows. Don't forget that spatial file

5:49is on your desktop and it now no longer has

5:51a link to a file that could have been in a

5:53SharePoint folder that you might have been

5:55updating from time to time. So that's just

5:57something to be aware of, but otherwise, I

5:59think this is a great feature. Let me know

6:00what you think in the comments below. If

6:02you found this video useful, you know what

6:03to

6:03do, hit the plus button. If you don't like

6:05this video, then by all means hit the

6:07dislike

6:08button twice and I'll catch you in the next

6:09video.

6:10[BLANK_AUDIO]