1
00:00:02,320 --> 00:00:04,770
Now let's do a different query,

2
00:00:04,840 --> 00:00:12,650
we still have that text index on the description and let's search for awesome T-shirt. If we do this,

3
00:00:12,650 --> 00:00:16,580
I find both entries because I have awesome here too in the first product.

4
00:00:16,670 --> 00:00:22,510
Now finding both might be fine but you typically would want to order this differently,

5
00:00:22,550 --> 00:00:27,980
for us humans it's pretty clear that the second product is the better hit because there we have both awesome

6
00:00:28,070 --> 00:00:34,160
and t-shirt so would be nice if that would come first in our result list and we can actually get there

7
00:00:34,820 --> 00:00:40,400
because mongodb does something interesting or special when managing such a text index or when searching

8
00:00:40,400 --> 00:00:47,900
for text in a text index, we can find out how it scores its results and let's first of all find out with

9
00:00:47,900 --> 00:00:49,120
the help of projection.

10
00:00:49,220 --> 00:00:56,420
So here I'll add a second argument to my find method to project the results and I will simply output

11
00:00:56,510 --> 00:00:58,110
a score field here

12
00:00:59,430 --> 00:01:08,220
where I can use a special meta operator to add text score. This is a meta field added or managed by mongodb

13
00:01:08,270 --> 00:01:09,920
for text searches

14
00:01:09,930 --> 00:01:11,090
here with the $text

15
00:01:11,090 --> 00:01:14,470
operator on a text index.

16
00:01:14,470 --> 00:01:16,240
If I do this, you see the score.

17
00:01:16,410 --> 00:01:18,340
mongodb assigned to a result

18
00:01:18,580 --> 00:01:22,550
and here, it even automatically sort it by that score now as you can tell.

19
00:01:22,840 --> 00:01:30,680
So here we see that the score is much higher for T-shirt and awesome than for just awesome and it automatically

20
00:01:30,700 --> 00:01:31,620
sort it for us here

21
00:01:31,630 --> 00:01:39,430
but if we want to make sure, we can of course also add sort here and we can then sort by score where

22
00:01:39,430 --> 00:01:48,020
score is a document with $meta and then textScore written like this as a text and I should add

23
00:01:48,020 --> 00:01:51,640
a dot in front of pretty. And now, it's definitely sorted,

24
00:01:51,650 --> 00:01:58,220
it was before but now we enforce sorting by looking at that meta text score which mongodb manages

25
00:01:58,220 --> 00:01:58,980
for us.

26
00:01:59,210 --> 00:02:03,630
So this is simply a construct you can remember for when working with text and text indexes,

27
00:02:03,800 --> 00:02:10,400
you can extract this meta information text score and both these words have to be written like this from

28
00:02:10,490 --> 00:02:15,760
mongodb so to say, it manages that for us and we can see how it scores the results

29
00:02:15,770 --> 00:02:18,890
and as you can see, we can then use that to order too.