# Dremio capabilities on UNStructured data?

**URL:** <https://community.dremio.com/t/dremio-capabilities-on-unstructured-data/4414>\
**Category:** Uncategorized\
**Created:** [October 24, 2019, 3:29pm UTC](https://community.dremio.com/t/dremio-capabilities-on-unstructured-data/4414 "2019-10-24T15:29:59Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![KThig](https://avatars.discourse-cdn.com/v4/letter/k/eb8c5e/32.png) [@KThig](https://community.dremio.com/u/KThig)\
**Post date:** [October 24, 2019, 3:29pm UTC](https://community.dremio.com/t/dremio-capabilities-on-unstructured-data/4414/1 "2019-10-24T15:29:59Z")

</div>

I am evaluating Dremio for one of my client partner’s use case and wondering if Dremio has any capabilities for unstructured data? Any supporting documentation can also be a help

---

<div class="post-metadata">

**Author:** ![balaji.ramaswamy](https://sea2.discourse-cdn.com/flex020/user_avatar/community.dremio.com/balaji.ramaswamy/32/843_2.png) [@balaji.ramaswamy](https://community.dremio.com/u/balaji.ramaswamy)\
**Post date:** [October 24, 2019, 4:53pm UTC](https://community.dremio.com/t/dremio-capabilities-on-unstructured-data/4414/2 "2019-10-24T16:53:38Z")

</div>

@KThig

Any specific sources you have in mind like Elasticsearch, Mongo, JSON etc?

---

<div class="post-metadata">

**Author:** ![ssky](https://avatars.discourse-cdn.com/v4/letter/s/9fc348/32.png) [@ssky](https://community.dremio.com/u/ssky)\
**Post date:** [November 6, 2019, 1:32pm UTC](https://community.dremio.com/t/dremio-capabilities-on-unstructured-data/4414/3 "2019-11-06T13:32:43Z")

</div>

Hi,

I am having the same need about the json files

Is dremio have the capabilities to parse json files dynamically and recursively ?

is there predefined functions to help parse a json file and store the result in a relational schema ?

---

<div class="post-metadata">

**Author:** ![balaji.ramaswamy](https://sea2.discourse-cdn.com/flex020/user_avatar/community.dremio.com/balaji.ramaswamy/32/843_2.png) [@balaji.ramaswamy](https://community.dremio.com/u/balaji.ramaswamy)\
**Post date:** [November 6, 2019, 2:32pm UTC](https://community.dremio.com/t/dremio-capabilities-on-unstructured-data/4414/4 "2019-11-06T14:32:27Z")

</div>

@ssky

You should be able to promote the JSON folder as a PDS and then Dremio sees it like a relational schema. On top of that if you create reflections, it would Dremio generated Parquert files

Thanks  
@balaji.ramaswamy

---

<div class="post-metadata">

**Author:** ![ssky](https://avatars.discourse-cdn.com/v4/letter/s/9fc348/32.png) [@ssky](https://community.dremio.com/u/ssky)\
**Post date:** [November 13, 2019, 9:15am UTC](https://community.dremio.com/t/dremio-capabilities-on-unstructured-data/4414/5 "2019-11-13T09:15:13Z")

</div>

Hi balaji,

Dremio sees the Json as physical dataset, that’s fine

but let me explain the problem with a sample used in a dremio tutorial

The relation schema expected is at least tree relational tables  
table 1 : business  
table 2 : business type  
table 3 : service hours

So , if I want to get this schema, I have to parse the text of each record in Dremio  
to get all service hours record and store them in a new table  
and things can get tougher if there’s multiple imbrication levels !

So my question : is dremio capable to do this transformation easily ?

{  
“business\_id”: “vcNAWiLM4dR7D2nwwJ7nCA”,  
“full\_address”: “4840 E Indian School Rd\nSte 101\nPhoenix, AZ 85018”,  
“hours”: {  
“Tuesday”: {  
“close”: “17:00”,  
“open”: “08:00”  
},  
“Friday”: {  
“close”: “17:00”,  
“open”: “08:00”  
},  
“Monday”: {  
“close”: “17:00”,  
“open”: “08:00”  
},  
“Wednesday”: {  
“close”: “17:00”,  
“open”: “08:00”  
},  
“Thursday”: {  
“close”: “17:00”,  
“open”: “08:00”  
}  
},  
“open”: true,  
“categories”: [  
“Doctors”,  
“Health & Medical”  
],  
“city”: “Phoenix”,  
“review\_count”: 7,  
“name”: “Eric Goldberg, MD”,  
“neighborhoods”: [],  
“longitude”: -111.983758,  
“state”: “AZ”,  
“stars”: 3.5,  
“latitude”: 33.499313,  
“attributes”: {  
“By Appointment Only”: true  
},  
“type”: “business”  
}