# XML file support

**URL:** <https://community.dremio.com/t/xml-file-support/1833>\
**Category:** Uncategorized\
**Created:** [September 13, 2018, 8:31am UTC](https://community.dremio.com/t/xml-file-support/1833 "2018-09-13T08:31:39Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![Ben\_Spencer](https://avatars.discourse-cdn.com/v4/letter/b/e274bd/32.png) [@Ben\_Spencer](https://community.dremio.com/u/Ben_Spencer)\
**Post date:** [September 13, 2018, 8:31am UTC](https://community.dremio.com/t/xml-file-support/1833/1 "2018-09-13T08:31:39Z")

</div>

Hi, Just evaluating Dremio first time today. I am looking to create a dataset out of a drive of XML files where there may be thousands of records per file, like a daily xml log. Is there a simple way that Dremio can parses these XML files? I seem to be missing something. Like I said, first day!

---

<div class="post-metadata">

**Author:** ![can](https://sea2.discourse-cdn.com/flex020/user_avatar/community.dremio.com/can/32/40_2.png) [@can](https://community.dremio.com/u/can)\
**Post date:** [September 13, 2018, 6:22pm UTC](https://community.dremio.com/t/xml-file-support/1833/2 "2018-09-13T18:22:07Z")

</div>

Hey @Ben_Spencer, Dremio does not support parsing XML files. We’d recommend converting them to some other format before analyzing with Dremio.

---

<div class="post-metadata">

**Author:** ![Ben\_Spencer](https://avatars.discourse-cdn.com/v4/letter/b/e274bd/32.png) [@Ben\_Spencer](https://community.dremio.com/u/Ben_Spencer)\
**Post date:** [September 14, 2018, 2:50am UTC](https://community.dremio.com/t/xml-file-support/1833/3 "2018-09-14T02:50:59Z")

</div>

Hey Can, thanks for responding. Would this be on the cards anytime? It seems like this would be just one of a number of parsers/interpretors that could makes sense and Dremio doesn’t cover.

---

<div class="post-metadata">

**Author:** ![david.lee](https://avatars.discourse-cdn.com/v4/letter/d/ed655f/32.png) [@david.lee](https://community.dremio.com/u/david.lee)\
**Post date:** [September 19, 2018, 6:56pm UTC](https://community.dremio.com/t/xml-file-support/1833/4 "2018-09-19T18:56:45Z")

</div>

You can’t really analyze XML in a structured manner without a XSD file…

Is Address a string or a list of strings???

```
<Address>1600 Pennsylvania Ave</Address>
vs
<Address>1600 Pennsylvania Ave</Address>
<Address>10 Downing Street</Address>

```

I wrote something to convert XML to JSON

> **[davlee1972/xml\_to\_json](https://github.com/davlee1972/xml_to_json)**
>
> xml to json parser. Contribute to davlee1972/xml\_to\_json development by creating an account on GitHub.

  

> **[sissaschool/xmlschema](https://github.com/sissaschool/xmlschema)**
>
> XML Schema validator and data conversion library for Python - sissaschool/xmlschema

Even after all this JSON can still be problematic…

```
"Address":"1600 Pennsylvania Ave"
"Zip":"20500"
vs
"Address":"10 Downing Street"
"Zip": null

```

What datatype is Zip? string, int, boolean??

The only nested data structure that really works is parquet which includes a schema definition…  
Someday I’ll write a XML to Parquet converter with the help of XSD files.

---

<div class="post-metadata">

**Author:** ![can](https://sea2.discourse-cdn.com/flex020/user_avatar/community.dremio.com/can/32/40_2.png) [@can](https://community.dremio.com/u/can)\
**Post date:** [September 19, 2018, 7:02pm UTC](https://community.dremio.com/t/xml-file-support/1833/5 "2018-09-19T19:02:57Z")

</div>

@Ben_Spencer probably not a top priority at the moment. We’ll post any updates on this thread.  
@david.lee appreciate the input!