# Hail index file compression type

**URL:** <https://discuss.hail.is/t/hail-index-file-compression-type/4196>\
**Category:** Development\
**Created:** [August 16, 2025, 11:21pm UTC](https://discuss.hail.is/t/hail-index-file-compression-type/4196 "2025-08-16T23:21:02Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![PHILIPP111007](https://yyz2.discourse-cdn.com/flex036/user_avatar/discuss.hail.is/philipp111007/32/1314_2.png) [@PHILIPP111007](https://discuss.hail.is/u/PHILIPP111007)\
**Post date:** [August 16, 2025, 11:21pm UTC](https://discuss.hail.is/t/hail-index-file-compression-type/4196/1 "2025-08-16T23:21:02Z")

</div>

Hello, what type of compression does the index file have? (matrix\_table/index/part-0/index)

Sincerely, Roshchin Philipp

---

<div class="post-metadata">

**Author:** ![PHILIPP111007](https://yyz2.discourse-cdn.com/flex036/user_avatar/discuss.hail.is/philipp111007/32/1314_2.png) [@PHILIPP111007](https://discuss.hail.is/u/PHILIPP111007)\
**Post date:** [August 17, 2025, 8:08pm UTC](https://discuss.hail.is/t/hail-index-file-compression-type/4196/2 "2025-08-17T20:08:57Z")

</div>

i need to parse data in index file by Python

---

<div class="post-metadata">

**Author:** ![PHILIPP111007](https://yyz2.discourse-cdn.com/flex036/user_avatar/discuss.hail.is/philipp111007/32/1314_2.png) [@PHILIPP111007](https://discuss.hail.is/u/PHILIPP111007)\
**Post date:** [August 19, 2025, 8:52am UTC](https://discuss.hail.is/t/hail-index-file-compression-type/4196/3 "2025-08-19T08:52:13Z")

</div>

without Pandas and Spark, only Python

Thank you!

---

<div class="post-metadata">

**Author:** ![PHILIPP111007](https://yyz2.discourse-cdn.com/flex036/user_avatar/discuss.hail.is/philipp111007/32/1314_2.png) [@PHILIPP111007](https://discuss.hail.is/u/PHILIPP111007)\
**Post date:** [August 19, 2025, 9:26am UTC](https://discuss.hail.is/t/hail-index-file-compression-type/4196/4 "2025-08-19T09:26:16Z")

</div>

```python
import chardet

part_path = "/data/table.ht/rows/parts/part-00-beab4b89-aff6-43de-b358-feed3f1c07a3"

with open(part_path, "rb") as file:
    content = file.read()

detected_encoding = chardet.detect(content)['encoding']
detected_encoding
# 'ISO-8859-1'

decoded_content = content.decode("ISO-8859-1")
decoded_content
# '\x0e\x00\x00\x00\x01\x00\x00\x00(µ/ý \x01\t\x00\x00\x00\r\x00\x00\x00\x00\x00\x00\x00(µ/ý \x00\x01\x00\x00'

```

I need to understand how to decode this data

---

<div class="post-metadata">

**Author:** ![patrick-schultz](https://yyz2.discourse-cdn.com/flex036/user_avatar/discuss.hail.is/patrick-schultz/32/265_2.png) [@patrick-schultz](https://discuss.hail.is/u/patrick-schultz)\
**Post date:** [October 29, 2025, 7:27pm UTC](https://discuss.hail.is/t/hail-index-file-compression-type/4196/5 "2025-10-29T19:27:58Z")

</div>

Hi @PHILIPP111007,

The index file is a binary file, and is part of our internal format which we don’t document, and can’t promise won’t change in the future. Can I ask why you need to decode it in python?