glom

Restructuring data, the Python way

Real applications have real data, and real data nests. Objects inside of objects inside of lists of objects.

glom is a new and powerful way to handle real-world data, featuring:

Path-based access for nested data structures
Readable, meaningful error messages
Declarative data transformation, using lightweight, Pythonic specifications
Built-in data exploration and debugging features

All of that and more, available as a fully-documented, pure-Python package, tested on Python 3.7-3.14, as well as PyPy3. Installation is as easy as:

  pip install glom

And when you install glom, you also get the glom command-line interface, letting you experiment at the console, but never limiting you to shell scripts:

Usage: glom [FLAGS] [spec [target]]

Command-line interface to the glom library, providing nested data access and data
restructuring with the power of Python.

Flags:

  --help / -h                     show this help message and exit
  --target-file TARGET_FILE       path to target data source (optional)
  --target-format TARGET_FORMAT
                                  format of the source data (json, python, toml,
                                  or yaml) (defaults to 'json')
  --spec-file SPEC_FILE           path to glom spec definition (optional)
  --spec-format SPEC_FORMAT       format of the glom spec definition (json, python,
                                    python-full) (defaults to 'python')
  --indent INDENT                 number of spaces to indent the result, 0 to disable
                                    pretty-printing (defaults to 2)
  --debug                         interactively debug any errors that come up
  --inspect                       interactively explore the data

Anything you can do at the command line readily translates to Python code, so you've always got a path forward when complexity starts to ramp up.

Examples

Without glom

>>> data = {'a': {'b': {'c': 'd'}}}
>>> data['a']['b']['c']
'd'
>>> data2 = {'a': {'b': None}}
>>> data2['a']['b']['c']
Traceback (most recent call last):
...
TypeError: 'NoneType' object is not subscriptable

With glom

>>> glom(data, 'a.b.c')
'd'
>>> glom(data2, 'a.b.c')
Traceback (most recent call last):
...
PathAccessError: could not access 'c', index 2 in path Path('a', 'b', 'c'), got error: ...

Learn more

If all this seems interesting, continue exploring glom below:

All of the links above are overflowing with examples, but should you find anything about the docs, or glom itself, lacking, please submit an issue!

In the meantime, just remember: When you've got nested data, glom it! ☄️

Metadata

Release files for glom 25.12.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for glom 25.12.0
File	Size	Uploaded
glom-25.12.0.tar.gz	196.1 kB	Dec 29, 2025	Details

Built distribution (wheel)

Table of built distributions (wheels) for glom 25.12.0
File	Interpreter	ABI	Platform	Reset
glom-25.12.0-py3-none-any.whl 103.3 kB Dec 29, 2025	Python 3	none	any	Details

Total release size: 299.4 kB

Release files / glom-25.12.0.tar.gz

Download URL	glom-25.12.0.tar.gz
Size	196.1 kB
Tags	Source
SHA-256 checksum How to use checksums	`1ae7da88be3693df40ad27bdf57a765a55c075c86c971bcddd67927403eb0069`
BLAKE2b-256 checksum How to use checksums	`78748387f95565ba7c30cd152a585b275ebb9a834d1d32782425c5d2fe0a102c`
Upload date	Dec 29, 2025
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/5.1.1 CPython/3.10.12`

Release files / glom-25.12.0-py3-none-any.whl

Download URL	glom-25.12.0-py3-none-any.whl
Size	103.3 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`b9f21e77f71a6576a43864e85066b8cc3f0f778d0d50961563f8981377a6dcb1`
BLAKE2b-256 checksum How to use checksums	`a7e64129d9a3baa72d747533bb33376543ccadd9a7f9944e5a6e3ae2e245f5d6`
Upload date	Dec 29, 2025
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/5.1.1 CPython/3.10.12`