Erotemic/ubelt

Ubelt is a utility library for Python with a stdlib like feel.

Elevator Pitch:

Is the Python standard library good? Yes. Could it's conciseness be improved? Yes. Ubelt aims to provide a quicker way to express things you can do in the standard library. Progress? ub.ProgIter. Hashing? ub.hash_data / ub.hash_file. Caching? ub.Cacher / ub.CacheStamp. Shell commands? ub.cmd. There are similar functions for downloading data, futures-based parallel (or serial) job execution, pretty reprs, path management, iteration, and one of my favorites: set operation enriched dictionaries: ub.udict.

There are 120ish functions and classes to help make your code shorter and easier to express concisely. The library is fast to install and import, all dependencies are optional. As of 2023 it is 6 years old, regularly maintained, and mature. It is well tested and has moderate usage.

To learn more, the function usefulness chart is a good place to start. This shows how often I use particular functions, and while some of the less used ones are candidates for removal, some of them still worth checking out. For a slightly slower start, read the introduction:

Introduction:

Ubelt is a lightweight library of robust, tested, documented, and simple functions that extend the Python standard library. It has a flat API that all behaves similarly on Windows, Mac, and Linux (up to some small unavoidable differences). Almost every function in ubelt was written with a doctest. This provides helpful documentation and example usage as well as helping achieve 100% test coverage (with minor exceptions on Windows).

• Goal: provide simple functions that accomplish common tasks not yet addressed by the python standard library.
• Constraints: Must be low-impact pure python; it should be easy to install and use.
• Method: All functions are written with docstrings and doctests to ensure that a baseline level of documentation and testing always exists (even if functions are copy/pasted into other libraries)
• Motto: Good utilities lift all codes.

Read the docs here: http://ubelt.readthedocs.io/en/latest/

These are some of the tasks that ubelt's API enables:

• extended pathlib with expand, ensuredir, endswith, augment, delete (ub.Path)
• get paths to cross platform data/cache/config directories (ub.Path.appdir, ...)
• perform set operations on dictionaries (SetDict)
• a dictionary with extended helper methods like subdict, take, peek_value, invert, sorted_keys, sorted_vals (UDict)
• hash common data structures like list, dict, int, str, etc. (hash_data)
• hash files (hash_file)
• cache a block of code (Cacher, CacheStamp)
• time a block of code (Timer)
• show loop progress with less overhead than tqdm (ProgIter)
• download a file with optional caching and hash verification (download, grabdata)
• run shell commands (cmd)
• find a file or directory in candidate locations (find_path, find_exe)
• string-repr for nested data structures (urepr)
• color text with ANSI tags (color_text)
• horizontally concatenate multiline strings (hzcat)
• create cross platform symlinks (symlink)
• import a module using the path to that module (import_module_from_path)
• check if a particular flag or value is on the command line (argflag, argval)
• memoize functions (memoize, memoize_method, memoize_property)
• build ordered sets (oset)
• argmax/min/sort on lists and dictionaries (argmin, argsort,)
• get a histogram of items or find duplicates in a list (dict_hist, find_duplicates)
• group a sequence of items by some criterion (group_items)

Ubelt is small. Its top-level API is defined using roughly 40 lines:

from ubelt.util_arg import (argflag, argval,)
from ubelt.util_cache import (CacheStamp, Cacher,)
from ubelt.util_colors import (NO_COLOR, color_text, highlight_code,)
from ubelt.util_const import (NoParam,)
from ubelt.util_cmd import (cmd,)
from ubelt.util_dict import (AutoDict, AutoOrderedDict, SetDict, UDict, ddict,
                             dict_diff, dict_hist, dict_isect, dict_subset,
                             dict_union, dzip, find_duplicates, group_items,
                             invert_dict, map_keys, map_vals, map_values,
                             named_product, odict, sdict, sorted_keys,
                             sorted_vals, sorted_values, udict, varied_values,)
from ubelt.util_deprecate import (schedule_deprecation,)
from ubelt.util_download import (download, grabdata,)
from ubelt.util_download_manager import (DownloadManager,)
from ubelt.util_func import (compatible, identity, inject_method,)
from ubelt.util_repr import (ReprExtensions, urepr,)
from ubelt.util_futures import (Executor, JobPool,)
from ubelt.util_io import (delete, touch,)
from ubelt.util_links import (symlink,)
from ubelt.util_list import (allsame, argmax, argmin, argsort, argunique,
                             boolmask, chunks, compress, flatten, iter_window,
                             iterable, peek, take, unique, unique_flags,)
from ubelt.util_hash import (hash_data, hash_file,)
from ubelt.util_import import (import_module_from_name,
                               import_module_from_path, modname_to_modpath,
                               modpath_to_modname, split_modpath,)
from ubelt.util_indexable import (IndexableWalker, indexable_allclose,)
from ubelt.util_memoize import (memoize, memoize_method, memoize_property,)
from ubelt.util_mixins import (NiceRepr,)
from ubelt.util_path import (Path, TempDir, augpath, ensuredir, expandpath,
                             shrinkuser, userhome,)
from ubelt.util_platform import (DARWIN, LINUX, POSIX, WIN32, find_exe,
                                 find_path, platform_cache_dir,
                                 platform_config_dir, platform_data_dir,)
from ubelt.util_str import (codeblock, hzcat, indent, paragraph,)
from ubelt.util_stream import (CaptureStdout, CaptureStream, TeeStringIO,)
from ubelt.util_time import (Timer, timeparse, timestamp,)
from ubelt.util_zip import (split_archive, zopen,)
from ubelt.orderedset import (OrderedSet, oset,)
from ubelt.progiter import (ProgIter,)

Installation:

Ubelt is distributed on pypi as a universal wheel and can be pip installed on Python 3.6+. Installations are tested on CPython and PyPy implementations. For Python 2.7 and 3.5, the last supported version was 0.11.1.

pip install ubelt

Note that our distributions on pypi are signed with GPG. The signing public key is D297D757; this should agree with the value in dev/public_gpg_key.

Function Usefulness

When I had to hand pick a set of functions that I thought were the most useful I chose these and provided some comment on why:

import ubelt as ub

ub.Path  # inherits from pathlib.Path with quality of life improvements
ub.UDict  # inherits from dict with keywise set operations and quality of life improvements
ub.Cacher  # configuration based on-disk cachine
ub.CacheStamp  # indirect caching with corruption detection
ub.hash_data  # hash mutable python containers, useful with Cacher to config strings
ub.cmd  # combines the best of subprocess.Popen and os.system
ub.download  # download a file with a single command. Also see grabdata for the same thing, but caching from CacheStamp.
ub.JobPool   # easy multi-threading / multi-procesing / or single-threaded processing
ub.ProgIter  # a minimal progress iterator. It's single threaded, informative, and faster than tqdm.
ub.memoize  # like ``functools.cache``, but uses ub.hash_data if the args are not hashable.
ub.urepr  # readable representations of nested data structures

But a better way might to objectively measure the frequency of usage and built a histogram of usefulness. I generated this histogram using python dev/maintain/gen_api_for_docs.py, which roughly counts the number of times I've used a ubelt function in another project. Note: this measure is biased towards older functions.

Examples

The most up to date examples are the doctests. We also have a Jupyter notebook: https://github.com/Erotemic/ubelt/blob/main/docs/notebooks/Ubelt%20Demo.ipynb

Here are some examples of some features inside ubelt

Paths

Ubelt extends pathlib.Path by adding several new (often chainable) methods. Namely, augment, delete, expand, ensuredir, shrinkuser. It also modifies behavior of touch to be chainable. (New in 1.0.0)

>>> # Ubelt extends pathlib functionality
>>> import ubelt as ub
>>> dpath = ub.Path('~/.cache/ubelt/demo_path').expand().ensuredir()
>>> fpath = dpath / 'text_file.txt'
>>> aug_fpath = fpath.augment(suffix='.aux', ext='.jpg').touch()
>>> aug_dpath = dpath.augment('demo_path2')
>>> assert aug_fpath.read_text() == ''
>>> fpath.write_text('text data')
>>> assert aug_fpath.exists()
>>> assert not aug_fpath.delete().exists()
>>> assert dpath.exists()
>>> assert not dpath.delete().exists()
>>> print(f'{fpath.shrinkuser()}')
>>> print(f'{dpath.shrinkuser()}')
>>> print(f'{aug_fpath.shrinkuser()}')
>>> print(f'{aug_dpath.shrinkuser()}')
~/.cache/ubelt/demo_path/text_file.txt
~/.cache/ubelt/demo_path
~/.cache/ubelt/demo_path/text_file.aux.jpg
~/.cache/ubelt/demo_pathdemo_path2

Hashing

The ub.hash_data constructs a hash for common Python nested data structures. Extensions to allow it to hash custom types can be registered. By default it handles lists, dicts, sets, slices, uuids, and numpy arrays.

>>> import ubelt as ub
>>> data = [('arg1', 5), ('lr', .01), ('augmenters', ['flip', 'translate'])]
>>> ub.hash_data(data, hasher='sha256')
0d95771ff684756d7be7895b5594b8f8484adecef03b46002f97ebeb1155fb15

Support for torch tensors and pandas data frames are also included, but needs to be explicitly enabled. There also exists an non-public plugin architecture to extend this function to arbitrary types. While not officially supported, it is usable and will become better integrated in the future. See ubelt/util_hash.py for details.

Caching

Cache intermediate results from blocks of code inside a script with minimal boilerplate or modification to the original code.

For direct caching of data, use the Cacher class. By default results will be written to the ubelt's appdir cache, but the exact location can be specified via dpath or the appname arguments. Additionally, process dependencies can be specified via the depends argument, which allows for implicit cache invalidation. As far as I can tell, this is the most concise way (4 lines of boilerplate) to cache a block of code with existing Python syntax (as of 2022-06-03).

>>> import ubelt as ub
>>> depends = ['config', {'of': 'params'}, 'that-uniquely-determine-the-process']
>>> cacher = ub.Cacher('test_process', depends=depends, appname='myapp')
>>> # start fresh
>>> cacher.clear()
>>> for _ in range(2):
>>>     data = cacher.tryload()
>>>     if data is None:
>>>         myvar1 = 'result of expensive process'
>>>         myvar2 = 'another result'
>>>         data = myvar1, myvar2
>>>         cacher.save(data)
>>> myvar1, myvar2 = data

For indirect caching, use the CacheStamp class. This simply writes a "stamp" file that marks that a process has completed. Additionally you can specify criteria for when the stamp should expire. If you let CacheStamp know about the expected "product", it will expire the stamp if that file has changed, which can be useful in situations where caches might becomes corrupt or need invalidation.

>>> import ubelt as ub
>>> dpath = ub.Path.appdir('ubelt/demo/cache').delete().ensuredir()
>>> params = {'params1': 1, 'param2': 2}
>>> expected_fpath = dpath / 'file.txt'
>>> stamp = ub.CacheStamp('name', dpath=dpath, depends=params,
>>>                      hasher='sha256', product=expected_fpath,
>>>                      expires='2101-01-01T000000Z', verbose=3)
>>> # Start fresh
>>> stamp.clear()
>>>
>>> for _ in range(2):
>>>     if stamp.expired():
>>>         expected_fpath.write_text('expensive process')
>>>         stamp.renew()

See https://ubelt.readthedocs.io/en/latest/ubelt.util_cache.html for more details about Cacher and CacheStamp.

Loop Progress

ProgIter is a no-threads attached Progress meter that writes to stdout. It is a mostly drop-in alternative to tqdm. The advantage of ``ProgIter`` is that it does not use any python threading, and therefore can be safer with code that makes heavy use of multiprocessing.

Note: ProgIter is also defined in a standalone module: pip install progiter)

>>> import ubelt as ub
>>> def is_prime(n):
...     return n >= 2 and not any(n % i == 0 for i in range(2, n))
>>> for n in ub.ProgIter(range(1000), verbose=2):
>>>     # do some work
>>>     is_prime(n)
    0/1000... rate=0.00 Hz, eta=?, total=0:00:00, wall=14:05 EST
    1/1000... rate=82241.25 Hz, eta=0:00:00, total=0:00:00, wall=14:05 EST
  257/1000... rate=177204.69 Hz, eta=0:00:00, total=0:00:00, wall=14:05 EST
  642/1000... rate=94099.22 Hz, eta=0:00:00, total=0:00:00, wall=14:05 EST
 1000/1000... rate=71886.74 Hz, eta=0:00:00, total=0:00:00, wall=14:05 EST

Command Line Interaction

The builtin Python subprocess.Popen module is great, but it can be a bit clunky at times. The os.system command is easy to use, but it doesn't have much flexibility. The ub.cmd function aims to fix this. It is as simple to run as os.system, but it returns a dictionary containing the return code, standard out, standard error, and the Popen object used under the hood.

This utility is designed to provide as consistent as possible behavior across different platforms. We aim to support Windows, Linux, and OSX.

>>> import ubelt as ub
>>> info = ub.cmd('gcc --version')
>>> print(ub.urepr(info))
{
    'command': 'gcc --version',
    'err': '',
    'out': 'gcc (Ubuntu 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609\nCopyright (C) 2015 Free Software Foundation, Inc.\nThis is free software; see the source for copying conditions.  There is NO\nwarranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.\n\n',
    'proc': <subprocess.Popen object at 0x7ff98b310390>,
    'ret': 0,
}

Also note the use of ub.urepr (previously ub.repr2) to nicely format the output dictionary.

Additionally, if you specify verbose=True, ub.cmd will simultaneously capture the standard output and display it in real time (i.e. it will "tee" the output).

>>> import ubelt as ub
>>> info = ub.cmd('gcc --version', verbose=True)
gcc (Ubuntu 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
Copyright (C) 2015 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

A common use case for ub.cmd is parsing version numbers of programs

>>> import ubelt as ub
>>> cmake_version = ub.cmd('cmake --version')['out'].splitlines()[0].split()[-1]
>>> print('cmake_version = {!r}'.format(cmake_version))
cmake_version = 3.11.0-rc2

This allows you to easily run a command line executable as part of a python process, see what it is doing, and then do something based on its output, just as you would if you were interacting with the command line itself.

The idea is that ub.cmd removes the need to think about if you need to pass a list of args, or a string. Both will work.

New in 1.0.0, a third variant with different consequences for executing shell commands. Using the system=True kwarg will directly use os.system instead of Popen entirely. In this mode it is not possible to tee the output because the program is executing directly in the foreground. This is useful for doing things like spawning a vim session and returning if the user manages to quit vim.

Downloading Files

The function ub.download provides a simple interface to download a URL and save its data to a file.

>>> import ubelt as ub
>>> url = 'http://i.imgur.com/rqwaDag.png'
>>> fpath = ub.download(url, verbose=0)
>>> print(ub.shrinkuser(fpath))
~/.cache/ubelt/rqwaDag.png

The function ub.grabdata works similarly to ub.download, but whereas ub.download will always re-download the file, ub.grabdata will check if the file exists and only re-download it if it needs to.

>>> import ubelt as ub
>>> url = 'http://i.imgur.com/rqwaDag.png'
>>> fpath = ub.grabdata(url, verbose=0, hash_prefix='944389a39')
>>> print(ub.shrinkuser(fpath))
~/.cache/ubelt/rqwaDag.png

New in version 0.4.0: both functions now accepts the hash_prefix keyword argument, which if specified will check that the hash of the file matches the provided value. The hasher keyword argument can be used to change which hashing algorithm is used (it defaults to "sha512").

Dictionary Set Operations

Dictionary operations that are analogous to set operations. See each funtions documentation for more details on the behavior of the values. Typically the last seen value is given priority.

I hope Python decides to add these to the stdlib someday.

• ubelt.dict_union corresponds to set.union.
• ubelt.dict_isect corresponds to set.intersection.
• ubelt.dict_diff corresponds to set.difference.

>>> d1 = {'a': 1, 'b': 2, 'c': 3}
>>> d2 = {'c': 10, 'e': 20, 'f': 30}
>>> d3 = {'e': 10, 'f': 20, 'g': 30, 'a': 40}
>>> ub.dict_union(d1, d2, d3)
{'a': 40, 'b': 2, 'c': 10, 'e': 10, 'f': 20, 'g': 30}

>>> ub.dict_isect(d1, d2)
{'c': 3}

>>> ub.dict_diff(d1, d2)
{'a': 1, 'b': 2}

New in Version 1.2.0: Ubelt now contains a dictionary subclass with set operations that can be invoked as ubelt.SetDict or ub.sdict. Note that n-ary operations are supported.

>>> d1 = ub.sdict({'a': 1, 'b': 2, 'c': 3})
>>> d2 = {'c': 10, 'e': 20, 'f': 30}
>>> d3 = {'e': 10, 'f': 20, 'g': 30, 'a': 40}
>>> d1 | d2 | d3
{'a': 40, 'b': 2, 'c': 10, 'e': 10, 'f': 20, 'g': 30}

>>> d1 & d2
{'c': 3}

>>> d1 - d2
{'a': 1, 'b': 2}

>>> ub.sdict.intersection({'a': 1, 'b': 2, 'c': 3}, ['b', 'c'], ['c', 'e'])
{'c': 3}

Note this functionality and more is available in ubelt.UDict or ub.udict.

Grouping Items

Given a list of items and corresponding ids, create a dictionary mapping each id to a list of its corresponding items. In other words, group a sequence of items of type VT and corresponding keys of type KT given by a function or corresponding list, group them into a Dict[KT, List[VT] such that each key maps to a list of the values associated with the key. This is similar to pandas.DataFrame.groupby.

Group ids can be specified by a second list containing the id for each corresponding item.

>>> import ubelt as ub
>>> # Group via a corresonding list
>>> item_list    = ['ham',     'jam',   'spam',     'eggs',    'cheese', 'bannana']
>>> groupid_list = ['protein', 'fruit', 'protein',  'protein', 'dairy',  'fruit']
>>> dict(ub.group_items(item_list, groupid_list))
{'dairy': ['cheese'], 'fruit': ['jam', 'bannana'], 'protein': ['ham', 'spam', 'eggs']}

They can also be given by a function that is executed on each item in the list

>>> import ubelt as ub
>>> # Group via a function
>>> item_list    = ['ham',     'jam',   'spam',     'eggs',    'cheese', 'bannana']
>>> def grouper(item):
...     return item.count('a')
>>> dict(ub.group_items(item_list, grouper))
{1: ['ham', 'jam', 'spam'], 0: ['eggs', 'cheese'], 3: ['bannana']}

Dictionary Histogram

Find the frequency of items in a sequence. Given a list or sequence of items, this returns a dictionary mapping each unique value in the sequence to the number of times it appeared. This is similar to pandas.DataFrame.value_counts.

>>> import ubelt as ub
>>> item_list = [1, 2, 39, 900, 1232, 900, 1232, 2, 2, 2, 900]
>>> ub.dict_hist(item_list)
{1232: 2, 1: 1, 2: 4, 900: 3, 39: 1}

Each item can also be given a weight

>>> import ubelt as ub
>>> item_list = [1, 2, 39, 900, 1232, 900, 1232, 2, 2, 2, 900]
>>> weights   = [1, 1,  0,   0,    0,   0,  0.5, 0, 1, 1, 0.3]
>>> ub.dict_hist(item_list, weights=weights)
{1: 1, 2: 3, 39: 0, 900: 0.3, 1232: 0.5}

Dictionary Manipulation

Map functions across dictionarys to transform the keys or values in a dictionary. The ubelt.map_keys function applies a function to each key in a dictionary and returns this transformed copy of the dictionary. Key conflict behavior currently raises and error, but may be configurable in the future. The ubelt.map_vals function is the same except the function is applied to each value instead. I these functions are useful enough to be ported to Python itself.

>>> import ubelt as ub
>>> dict_ = {'a': [1, 2, 3], 'bb': [], 'ccc': [2,]}
>>> dict_keymod = ub.map_keys(len, dict_)
>>> dict_valmod = ub.map_vals(len, dict_)
>>> print(dict_keymod)
>>> print(dict_valmod)
{1: [1, 2, 3], 2: [], 3: [2]}
{'a': 3, 'bb': 0, 'ccc': 1}

Take a subset of a dictionary. Note this is similar to ub.dict_isect, except this will raise an error if the given keys are not in the dictionary.

>>> import ubelt as ub
>>> dict_ = {'K': 3, 'dcvs_clip_max': 0.2, 'p': 0.1}
>>> subdict_ = ub.dict_subset(dict_, ['K', 'dcvs_clip_max'])
>>> print(subdict_)
{'K': 3, 'dcvs_clip_max': 0.2}

The ubelt.take function works on dictionarys (and lists). It is similar to ubelt.dict_subset, except that it returns just a list of the values, and discards information about the keys. It is also possible to specify a default value.

>>> import ubelt as ub
>>> dict_ = {1: 'a', 2: 'b', 3: 'c'}
>>> print(list(ub.take(dict_, [1, 3, 4, 5], default=None)))
['a', 'c', None, None]

Invert the mapping defined by a dictionary. By default invert_dict assumes that all dictionary values are distinct (i.e. the mapping is one-to-one / injective).

>>> import ubelt as ub
>>> mapping = {0: 'a', 1: 'b', 2: 'c', 3: 'd'}
>>> ub.invert_dict(mapping)
{'a': 0, 'b': 1, 'c': 2, 'd': 3}

However, by specifying unique_vals=False the inverted dictionary builds a set of keys that were associated with each value.

>>> import ubelt as ub
>>> mapping = {'a': 0, 'A': 0, 'b': 1, 'c': 2, 'C': 2, 'd': 3}
>>> ub.invert_dict(mapping, unique_vals=False)
{0: {'A', 'a'}, 1: {'b'}, 2: {'C', 'c'}, 3: {'d'}}

New in Version 1.2.0: Ubelt now contains a dictionary subclass ubelt.UDict with these quality of life operations (and also inherits from ubelt.SetDict). The alias ubelt.udict can be used for quicker access.

>>> import ubelt as ub
>>> d1 = ub.udict({'a': 1, 'b': 2, 'c': 3})
>>> d1 & {'a', 'c'}
{'a': 1, 'c': 3}

>>> d1.map_keys(ord)
{97: 1, 98: 2, 99: 3}
>>> d1.invert()
{1: 'a', 2: 'b', 3: 'c'}
>>> d1.subdict(['b', 'c', 'e'], default=None)
{'b': 2, 'c': 3, 'e': None}
>>> d1.sorted_keys()
OrderedDict([('a', 1), ('b', 2), ('c', 3)])
>>> d1.peek_key()
'a'
>>> d1.peek_value()
1

Next time you have a default configuration dictionary like and you allow the developer to pass keyword arguments to modify these behaviors, consider using dictionary intersection (&) to separate out only the relevant parts and dictionary union (|) to update those relevant parts. You can also use dictionary differences (-) if you need to check for unused arguments.

import ubelt as ub

def run_multiple_algos(**kwargs):
    algo1_defaults = {'opt1': 10, 'opt2': 11}
    algo2_defaults = {'src': './here/', 'dst': './there'}

    kwargs = ub.udict(kwargs)

    algo1_specified = kwargs & algo1_defaults
    algo2_specified = kwargs & algo2_defaults

    algo1_config = algo1_defaults | algo1_specified
    algo2_config = algo2_defaults | algo2_specified

    unused_kwargs = kwargs - (algo1_defaults | algo2_defaults)

    print('algo1_specified = {}'.format(ub.urepr(algo1_specified, nl=1)))
    print('algo2_specified = {}'.format(ub.urepr(algo2_specified, nl=1)))
    print(f'algo1_config={algo1_config}')
    print(f'algo2_config={algo2_config}')
    print(f'The following kwargs were unused {unused_kwargs}')

print(chr(10))
print('-- Run with some specified --')
run_multiple_algos(src='box', opt2='fox')
print(chr(10))
print('-- Run with extra unspecified --')
run_multiple_algos(a=1, b=2)

Produces:

-- Run with some specified --
algo1_specified = {
    'opt2': 'fox',
}
algo2_specified = {
    'src': 'box',
}
algo1_config={'opt1': 10, 'opt2': 'fox'}
algo2_config={'src': 'box', 'dst': './there'}
The following kwargs were unused {}


-- Run with extra unspecified --
algo1_specified = {}
algo2_specified = {}
algo1_config={'opt1': 10, 'opt2': 11}
algo2_config={'src': './here/', 'dst': './there'}
The following kwargs were unused {'a': 1, 'b': 2}

Find Duplicates

Find all duplicate items in a list. More specifically, ub.find_duplicates searches for items that appear more than k times, and returns a mapping from each duplicate item to the positions it appeared in.

>>> import ubelt as ub
>>> items = [0, 0, 1, 2, 3, 3, 0, 12, 2, 9]
>>> ub.find_duplicates(items, k=2)
{0: [0, 1, 6], 2: [3, 8], 3: [4, 5]}

Cross-Platform Config and Cache Directories

If you have an application which writes configuration or cache files, the standard place to dump those files differs depending if you are on Windows, Linux, or Mac. Ubelt offers a unified functions for determining what these paths are.

New in version 1.0.0: the ub.Path.appdir classmethod provides a way to achieve the above with a chainable object oriented interface.

The ub.Path.appdir(..., type='cache'), ub.Path.appdir(..., type='config'), and ub.Path.appdir(..., type='data') functions find the correct platform-specific location for these files and calling ensuredir ensures that the directories exist.

The config root directory is ~/AppData/Roaming on Windows, ~/.config on Linux and ~/Library/Application Support on Mac. The cache root directory is ~/AppData/Local on Windows, ~/.config on Linux and ~/Library/Caches on Mac.

Example usage on Linux might look like this:

>>> import ubelt as ub
>>> print(ub.Path.appdir('my_app').ensuredir().shrinkuser())  # default is cache
~/.cache/my_app
>>> print(ub.Path.appdir('my_app', type='config').ensuredir().shrinkuser())
~/.config/my_app

Symlinks

The ub.symlink function will create a symlink similar to os.symlink. The main differences are that 1) it will not error if the symlink exists and already points to the correct location. 2) it works* on Windows (*hard links and junctions are used if real symlinks are not available)

>>> import ubelt as ub
>>> dpath = ub.Path('ubelt', 'demo_symlink')
>>> real_path = dpath / 'real_file.txt'
>>> link_path = dpath / 'link_file.txt'
>>> real_path.write_text('foo')
>>> ub.symlink(real_path, link_path)

AutoDict - Autovivification

While the collections.defaultdict is nice, it is sometimes more convenient to have an infinitely nested dictionary of dictionaries.

>>> import ubelt as ub
>>> auto = ub.AutoDict()
>>> print('auto = {!r}'.format(auto))
auto = {}
>>> auto[0][10][100] = None
>>> print('auto = {!r}'.format(auto))
auto = {0: {10: {100: None}}}
>>> auto[0][1] = 'hello'
>>> print('auto = {!r}'.format(auto))
auto = {0: {1: 'hello', 10: {100: None}}}

String-based imports

Ubelt contains functions to import modules dynamically without using the python import statement. While importlib exists, the ubelt implementation is simpler to user and does not have the disadvantage of breaking pytest.

Note ubelt simply provides an interface to this functionality, the core implementation is in xdoctest (over as of version 0.7.0, the code is statically copied into an autogenerated file such that ubelt does not actually depend on xdoctest during runtime).

>>> import ubelt as ub
>>> try:
>>>     # This is where I keep ubelt on my machine, so it is not expected to work elsewhere.
>>>     module = ub.import_module_from_path(ub.expandpath('~/code/ubelt/ubelt'))
>>>     print('module = {!r}'.format(module))
>>> except OSError:
>>>     pass
>>>
>>> module = ub.import_module_from_name('ubelt')
>>> print('module = {!r}'.format(module))
>>> #
>>> try:
>>>     module = ub.import_module_from_name('does-not-exist')
>>>     raise AssertionError
>>> except ModuleNotFoundError:
>>>     pass
>>> #
>>> modpath = ub.Path(ub.util_import.__file__)
>>> print(ub.modpath_to_modname(modpath))
>>> modname = ub.util_import.__name__
>>> assert ub.Path(ub.modname_to_modpath(modname)).resolve() == modpath.resolve()

module = <module 'ubelt' from '/home/joncrall/code/ubelt/ubelt/__init__.py'>
>>> module = ub.import_module_from_name('ubelt')
>>> print('module = {!r}'.format(module))
module = <module 'ubelt' from '/home/joncrall/code/ubelt/ubelt/__init__.py'>

Related to this functionality are the functions ub.modpath_to_modname and ub.modname_to_modpath, which statically transform (i.e. no code in the target modules is imported or executed) between module names (e.g. ubelt.util_import) and module paths (e.g. ~/.local/conda/envs/cenv3/lib/python3.5/site-packages/ubelt/util_import.py).

>>> import ubelt as ub
>>> modpath = ub.util_import.__file__
>>> print(ub.modpath_to_modname(modpath))
ubelt.util_import
>>> modname = ub.util_import.__name__
>>> assert ub.modname_to_modpath(modname) == modpath

Horizontal String Concatenation

Sometimes its just prettier to horizontally concatenate two blocks of text.

>>> import ubelt as ub
>>> B = ub.urepr([[1, 2], [3, 4]], nl=1, cbr=True, trailsep=False)
>>> C = ub.urepr([[5, 6], [7, 8]], nl=1, cbr=True, trailsep=False)
>>> print(ub.hzcat(['A = ', B, ' * ', C]))
A = [[1, 2], * [[5, 6],
     [3, 4]]    [7, 8]]

Timing

Quickly time a single line.

>>> import math
>>> import ubelt as ub
>>> timer = ub.Timer('Timer demo!', verbose=1)
>>> with timer:
>>>     math.factorial(100000)
tic('Timer demo!')
...toc('Timer demo!')=0.1453s

External tools

Some of the tools in ubelt also exist as standalone modules. I haven't decided if its best to statically copy them into ubelt or require on pypi to satisfy the dependency. There are some tools that are not used by default unless you explicitly allow for them.

Code that is currently statically included (vendored):

• ProgIter - https://github.com/Erotemic/progiter
• OrderedSet - https://github.com/LuminosoInsight/ordered-set

Code that is completely optional, and only used in specific cases:

• Numpy - ub.urepr will format a numpy array nicely by default
• xxhash - this can be specified as a hasher to ub.hash_data
• Pygments - used by the util_color module.
• dateutil - used by the util_time module.

Similar Tools

UBelt is one of many Python utility libraries. A selection of similar libraries are listed here.

Libraries that contain a broad scope of utilities:

• Boltons: https://github.com/mahmoud/boltons
• Toolz: https://github.com/pytoolz/toolz
• CyToolz: https://github.com/pytoolz/cytoolz/
• UnStdLib: https://github.com/shazow/unstdlib.py

Libraries that contain a specific scope of utilities:

• More-Itertools: iteration tools: https://pypi.org/project/more-itertools/
• Funcy: functional tools: https://github.com/Suor/funcy
• Rich: pretty CLI displays - https://github.com/willmcgugan/rich
• tempora: time related tools - https://github.com/jaraco/tempora

Libraries that contain one specific data structure or utility:

• Benedict: dictionary tools - https://pypi.org/project/python-benedict/
• tqdm: progress bars - https://pypi.org/project/tqdm/
• pooch: data downloading - https://pypi.org/project/pooch/
• timerit: snippet timing for benchmarks - https://github.com/Erotemic/timerit

Ubelt is included in the the [bestof-python list](https://github.com/ml-tooling/best-of-python), which contains many other tools that you should check out.

History:

Ubelt is a migration of the most useful parts of utool(https://github.com/Erotemic/utool) into a standalone module with minimal dependencies.

The utool library contains a number of useful utility functions, but it also contained non-useful functions, as well as the kitchen sink. A number of the functions were too specific or not well documented. The ubelt is a port of the simplest and most useful parts of utool.

Note that there are other cool things in utool that are not in ubelt. Notably, the doctest harness ultimately became xdoctest. Code introspection and dynamic analysis tools were ported to xinspect. The more IPython-y tools were ported to xdev. Parts of it made their way into scriptconfig. The init-file generation was moved to mkinit. Some vim and system-y things can be found in vimtk.

Development on ubelt started 2017-01-30 and development of utool mostly stopped on utool was stopped later that year, but received patches until about 2020. Ubelt achieved 1.0.0 and removed support for Python 2.7 and 3.5 on 2022-01-07.

Notes.

PRs are welcome.

Also check out my other projects which are powered by ubelt:

• xinspect https://github.com/Erotemic/xinspect
• xdev https://github.com/Erotemic/xdev
• vimtk https://github.com/Erotemic/vimtk
• graphid https://github.com/Erotemic/graphid
• ibeis https://github.com/Erotemic/ibeis
• kwarray https://github.com/Kitware/kwarray
• kwimage https://github.com/Kitware/kwimage
• kwcoco https://github.com/Kitware/kwcoco

And my projects related to ubelt:

• ProgIter https://github.com/Erotemic/progiter
• Timerit https://github.com/Erotemic/timerit
• mkinit https://github.com/Erotemic/mkinit
• xdoctest https://github.com/Erotemic/xdoctest