The purpose of this document is to list all of the notable changes to this project. The format was inspired by (but doesn’t strictly adhere to) Keep a Changelog. This project adheres to semantic versioning.

Release 10.0 (2021-09-17)

Noteworthy changes:

  • Merged pull request #45 to resolve the issue caused by the conditional pyreadline requirement on Windows not supporting Python 3.9+.
  • Updated the readme to use Python 3 in the example (reported in issue #56). Also added a mention of the humanfriendly --demo command.
  • Removed the humanfriendly.compat.unittest alias that presumably no-one is using at this point; it had been rendered useless quite a long time ago (requested in issue #53).

Internal changes:

  • Merged pull request #54 which migrates the humanfriendly project from Travis CI to GitHub Actions and from to Codecov.
  • Fixed a deprecation warning concerning setup.cfg and some Sphinx documentation errors.

Release 9.2 (2021-06-11)

Maintenance release:

  • Merged pull request #46 which fixes several flake8 warnings.
  • Merged pull request #49 which marks Python 3.9 support final.
  • Merged pull request #51 which helps to stabilize the test suite.
  • Merged pull request #52 which updates the humanfriendly.sphinx module to include Sphinx extension metadata that has become mandatory in a recent Sphinx release. After merging the pull request I added additional metadata including the version.

Release 9.1 (2020-12-10)

Added on_macos() function to detect Apple MacOS (I need this in an upcoming coloredlogs release and don’t want to have to think about how to detect MacOS again in the future 😇).

Release 9.0 (2020-12-01)

The major version number was bumped because the bug fix for pluralize() is backwards incompatible and (even though this seems like very “cosmetic” functionality) version numbers are cheap, so who cares 😉.

Bug fixes:


  • Enhanced concatenate() to support conjunction and serial_comma keyword arguments (as suggested in issue #30).
  • Added pluralize_raw() to select singular or plural form without prefixing the count to the text that is returned.

Release 8.2 (2020-04-19)

Added a simple case insensitive dictionary implementation, for details refer to the new module.

Release 8.1 (2020-03-06)


Internal changes:

  • Refactored the test suite to import all names separately instead of referring to identifiers via their modules (my preferences have changed since this code was written a long time ago).

Release 8.0 (2020-03-02)

This release is backwards incompatible in several ways, see the notes below.


  • Adopt functools.wraps() to make decorator functions more robust.

  • Make the Spinner class more customizable. The interval at which spinners are updated and the characters used to draw the animation of spinners can now be customized by callers. This was triggered by executor issue #2.


    The text cursor hiding behavior of spinners has been removed because it was found to be problematic (sometimes the text cursor would be hidden but not made visible again, which is disorienting to say the least).

  • Improve test skipping based on exception types.

    The humanfriendly.testing.TestCase class was originally created to enable skipping of tests that raise specific exception types on Python 2.6. This involved patching test methods, which had the unfortunate side effect of generating confusing pytest output on test failures.

    Since then unittest2 was integrated which provided real skipping of tests however removing the old test skipping support from the humanfriendly.testing module would have resulted in a backwards incompatible change, so I never bothered. I’ve now decided to bite the bullet and get this over with:

    1. I’ve implemented an alternative (finer grained) strategy based on a decorator function that applies to individual test methods, for details see humanfriendly.testing.skip_on_raise().
    2. I’ve removed the test method wrapping from the humanfriendly.testing.TestCase class.


    This change is backwards incompatible, in fact it breaks the test suites of two other projects of mine (executor and vcs-repo-mgr) because they depend on the old test method wrapping approach. Both test suites will need to be migrated to the skip_on_raise() decorator.

Internal changes:

Release 7.3 (2020-03-02)


Added the humanfriendly.deprecation.deprecated_args() decorator function which makes it easy to switch from positional arguments to keyword arguments without dropping backwards compatibility.


I’m still working on the humanfriendly 8.0 release which was going to break backwards compatibility in several ways if it wasn’t for the tools provided by the new humanfriendly.deprecation module.

Release 7.2 (2020-03-01)


Support for backwards compatible aliases that emit deprecation warnings (humanfriendly.deprecation).


I’m currently working on several large refactorings that involve moving things around between modules and dreaded having to extend the existing maze of (almost but not quite) cyclic import dependencies between modules. This new functionality will be adopted to untangle the existing maze in the upcoming humanfriendly 8.0 release, which bumps the major version number due to this very large change in how backwards compatibility is implemented. It is my hope that this new functionality will prove to be robust enough to unburden me from the less elegant aspects of preserving backwards compatibility 😁.


Get rid of broken references and noise in the online documentation once and for all:

  • Sphinx was emitting a screen full of warnings about unknown references. These were bothering me because testing the integration between Sphinx and humanfriendly.deprecation involved lots of broken references as well.
  • Additionally the humanfriendly.compat module introduced a lot of noise into the generated documentation because imported classes and their members were being included in the documentation, this is now also fixed.
  • Finally I decided to start using sphinx-build -nW to complain loudly when even just one broken reference is found. This should encourage the discipline to never introduce broken references again!


Fixed unittest deprecation warnings in the test suite.

Release 7.1.1 (2020-02-18)

Fix Python 3 incompatibility (distutils.spawn).

Much to my dismay this morning I ran into the following traceback on a Python 3.6 installation that is based on native Ubuntu (Debian) packages:

Traceback (most recent call last):
  File "...", line 1, in <module>
    from coloredlogs.syslog import enable_system_logging
  File ".../coloredlogs/", line 138, in <module>
    from humanfriendly import coerce_boolean
  File ".../humanfriendly/", line 25, in <module>
    from humanfriendly.tables import format_pretty_table as format_table
  File ".../humanfriendly/", line 32, in <module>
    from humanfriendly.terminal import (
  File ".../humanfriendly/", line 26, in <module>
    import distutils.spawn
ModuleNotFoundError: No module named 'distutils.spawn'

To enable local development and testing against lots of Python releases I use deadsnakes to install Python 2.7, 3.4, 3.5, 3.6, 3.7 and 3.8 at the same time. Before committing 335a69bae5 I did check the availability of the distutils.spawn module against my locally installed interpreters:

$ ls -l /usr/lib/python*/distutils/
-rw-r--r-- 1 root root 8.5K Nov  7 11:07 /usr/lib/python2.7/distutils/
-rw-r--r-- 1 root root 7.4K Mar 29  2019 /usr/lib/python3.4/distutils/
-rw-r--r-- 1 root root 7.3K Nov 24 02:35 /usr/lib/python3.5/distutils/
-rw-r--r-- 1 root root 7.3K Oct 28 17:30 /usr/lib/python3.6/distutils/
-rw-r--r-- 1 root root 7.7K Oct 28 17:30 /usr/lib/python3.7/distutils/
-rw-r--r-- 1 root root 7.7K Oct 28 17:30 /usr/lib/python3.8/distutils/

I took this to mean it would be available on all these versions. Furthermore the tests on Travis CI passed as well. I think this is because deadsnakes as well as Travis CI are closer to upstream (the official Python releases) whereas Debian and Ubuntu make significant customizations…

In any case this new commit should fix the issue by using shutil.which() on Python 3 instead.

Release 7.1 (2020-02-16)


  • Enable Windows native support for ANSI escape sequences. This was brought to my attention in coloredlogs issue #71 and coloredlogs pull request #72. My experiences with ANSI escape sequences started out as part of the coloredlogs package but eventually I moved the support for ANSI escape sequences to the humanfriendly package. This explains how it now makes sense to integrate the Windows native ANSI escape sequence support in humanfriendly as well.

Bug fixes:

  • Accept pluralized disk size units (#26). I’m not claiming this is a full solution to the problem, far from it. It does lessen the pain a bit (IMHO).
  • Make sure the selected pager is available before trying to run it. While testing humanfriendly on Windows 10 I noticed that humanfriendly --help resulted in nothing but a traceback, because less wasn’t available. That’s not human friendly at all 😕 (even if it is Windows 😈).

Release 7.0 (2020-02-16)

After an unplanned but extended hiatus from the development and maintenance of my open source projects I’m now finally starting to pick up some momentum, so I’m trying to make the best of it:

  • Merge pull request #24: Fix bug in parse_length() that rounded floats.
  • Merge pull request #32: Update hyperlinks in readme.
  • Merge pull request #33: Drop support for Python 2.6 and 3.0-3.4
  • Merge pull request #35: SVG badge in readme.
  • Merge pull request #36: Add support for nanoseconds and microseconds time units
  • Fixed format_rst_table() omission from humanfriendly.tables.__all__.
  • Start testing on Python 3.8 and 3.9-dev.

Release 6.1 (2020-02-10)

  • Added a :pypy:`…` role for easy linking to packages on the Python Package Index, for details refer to humanfriendly.sphinx.pypi_role().
  • Wasted quite a bit of time debugging a MacOS failure on Travis CI caused by a broken pip installation, fixed by using to bootstrap an installation that actually works 😉.

Release 6.0 (2020-02-09)

Noteworthy changes:

  • Enable MockedProgram to customize the shell script code of mocked programs. This was added to make it easy to mock a program that is expected to generate specific output (I’m planning to use this in the linux-utils test suite).
  • Defined __all__ for all public modules that previously lacked “export control” and decided to bump the major version number as a precaution:
    • These changes should not have any impact on backwards compatibility, unless I forgot entries, in which case callers can get ImportError exceptions…
    • Imports of public modules were previously exported (implicitly) and this pollutes code completion suggestions which in turn can encourage bad practices (not importing things using their “canonical” name).
    • I started developing the humanfriendly package years before I learned about the value of defining __all__ and so some modules lacked a definition until now. I decided that now was as good a time as any to add those definitions 😇.

Miscellaneous changes:

  • Simplified the headings in docs/api.rst so that only the module names remain. This was done because Sphinx doesn’t support nested links in HTML output and thus generated really weird “Table of Contents” listings.
  • Fixed the reStructuredText references in the documentation of prompt_for_choice(). This function is imported from humanfriendly.prompts to humanfriendly (for backwards compatibility) where it can’t use relative references to refer to the other functions in the humanfriendly.prompts module.
  • Changed the Makefile to default to Python 3 for development, make sure flake8 is always up-to-date and silence the few targets whose commands were not already silenced.
  • Embedded quite a few Python API references into recent changelog entries, just because I could (I ❤️ what hyperlinks can do for the usability of technical documentation, it gives a lot more context).

Release 5.0 (2020-02-06)

  • Added custom :man:`…` role for easy linking to Linux manual pages to the humanfriendly.sphinx module.

  • Changed rendering of pretty tables to expand tab characters to spaces:

    Until now pretty tables did not take the variable width of tab characters into account which resulted in tables whose “line drawing characters” were visually misaligned. Tabs are now expanded to spaces using str.expandtabs().

  • Stop testing on Python 2.6 and drop official support. The world (including Travis CI) has moved on and preserving Python 2.6 compatibility was clearly starting to drag the project down…

I decided to bump the major version number because each of these changes can be considered backwards incompatible in one way or another and version numbers are cheap anyway so there 😛.

Release 4.18 (2019-02-21)

  • Added humanfriendly.text.generate_slug() function.
  • Fixed “invalid escape sequence” DeprecationWarning (pointed out by Python >= 3.6).
  • Fought Travis CI (for way too long) in order to restore Python 2.6, 2.7, 3.4, 3.5, 3.6 and 3.7 compatibility in the Travis CI configuration (unrelated to the humanfriendly package itself).

Release 4.17 (2018-10-20)

  • Add Python 3.7 to versions tested on Travis CI and using tox and document compatibility with Python 3.7.

  • Add rudimentary caching decorator for functions:

    Over the years I’ve used several variations on this function in multiple projects and I’d like to consolidate all of those implementations into a single one that’s properly tested and documented.

    Due to the simplicity and lack of external dependencies it seemed kind of fitting to include this in the humanfriendly package, which has become a form of extended standard library for my Python projects 😇.

Release 4.16.1 (2018-07-21)

Yet another ANSI to HTML improvement: Emit an ANSI reset code before emitting ANSI escape sequences that change styles, so that previously activated styles don’t inappropriately “leak through” to the text that follows.

Release 4.16 (2018-07-21)

More HTML to ANSI improvements:

Release 4.15.1 (2018-07-14)

Bug fixes for HTML to ANSI conversion.

HTML entities were being omitted from conversion because I had neglected to define the handle_charref() and handle_entityref() methods (whose definitions are so conveniently given in the documentation of the HTMLParser class 😇).

Release 4.15 (2018-07-14)

Added the humanfriendly.terminal.html_to_ansi() function which is a shortcut for the humanfriendly.terminal.HTMLConverter class that’s based on html.parser.HTMLParser.

This new functionality converts HTML with simple text formatting tags like <b> for bold, <i> for italic, <u> for underline, <span> for colors, etc. to text with ANSI escape sequences.

I’m still working on that awesome new project (update: see chat-archive), this functionality was born there but seemed like a useful addition to the humanfriendly package, given the flexibility that this provides 😇.

Release 4.14 (2018-07-13)

Support for 24-bit (RGB) terminal colors. Works by accepting a tuple or list with three integers representing an RGB (red, green, blue) color.

Release 4.13 (2018-07-09)

Support for italic text rendering on the terminal.

I’m working on an awesome new project (update: see chat-archive) that’s almost ready to publish, but then I noticed that I couldn’t render italic text on the terminal using the humanfriendly package. I checked and sure enough my terminal supported it just fine, so I didn’t see any reason not to fix this now 😇.

Release 4.12.1 (2018-05-10)

It was reported in issue #28 that humanfriendly --demo didn’t work on Python 3 due to two unrelated TypeError exceptions. First I added a failing regression test to the test suite (here’s the failing build) and then I applied the changes suggested in issue #28, confirming that both issues are indeed fixed because the test now passes (here’s the successful build).

Release 4.12 (2018-04-26)

Release 4.11 (2018-04-26)

Added this changelog as requested in #23.

I’ve held off on having to keep track of changelogs in my open source programming projects until now (2018) because it’s yet another piece of bookkeeping that adds overhead to project maintenance versus just writing the damn code and throwing it up on GitHub :-p. However all that time I felt bad for not publishing change logs and I knew that requests would eventually come in and indeed in the past months I’ve received two requests in #23 and in issue #55 of coloredlogs.

I actually wrote a Python script that uses the git tag and git for-each-ref commands to automatically generate a CHANGELOG.rst “prototype” (requiring manual editing to clean it up) to bootstrap the contents of this document. I’m tempted to publish that now but don’t want to get sidetracked even further :-).

Release 4.10 (2018-03-31)

Added the humanfriendly.Timer.sleep() method to sleep “no more than” the given number of seconds.

Release 4.9 (2018-03-28)

Added the humanfriendly.tables.format_rst_table() function to render RST (reStructuredText) tables.

Release 4.8 (2018-01-20)

Added the humanfriendly.coerce_pattern() function. I previously created this for vcs-repo-mgr and now need the same thing in qpass so I’m putting it in humanfriendly :-) because it kind of fits with the other coercion functions.

Release 4.7 (2018-01-14)

Release 4.6 (2018-01-04)

Fixed issue #21 by implementing support for bright (high intensity) terminal colors.

Release 4.5 (2018-01-04)

Fixed issue #16 by merging pull request #17: Extend byte ranges, add RAM output to command line.

In the merge commit I removed the --format-bytes option that #17 added and instead implemented a --binary option which changes --format-size to use binary multiples of bytes (base-2) instead of decimal multiples of bytes (base-10).

Release 4.4.2 (2018-01-04)

  • Fixed ImportError exception on Windows due to interactive prompts (fixes #19 by merging #20.).
  • Enable MacOS builds on Travis CI and document MacOS compatibility.
  • Change Sphinx documentation theme.

Release 4.4.1 (2017-08-07)

Include the Sphinx documentation in source distributions (same rationales as for the similar change made to ‘coloredlogs’ and ‘verboselogs’).

Release 4.4 (2017-07-16)

Added the make_dirs() and touch() functions.

Release 4.3 (2017-07-10)

Don’t log duplicate output in run_cli().

Release 4.2 (2017-07-10)

Automatically reconfigure logging in run_cli().

Release 4.1 (2017-07-10)

Improve run_cli() to always log standard error as well.

Release 4.0 (2017-07-10)

Backwards incompatible improvements to run_cli().

I just wasted quite a bit of time debugging a Python 3.6 incompatibility in deb-pkg-tools (see build 251688788) which was obscured by my naive implementation of the run_cli() function. This change is backwards incompatible because run_cli() now intercepts all exceptions whereas previously it would only intercept SystemExit.

Release 3.8 (2017-07-02)

Make it easy to mock the $HOME directory.

Release 3.7 (2017-07-01)

Enable customizable skipping of tests.

Release 3.6.1 (2017-06-24)

Improved the robustness of the PatchedAttribute and PatchedItem classes.

Release 3.6 (2017-06-24)

  • Made the retry limit in interactive prompts configurable.
  • Refactored the makefile and Travis CI configuration.

Release 3.5 (2017-06-24)

Added humanfriendly.testing.TestCase.assertRaises() enhancements.

Release 3.4.1 (2017-06-24)

Bug fix for Python 3 syntax incompatibility.

Release 3.4 (2017-06-24)

Promote the command line testing function to the public API.

Release 3.3 (2017-06-24)

Release 3.2 (2017-05-18)

Added the humanfriendly.terminal.output() function to auto-encode terminal output to avoid encoding errors and applied the use of this function in various places throughout the package.

Release 3.1 (2017-05-06)

Improved usage message parsing and rendering.

While working on a new project I noticed that the join_lines() call in render_usage() could corrupt lists as observed here: (check the part after ‘Supported locations include:’)

To be honest I’m not even sure why I added that join_lines() call to begin with and I can’t think of any good reasons to keep it there, so gone it is!

Release 3.0 (2017-05-04)

  • Added support for min, mins abbreviations for minutes based on #14.
  • Added Python 3.6 to supported versions on Travis CI and in documentation.

I’ve decided to bump the major version number after merging pull request #14 because the humanfriendly.time_units data structure was changed. Even though this module scope variable isn’t included in the online documentation, nothing stops users from importing it anyway, so this change is technically backwards incompatible. Besides, version numbers are cheap. In fact, they are infinite! :-)

Release 2.4 (2017-02-14)

Make usage() and show_pager() more user friendly by changing how less as a default pager is invoked (with specific options).

Release 2.3.2 (2017-01-17)

Bug fix: Don’t hard code conditional dependencies in wheels.

Release 2.3.1 (2017-01-17)

Fix parse_usage() tripping up on commas in option labels.

Release 2.3 (2017-01-16)

  • Switch to monotonic clock for timers based on #13.
  • Change to everywhere.
  • Improve intersphinx references in documentation.
  • Minor improvements to setup script.

Release 2.2.1 (2017-01-10)

  • Improve use of timers as context managers by returning the timer object (as originally intended).
  • Minor improvements to reStructuredText formatting in various docstrings.

Release 2.2 (2016-11-30)

  • Fix and add a test for parse_date() choking on Unicode strings.
  • Only use “readline hints” in prompts when standard input is a tty.

Release 2.1 (2016-10-09)

Added clean_terminal_output() function to sanitize captured terminal output.

Release 2.0.1 (2016-09-29)

Update README.rst based on the changes in 2.0 by merging #12.

Release 2.0 (2016-09-29)

Proper support for IEEE 1541 definitions of units (fixes #4, merges #8 and #9).

Release 1.44.9 (2016-09-28)

  • Fix and add tests for the timespan formatting issues reported in issues #10 and #11.
  • Refactor Makefile, switch to py.test, add wheel support, etc.

Release 1.44.8 (2016-09-28)

  • Fixed issue #7 (TypeError when calling show_pager() on Python 3) and added a test.
  • Minor improvements to the script.
  • Stop testing tags on Travis CI.

Release 1.44.7 (2016-04-21)

Minor improvements to usage message reformatting.

Release 1.44.6 (2016-04-21)

Remove an undocumented .strip() call from join_lines().

Why I noticed this: It has the potential to eat significant white space in usage messages that are marked up in reStructuredText syntax.

Why I decided to change it: The behavior isn’t documented and on second thought I wouldn’t expect a function called join_lines() to strip any and all leading/trailing white space.

Release 1.44.5 (2016-03-20)

Improved the usage message parsing algorithm (also added a proper test). Refer to test_parse_usage_tricky() for an example of a usage message that is now parsed correctly but would previously confuse the dumb “parsing” algorithm in parse_usage().

Release 1.44.4 (2016-03-15)

Made usage message parsing a bit more strict. Admittedly this still needs a lot more love to make it more robust but I lack the time to implement this at the moment. Some day soon! :-)

Release 1.44.3 (2016-02-20)

Unbreak conditional importlib dependency after breakage observed here:

Release 1.44.2 (2016-02-20)

  • Make conditional importlib dependency compatible with wheels: While running tox tests of another project of mine that uses the humanfriendly package I noticed a traceback when importing the humanfriendly package (because importlib was missing). After some digging I found that tox uses pip to install packages and pip converts source distributions to wheel distributions before/during installation, thereby dropping the conditional importlib dependency.
  • Added the Sphinx extension trove classifier to the script.

Release 1.44.1 (2016-02-18)

  • Fixed a non-fatal but obviously wrong log format error in prompt_for_choice().
  • Added Python 3.5 to supported versions on Travis CI and in the documentation.

Release 1.44 (2016-02-17)

Added the humanfriendly.sphinx module with automagic usage message reformatting and a bit of code that I’d been copying and pasting between docs/ scripts for years to include magic methods, etc in Sphinx generated documentation.

Release 1.43.1 (2016-01-19)

Bug fix for Python 2.6 compatibility in script.

Release 1.43 (2016-01-19)

Replaced import_module() with a conditional dependency on importlib.

Release 1.42 (2015-10-23)

Added proper tests for ANSI escape sequence support.

Release 1.41 (2015-10-22)

  • Moved hard coded ANSI text style codes to a module level ANSI_TEXT_STYLES dictionary.
  • Improved the related error reporting based on the new dictionary.

Release 1.40 (2015-10-22)

Added support for custom delimiters in humanfriendly.text.split().

Release 1.39 (2015-10-22)

Added the humanfriendly.compat module to group Python 2 / 3 compatibility logic.

Release 1.38 (2015-10-22)

  • Added the prompt_for_confirmation() function to render (y/n) prompts.
  • Improved the prompt rendered by prompt_for_choice().
  • Extracted supporting prompt functionality to separate functions.

Release 1.37 (2015-10-22)

  • Added support for wrapping ANSI escape sequences in “readline hints”.
  • Work around incompatibility between flake8-pep257==1.0.3 and pep257==0.7.0.

Release 1.36 (2015-10-21)

Added message() and warning() functions to write informational and warning messages to the terminal (on the standard error stream).

Release 1.35 (2015-09-10)

Implemented the feature request in issue #6: Support for milleseconds in timespan parsing/formatting. Technically speaking this breaks backwards compatibility but only by dropping a nasty (not documented) implementation detail. Quoting from the old code:

# All of the first letters of the time units are unique, so
# although this check is not very strict I believe it to be
# sufficient.

That no longer worked with [m]illiseconds versus [m]inutes as was also evident from the feature request / bug report on GitHub.

Release 1.34 (2015-08-06)

Implemented and added checks to enforce PEP-8 and PEP-257 compliance.

Release 1.33 (2015-07-27)

Added format_length() and parse_length()` functions via pull request #5.

Release 1.32 (2015-07-19)

Added the humanfriendly.text.split() function.

Release 1.31 (2015-06-28)

Added support for rendering of usage messages to reStructuredText.

Release 1.30 (2015-06-28)

Started moving functions to separate modules.

Release 1.29 (2015-06-24)

Added the parse_timespan() function.

Release 1.28 (2015-06-24)

Extracted the “new” tokenize() function from the existing parse_size() function.

Release 1.27 (2015-06-03)

Changed table formatting to right-align table columns with numeric data (and pimped the documentation).

Release 1.26 (2015-06-02)

Make table formatting ‘smart’ by having it automatically handle overflow of columns by switching to a different more verbose vertical table layout.

Release 1.25.1 (2015-06-02)

  • Bug fix for a somewhat obscure UnicodeDecodeError in on Python 3.
  • Travis CI now also runs the test suite on PyPy.
  • Documented PyPy compatibility.

Release 1.25 (2015-05-27)

Added the humanfriendly.terminal.usage() function for nice rendering of usage messages on interactive terminals (try humanfriendly --help to see it in action).

Release 1.24 (2015-05-27)

Added the humanfriendly.terminal module with support for ANSI escape sequences, detecting interactive terinals, finding the terminal size, etc.

Release 1.23.1 (2015-05-26)

Bug fix for Python 3 compatibility in format_table().

Release 1.23 (2015-05-26)

Added format_table() function to format tabular data in simple textual tables.

Release 1.22 (2015-05-26)

Added additional string formatting functions compact(), dedent(), format(), is_empty_line() and trim_empty_lines().

Release 1.21 (2015-05-25)

Added support for formatting numbers with thousands separators.

Release 1.20 (2015-05-25)

  • Added a simple command line interface.
  • Added trove classifiers to

Release 1.19 (2015-05-23)

Made it possible to use spinners as context managers.

Release 1.18 (2015-05-23)

Added a Spinner.sleep() method.

Release 1.17 (2015-05-23)

  • Improved interaction between spinner & verbose log outputs: The spinner until now didn’t end each string of output with a carriage return because then the text cursor would jump to the start of the screen line and disturb the spinner, however verbose log output and the spinner don’t interact well because of this, so I’ve decided to hide the text cursor while the spinner is active.
  • Added another example to the documentation of parse_date().

Release 1.16 (2015-03-29)

  • Change spinners to use the ‘Erase in Line’ ANSI escape code to properly clear screen lines.
  • Improve performance of Travis CI and increase multiprocessing test coverage.

Release 1.15 (2015-03-17)

  • Added support for AutomaticSpinner that animates without requiring step() calls.
  • Changed the Python package layout so that all *.py files are in one directory.
  • Added tests for parse_path() and Timer.rounded.

Release 1.14 (2014-11-22)

  • Changed coerce_boolean() to coerce empty strings to False.
  • Added parse_path() function (a simple combination of standard library functions that I’ve repeated numerous times).

Release 1.13 (2014-11-16)

Added support for spinners with an embedded timer.

Release 1.12 (2014-11-16)

Added support for rounded timestamps.

Release 1.11 (2014-11-15)

Added coerce_boolean() function.

Release 1.10 (2014-11-15)

Improved pluralize() by making it handle the simple case of pluralizing by adding ‘s’.

Release 1.9.6 (2014-09-14)

Improved the documentation by adding a few docstring examples via pull request #3.

Release 1.9.5 (2014-06-29)

Improved the test suite by making the timing related tests less sensitive to slow test execution. See but the same thing can happen anywhere. When looked at from that perspective the fix I’m committing here really isn’t a fix, but I suspect it will be fine :-).

Release 1.9.4 (2014-06-29)

Release 1.9.3 (2014-06-29)

  • Automatically disable Spinner when stream.isatty() returns False.
  • Improve the makefile by adding install and coverage targets.
  • Remove the makefile generated by Sphinx (all we need from it is one command).
  • Add unit tests for prompt_for_choice() bringing coverage back up to 95%.

Release 1.9.2 (2014-06-29)

Added support for ‘B’ bytes unit to parse_size() via pull request #2.

Release 1.9.1 (2014-06-23)

Improved the prompt_for_choice() function by clearly presenting the default choice (if any).

Release 1.9 (2014-06-23)

Added the prompt_for_choice() function.

Release 1.8.6 (2014-06-08)

Enable Spinner to show progress counter (percentage).

Release 1.8.5 (2014-06-08)

Make Timer objects “resumable”.

Release 1.8.4 (2014-06-07)

Make the Spinner(label=...) argument optional.

Release 1.8.3 (2014-06-07)

Make it possible to override the label for individual steps of spinners.

Release 1.8.2 (2014-06-01)

Automatically rate limit Spinner instances.

Release 1.8.1 (2014-05-11)

  • Improve Python 3 compatibility: Make sure sequences passed to concatenate() are lists.
  • Submit test coverage from Travis CI to
  • Increase test coverage of concatenate(), Spinner() and Timer().
  • Use assertRaises() instead of try, except and isinstance() in test suite.

Release 1.8 (2014-05-10)

  • Added support for Python 3 thanks to a pull request.
  • Document the supported Python versions (2.6, 2.7 and 3.4).
  • Started using Travis CI to automatically run the test suite.

Release 1.7.1 (2013-09-22)

Bug fix for concatenate() when given only one item.

Release 1.7 (2013-09-22)

Added functions concatenate() and pluralize(), both originally developed in private scripts.

Release 1.6.1 (2013-09-22)

Bug fix: Don’t raise an error in format_path() if $HOME isn’t set.

Release 1.6 (2013-08-12)

Added a Spinner class that I originally developed for pip-accel.

Release 1.5 (2013-07-07)

Added a Timer class to easily keep track of long running operations.

Release 1.4.3 (2013-07-06)

Fixed various edge cases in format_path(), making it more robust.

Release 1.4.2 (2013-06-27)

Improved the project description in and added a link to online documentation on PyPI.

Release 1.4.1 (2013-06-27)

Renamed the package from human-friendly to humanfriendly.

Release 1.4 (2013-06-17)

Added the parse_date() function.

Release 1.3.1 (2013-06-17)

  • Improved the documentation by adding lots of examples.
  • Renamed the arguments to the format_size() and format_timespan() functions.

Release 1.3 (2013-06-17)

Added the format_timespan() function.

Release 1.2 (2013-06-17)

Started using Sphinx to generate API documentation from docstrings.

Release 1.1 (2013-06-17)

Added the format_path() function.

Release 1.0 (2013-06-17)

The initial commit of the project, created by gathering functions from various personal scripts that I wrote over the past years.