Releases: bbc/rd-apmm-python-lib-mediatimestamp
5.0.0
What's Changed
- Regenerate static files by @j616 in #81
- Add commit check workflow by @j616 in #82
- Regenerate static files by @GeorginaShippey in #83
- Regenerate static files by @j616 in #84
- Regenerate static files by @philipnbbc in #85
- Add to_float timestamp methods and further support to negative timestamps by @philipnbbc in #86
New Contributors
- @GeorginaShippey made their first contribution in #83
Full Changelog: 4.0.1...5.0.0
4.0.1
What's Changed
- Regenerate static files by @samdbmg in #74
- Add support for GHA by @AndrewGibb in #75
- Regenerate static files by @j616 in #77
- Update static files by @AndrewGibb in #76
- Regenerate static files by @AndrewGibb in #78
- Remove Jenkins files and update README.md by @AndrewGibb in #79
- Regenerate static files by @j616 in #80
Full Changelog: 4.0.0...4.0.1
Major Version 2
- Dropped all support for python 2.7
- Swapped the mutable and immutable defaults
- Removed the inclusion of the constants at the top level
- Removed all mutable timestamps
Final version supporting python 2.7
All python 2.7 support will be dropped after this version, except for essential bugfixes.
If you need future features please use Python 3.
v1.7.0
== 1.7.0
Added extend_to_encompass_timerange function for immutable.TimeRange.
Hardcode use of python3.4 in RPM spec file to workaround missing python3 soft link in recent (>3.4.9) centos python34 RPM.
== 1.6.0
Removed the PRESERVE_START and PRESERVE_END rounding options which are used in TimeRange.normalise().
== 1.5.2
Fixed bug in taking unions with empty ranges
== 1.5.1
Fixed bug in TimeOffset.to_count (both versions) that caused incorrect results when rounding down when the denominator of the rate properly divided neither the seconds part nor the nanoseconds part of the timestamp, nor the number of nanoseconds in a second, but did properly divide the whole timestamp expressed in nanoseconds. This bug has been present for some time but had not previously been seen because there was no test coverage for non-nearest rounding and none of our tests used rates with prime denominators other than 2.
Switched test runner in tox from nose2 to unittest
== 1.5.0
Added normalisation function for immutable.TimeRange
== 1.4.0
Added richer comparison functions for immutable.TimeRange
== 1.3.0
Added at_rate method to immutable.TimeRange which returns an iterable of Timestamps
== 1.2.0
Added new immutable timestamps, offsets, and ranges in their own namespace
New style immutable timestamps can be negative
New style immutable timestamps have a more useful repr
== 1.1.3
Remove unused custom install command from tox.ini.
Add Jenkins build trigger to rebuild master every day.
Allow build to run on a wider variety of slaves.
== 1.1.2
BUGFIX: upload all pydocs
== 1.1.1
Added upload of pydoc docs to Jenkinsfile
== 1.1.0
Added hypothesis submodule containing strategies compatible with the hypothesis library which can be used to generate timestamps when testing code which uses this library.
== 1.0.2
Updated docstrings to note that "seconds:nanoseconds" is preferred over "seconds.fraction".
== 1.0.1
Added MAX_NANOSEC and MAX_SECONDS to TimeOffset
Initial PyPi Release
- Initial version, porting timestamp components from nmos-common v.0.6.0
- Has Jenkinsfile for automation
- Has TimeRange
- TimeOffset.eq doesn't raise exceptions when called with a different type