* Given the nature of the project, the complexity of the codebase and the high level of quality assurance upstream, it is preferable to SRU a minor release rather than cherry-pick selected bug fixes.
* No specific test case, bugs fixed upstream hopefully come with unit/regression tests, and the release itself is extensively exercised upstream (both in an automated manner and manually) by a community of testers. Each minor release went through 2 release candidates.
* The libreoffice packages include autopkgtests, those should be run and verified to pass.
* General smoke testing of all the applications in the office suite should be carried out.
[Regression Potential]
* Two new minor releases with a total of 134 bug fixes always carry the potential for introducing regressions, even though they are bugfix-only releases, meaning that no new features were added, and no existing features were removed.
* A combination of autopkgtests and careful smoke testing as described above should provide reasonable confidence that no regressions sneaked in.
[Impact]
* LibreOffice 5.4.4 is the fourth bugfix release of the still 5.4 line. Version 5.4.2 is currently in 17.10. /wiki.documentf oundation. org/Releases/ 5.4.3/RC1# List_of_ fixed_bugs /wiki.documentf oundation. org/Releases/ 5.4.3/RC2# List_of_ fixed_bugs /wiki.documentf oundation. org/Releases/ 5.4.4/RC1# List_of_ fixed_bugs /wiki.documentf oundation. org/Releases/ 5.4.4/RC2# List_of_ fixed_bugs
For a list of fixed bugs compared to 5.4.2 see the list of bugs fixed in the RC1 and RC2 for 5.4.3 and the RC1 and RC2 for 5.4.4:
https:/
https:/
https:/
https:/
* Given the nature of the project, the complexity of the codebase and the high level of quality assurance upstream, it is preferable to SRU a minor release rather than cherry-pick selected bug fixes.
* Libreoffice 5.4.4 (RC2, which is effectively the final release) has been available for users to test since 2017-12-13 in the official "LibreOffice Pre-Releases" PPA (https:/ /launchpad. net/~libreoffic e/+archive/ ubuntu/ libreoffice- prereleases/ +packages), and it is in bionic-proposed since 2018-01-26.
[Test Case]
* No specific test case, bugs fixed upstream hopefully come with unit/regression tests, and the release itself is extensively exercised upstream (both in an automated manner and manually) by a community of testers. Each minor release went through 2 release candidates.
* The libreoffice packages include autopkgtests, those should be run and verified to pass.
* General smoke testing of all the applications in the office suite should be carried out.
[Regression Potential]
* Two new minor releases with a total of 134 bug fixes always carry the potential for introducing regressions, even though they are bugfix-only releases, meaning that no new features were added, and no existing features were removed.
* A combination of autopkgtests and careful smoke testing as described above should provide reasonable confidence that no regressions sneaked in.