Default to a sensible encoding #315
Conversation
|
There was some discussion around this over in #212 (comment) before we created these images -- @robilad do you think this makes sense while https://openjdk.java.net/jeps/8187041 is still in-progress? Do you think it would make sense to link to https://openjdk.java.net/jeps/8187041 in a comment here (both as a reminder for us and as a pointer for users looking for further reading)? |
|
The issue you refer to is unfortunately closed. Seems to have been a passing conversation that wasn't really addressed at the time. With the current defaults, |
|
So what are we doing about this? @tianon, noticed you reacted with a confused emoji; is there anything you need me to clarify about the pull request? |
|
Apologies, I lost track of this one. I've just pushed a rebase and an extra commit to update the template and version 13; if the CI passes, I think we're good to go. 👍 |
Changes: - docker-library/openjdk@2544422: Merge pull request docker-library/openjdk#315 from obsidiandynamics/master - docker-library/openjdk@554ff38: Update template and apply to both 12 and 13 - docker-library/openjdk@349703e: Default to a sensible encoding
Will work with files that contain unicode characters, closest to Debian's
C.UTF-8on distros where the latter tends to be omitted out of the box (Arch, Fedora, CentOS, RedHat, Oracle, etc.).Note: The default (if
LANGis unspecified) is justen_US(well, actually depending on the installation locale), which is ISO 8859-1 — essentially ASCII with an 8th bit. Given Java's native support for unicode, leaving the defaulten_USis less than ideal.