Assume strings are utf8 by default and fallback on encoding if necessary #318
Conversation
Before this patch, we always assumed the string was ascii-8bit which does not work correctly with real utf8 strings (french accented chars or chineese chars). This patch tries to use the string as utf8 and fallback if necessary to the the ascii-8bit assumption if necessary. This should help with ruby-git#295 and might replace the need for ruby-git#301. Change-Id: Idac63fa10e5aefafa1eb99a6be4138cac5f90ea0
|
Sorry for the removal of ruby 1.9.2 without prior discussion. At the moment tests cannot pass (even in master) for ruby 1.9.2 (see my comment on my second commit). I'd be glad to discuss it. |
ruby 1.9.2 has been out of support for 2,5 years (https://www.ruby-lang.org/en/news/2014/07/01/eol-for-1-8-7-and-1-9-2/) The motivation behind this removal is the fact that all dependencies of this gem (at least rake and rdoc) do not support ruby 1.9.2 in their latest versions. Keeping support would require to constraint their version to very old version (on ruby 1.9.2 only). Actually even 1.9.3 has been out of support for 1,5 years but it does not cause test problem yet. Change-Id: Iad7c3da74a19ebc37c3df4203f7e55a0b5aa70dc
|
@rvodden would you consider this PR if I fixed the conflicts? |
|
This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions. |
|
@perlun I am willing to fix the conflicts if you are interested in the PR. let me know |
|
@kamaradclimber I think #327 supersedes this. Could you try with the latest master and see if this assumption is right? If not, we can discuss it further - I'm open to changes that fixes bugs in the implementation. |
|
thanks for the answer. I'll close the PR and test latest master |
Before this patch, we always assumed the string was ascii-8bit which
does not work correctly with real utf8 strings (french accented chars or
chineese chars).
This patch tries to use the string as utf8 and fallback if necessary to
the the ascii-8bit assumption if necessary.
This should help with #295 and might replace the need for #301.
Change-Id: Idac63fa10e5aefafa1eb99a6be4138cac5f90ea0