On 2019-08-05 15:01:14, Kinney, Michael D wrote: > Laszlo, > > The context of this change is only to the PatchCheck.py tool. > and how that tool uses git show. > > I agree with the summary of very flexible capabilities in git > to help developers review different types of files. All of > those settings that were added to support UNI files in UTF-16 > file format were very valuable when we had to review the > text changes to those binary files. We should be using UTF-8 > now, and we can even extend PatchCheck.py to flag an error if > a UNI file is in UTF-16 format.
Ah. If we are to the point where we want to actively prevent utf-16 in the tree, then this sounds like a good idea. > I still prefer we make this change only to PatchCheck.py to > prevent false positives and print lines of text that can > not be found in a developer's working directory. I prefer > this one time change to PatchCheck.py instead of having to > update .gitattributes whenever the git show features are > extended to convert more binary files to text files. I think it's pretty rare for EDK II to add new binary file types, but I don't feel too strongly on this. I thought adding Laszlo's settings to .gitattributes might amount to solving two issues with one change. Since Laszlo acked this patch, I'll go ahead with: Reviewed-by: Jordan Justen <jordan.l.jus...@intel.com> > My expectation is that EDK II Maintainers need to review > if a binary file is ok or not for a repo. I would also be > ok with adding general rules to PatchCheck.py to generate > an error if a binary file is added/updated in one of the > text only repos (edk2, edk2-platforms) and not generate > an error if a binary file is added/updated in a repo that > supports binaries (edk-non-osi). -=-=-=-=-=-=-=-=-=-=-=- Groups.io Links: You receive all messages sent to this group. View/Reply Online (#45034): https://edk2.groups.io/g/devel/message/45034 Mute This Topic: https://groups.io/mt/32685494/21656 Group Owner: devel+ow...@edk2.groups.io Unsubscribe: https://edk2.groups.io/g/devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
