Git currently correctly detects them as binary; the changes are done to
avoid depending on the auto-detection and also for documentation.
Commands used to list all of the files that git detects as non-text
files:
$ git ls-files --eol | grep -e 'i/-text' -e 'w/-text'
i/-text w/-text attr/text=auto eol=lf etc-fixes/seccomp-join-bug/eecf35c-backports.zip
i/-text w/-text attr/text=auto eol=lf test/appimage/Leafpad-0.8.17-x86_64.AppImage
i/-text w/-text attr/text=auto eol=lf test/appimage/Leafpad-0.8.18.1.glibc2.4-x86_64.AppImage
i/-text w/-text attr/text=auto eol=lf test/filters/memwrexe
i/-text w/-text attr/text=auto eol=lf test/filters/memwrexe-32
i/-text w/-text attr/text=auto eol=lf test/filters/namespaces
i/-text w/-text attr/text=auto eol=lf test/filters/namespaces-32
Note: The committed seccomp filters do not have a file extension, so
ignore them for now.
This should make it easier to avoid whitespace errors, as long as the
editor used supports it (either natively or through a plugin).
See the editorconfig website for the editors that support it:
* https://editorconfig.org
Note: All text files appear to already be using LF and UTF-8 (or ASCII):
$ git ls-files --eol | grep -v -e '^i/lf w/lf' \
-e 'i/none w/none' -e 'i/-text w/-text'
i/ w/ attr/text=auto eol=lf ci/check/profiles/sort.py
$ git ls-files -z | xargs -0 file -i -h | sed 's/[^:]*: *//' |
grep -v -e 'charset=binary' -e 'charset=us-ascii' |
LC_ALL=C sort | uniq -c
1 text/html; charset=utf-8
2 text/plain; charset=utf-8
1 text/x-c; charset=utf-8
