author | Jonathan Nieder <jrnieder@gmail.com> | |
Fri, 6 May 2011 21:59:47 +0000 (16:59 -0500) | ||
committer | Junio C Hamano <gitster@pobox.com> | |
Fri, 6 May 2011 23:03:37 +0000 (16:03 -0700) | ||
commit | 8c5cea0095ee57c36fddfb9016810c230ce26557 | |
tree | 8d5ff7bf4b5f95f926d93a61be18fc0e29b62008 | tree | snapshot |
parent | f23e8decd544c8f81b40c98211f6cb028fa70cdc | commit | diff |
tests: check git does not barf on merge.ff values for future versions of git
Maybe some day in the future we will want to support a syntax
like
[merge]
ff = branch1
ff = branch2
ff = branch3
in addition to the currently permitted "true", "false", and "only"
values. So make sure we continue to treat such configurations as
though an unknown variable had been defined rather than erroring out,
until it is time to implement such a thing, so configuration files
using such a facility can be shared between present and future git.
While at it, add a few missing && and start the "combining --squash
and --no-ff" test with a known state so we can be sure it does not
succeed or fail for the wrong reason.
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Maybe some day in the future we will want to support a syntax
like
[merge]
ff = branch1
ff = branch2
ff = branch3
in addition to the currently permitted "true", "false", and "only"
values. So make sure we continue to treat such configurations as
though an unknown variable had been defined rather than erroring out,
until it is time to implement such a thing, so configuration files
using such a facility can be shared between present and future git.
While at it, add a few missing && and start the "combining --squash
and --no-ff" test with a known state so we can be sure it does not
succeed or fail for the wrong reason.
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
t/t7600-merge.sh | diff | blob | history |