Torsten Bögershausen [off-list ref] writes:
Am 2012-07-30 11:57, schrieb Michael J Gruber:
(Sorry being late)
That line:
quoted
skip_all="filesystem does not convert utf-8 nfd to nfc"
shouldn't it be the other way around?
skip_all="filesystem does not convert utf-8 nfc to nfd"
(and may be the following may be more easy to read:)
skip_all="filesystem does not decompose unicode"
Side note: t0050 uses another term:
"will test on a unicode corrupting filesystem"
I am OK with either of the last two. Thanks for noticing.
Junio C Hamano venit, vidit, dixit 05.08.2012 00:04:
Torsten Bögershausen [off-list ref] writes:
quoted
Am 2012-07-30 11:57, schrieb Michael J Gruber:
(Sorry being late)
That line:
quoted
skip_all="filesystem does not convert utf-8 nfd to nfc"
shouldn't it be the other way around?
skip_all="filesystem does not convert utf-8 nfc to nfd"
(and may be the following may be more easy to read:)
skip_all="filesystem does not decompose unicode"
Side note: t0050 uses another term:
"will test on a unicode corrupting filesystem"
I am OK with either of the last two. Thanks for noticing.
I had taken the test description from t3910. But that referred to Git's
undoing of the conversion done by the filesystem...
It's best to have it analogous to t0050 as in Junio's fixup.
Thanks!
Michael
[resent with cc: list...]