Josh Triplett [off-list ref] writes:
quoted
- The "everybody will have a single SP at the end" may or may not
last forever;
Trivially fixed if that ever changes, but given the nature of all of
these, that seems unlikely.
Why? Because we encourage to write tests that are expected to find
breakages, some of these test vector lines may have to show a broken
line that lacks SP after label + colon.
quoted
- With your scheme, if you already had _one_ trailing SPs in the
input, it would be hard to spot in the source;
Git makes them quite difficult to miss. :)
That is irrelevant, isn't it?
This is about protecting the source in the editor, before you run
"git show --whitespace=trailing-space", "git diff --check", etc.
On Mon, Dec 30, 2013 at 01:05:25PM -0800, Junio C Hamano wrote:
Josh Triplett [off-list ref] writes:
quoted
quoted
- With your scheme, if you already had _one_ trailing SPs in the
input, it would be hard to spot in the source;
Git makes them quite difficult to miss. :)
That is irrelevant, isn't it?
This is about protecting the source in the editor, before you run
"git show --whitespace=trailing-space", "git diff --check", etc.
That was exactly my point: such lines shouldn't exist, and rather than
including the trailing space and following it with a character that then
needs removing, it seems more sensible to me to omit the trailing space
and insert it via an almost identical sed line. Git already helps
ensure that trailing space won't exist on *any* line, including those; I
don't see how an extra character after the space (making it no longer
trailing space) makes it any more or less likely that those lines would
have trailing space.
In any case, I don't care enough to argue the point further; it was just
a style suggestion.
- Josh Triplett