►
From YouTube: Kubernetes SIG CLI 20210428 - bug scrub
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
Good
morning,
good
evening,
good
afternoon,
depending
on
where
you
are
today
is
april
28th-
and
this
is
another
of
our
monthly
sixth
july
boxcroft-
my
name
is
mate,
but
our
today's
host
is
actually
eddie
so
I'll
pass
the
baton
to
eddie.
B
Hello,
I'm
gonna
share
my
screen.
B
This
all
right,
so
matcha
opened
up
this
issue.
This
is
10
46..
So
let's
start
here.
This
is
to
hold
the
umbrella
issue
for
refactoring
the
cube
control
commands
matcha.
You
want
to
drive
this.
A
Yeah,
so
I'll
quickly,
if
you
can
open
the
weight
command.
So
this
is
something
that
it'll
take
at
least
a
couple
of
releases.
I'm
not
expecting
us
to
quickly
get
it
through
the
line,
but
I
would
like
to
start
this
thing
rolling.
So,
in
short,
we
currently
have
an
option
structure
and
with
every
option
structure
in
most
probably
all
of
the
commands,
there's
usually
complete,
validate,
run
methods.
A
This
is
good
because
you
can
actually
easily
unit
test
every
single
command.
The
downside
is
that
we
are
very
tightly
coupled
with
cobra
flags,
so
if,
at
some
point
in
time,
eventually
in
the
future,
we
would
love
to
change
from
cobra
to
some
new
fancy,
cli
library
that
would
be
more
powerful
than
cobra.
A
It
will
be
very
hard.
We
would
like
to
be
able
to
even
split
this
up
more
by
having
a
flag
structure
and
option
structure
such
that
black
structure
will
be
strictly
for
wiring.
The
flags
and
options
will
be
the
actual
logic
of
every
single
command.
Wade
is
a
perfect
example
of
static
command,
where
this
is
strictly.
A
There
are
interesting
stuff
also
in
the
white
flags,
where
we
are
using
printer
flags
resource
builder
flags,
but
that
would
be
something
that,
while
you
will
be
working
on,
every
single
command
you'll
have
to
tackle
one
by
one.
So
wait
is
probably
the
simplest
example
I'll
be
working
on
get
and
logs
I'll,
probably
put
information
that
I'm
picking
those
two
already
for
others
to
know,
and
we
would
like
to
push
this
forward.
A
What
I
would
like
to
achieve
at
some
point
in
time,
but
that
won't
be
a
problem
for
122,
maybe
closer
to
123
when
we
have
at
least
couple
commands
done.
I
would
like
to
turn
this
description
of
a
structure
into
a
cap
for
future
generation.
A
The
reason
for
that
is
currently
a
lot
of
the
a
lot
of
people
will
come
to
us
asking
what
the
command
looks
like
how
how
to
write
one,
how
to
work
with
this.
We
have
a
lot
of
dogs
and
myself
and
sean
and
eddie
and
phil.
We
were
talking
about
a
structure
of
every
single
command,
but
if
we
write
it
down
as
a
cap
that
hopefully
in
the
long
run,
should
help
people
understand
I'll
try
to
throw
a
little
bit
of
history
where
we
started
how
we
got
where
we
are
in
the
cap.
A
Once
we
have
a
rough
idea,
because
when
I
was
looking
at
the
weight
command
last
time
with
eddie,
what
we've
noticed
is
the
weight
command
does
not
have
validate
nor
they're
complete,
and
when
I
was
working
on
the
locks
earlier
today,
I
was
actually
inclined
to
get
rid
of
of
the
complete
method
in
favor
of
the
two
options,
which
is
basically
translating
flags
to
particular
structure
options.
A
So
it
might
be
possible
that
this
pattern
will
be
the
new
pattern
and
we
will
have
to
figure
out
whether
the
rest
stays,
but
ron
will
be
there,
as
is
the
question,
is
how
much
of
two
options
and
the
validate
we
want
to
keep
because
oftentimes,
the
validate
also
has
valet
also
has
validate,
is
also
returning
the
flack
name
and
I'm
worried
that
that
means
that
it
still
leaves
the
validation
very
tightly,
coupled,
maybe
not
necessarily
sean
said,
probably
don't
need
complete
and
validate
if
we
move
through
options
yeah,
that's
that's
what
I'm
seeing
from
looking
at
least
at
the
locks
command,
but
since
I
don't
have
a
clear
picture,
the
way
is
the
only
example
that
we
currently
have.
A
C
That's
great:
do
you
think
people
should
wait
to
jump
in
until
the
additional
example
that
you're
talking
about
like
until
you've
got
that
worked
out?
So
it's
clear
what
the
pattern
is
or
can
they
are
there?
Other
simple
commands
that
could
be
worked
on
right
away.
A
I'm
I
I'm
pretty
sure
that
there
are
other
simple
commands
logs
and
get
are
the
happy
guns,
rather
I'm
pretty
sure
that
there
are
simpler,
which
is,
I
don't
know,
api
versions,
api
resources.
They
should
be
pretty
straightforward.
The
amount
of
flags
that
there
are
is
rather
small.
A
The
question
is
whether
they
well
api
versions.
I
remember
the
api
versions,
have
any
flags?
A
Well,
actually,
api
version
does
not
have
any
flags.
So
that's
another
question
whether
we
want
to
have
something
api
resources
does
have
a
a
group
version.
I
think
yeah.
It
has
a
lot
more
because
it
has
api
group
and
a
couple.
Others
name
space,
no
header,
so
api
resources
should
be
pretty
straightforward.
A
Api
versions
is
something
that
we
might
want
to
have
a
closer
look.
Whether
we
want.
A
What
we
want
to
do
with
those
that
don't
have
any
flags
how
to
treat
those
special
cases
but
yeah.
B
So
my
recommendation
would
be
I'm
gonna.
I
want
to
remove
the
help,
one
and
good
first
issue
for
now.
I
think
that
us,
you
know
more
core
folks
should
take
in
a
command
or
two
and
see
if
we
can
each
implement
one
and
get
a
feel
for
it
kind
of
like
a
beta
run,
and
then
we
can,
you
know
open
it
up
and
have
more
examples
to
point
to
what
are
your
thoughts
on
that.
A
Yeah,
that's
fair,
like
I
said
I'll,
be
tackling
the
vlogs
and
get
in
the
next
weeks.
That
could
that
that
hopefully,
will
give
a
better
overview.
A
B
But
we
got
rid
of
that
one
actually
yeah,
so
if
anyone
else
wants
to
pick
one
up
and
and
have
a
go
at,
it
feel
free
to
claim
it
on
here.
For
now
I.
C
I
was
working
on
drain
and
describe
recently
so
I
could
take
one
of
those
if
you
think
it's
suitable,
but
I
haven't
looked
at
the
flag.
So
it's
not
the
part
I
was.
I
was
changing.
Well,
I
was
adding
a
flag,
so
it
was
pretty
simple.
A
Give
it
a
try
and
see
if
you
struggle
with
it
just
ping
any
of
us,
and
you
know
we
can
either
set
up
a
quick
call
or
try
to
figure
it
out.
It's
fine
to
open
it
work
in
progress
pr,
and
then
we
can
discuss
that
as
well.
Great.
B
A
There
were
the
content
command.
We,
we
did
got
the
command
when
you
were
working
on
it.
Last
time,
yeah
we.
B
We
redid
it
to
the
the
other
way
we
were
looking
at
them:
cool,
okay,
so
funny
enough,
we
have
like
four
new
issues
that
were
open.
In
the
past
day,
I've
I've
been
staying,
I've
been
able
to
stay
on
top
of
this
mostly
and
I've
started
writing
up
a
handbook
that
we
can
like
hand
off
to
other
people
for,
like
a
triagi
type
role.
B
Part
of
that
at
the
end
of
this
I'd
like
to
save
like
five
or
ten
minutes
to
talk
through
triage
party
and
like
what
we
would
see
as
like
good
filter
options
for
this.
So
let's
just
walk
backwards
through
the
newest
ones.
Here.
A
Because
the
cubical
side
of
validation
wasn't
as
good
as
the
server
side,
so
it
would
be
good
to
know
whether
the
error
that
he's
seeing
or
she's
seeing
is
coming
from
the
server
or
client
and
based
on
that,
we
can
then
quickly
identify
where
the
problem
is
coming
from,
because
in
122,
thanks
to,
I
think
we
could
pull
up
his
name,
because
I
was
pinging
got
here,
the
lore
he
eliminated
the.
B
What
v
is
that
for
better
error
message?
Eight.
A
A
We
just
had
issues
validating
your
request,
but
it
is
possible
that
the
server
will
properly
validate
that
so
just
ignore
client-side
validation,
push
it
to
server
and
in
most
cases
that
solves
a
lot
of
problems
because,
like
I
said,
that's
the
downside
that
we
have,
but
this
should
be
fixed.
B
B
B
D
A
I
was
also
going
through
caps
a
lot
of
caps
over
the
past
couple
of
days,
and
I
left
a
comment
there
for
dog
that,
since
a
lot
of
folks
is
raising
significant
concerns
about
the
xdg
directory
specification,
I
was
actually
inclined
to
complete
the
cap
and
put
there
all
the
suggestions
and
concerns
that
people
are
all
are
saying
and
merge
it
as
a
rejected,
because
I
was,
I
checked
that
it
is
possible
for
us
to
have
a
rejected
cap
and
we
would
use
that
rejected
cap
as
a
as
a
piece
of
docs
for
future
generations
and
future
questions
where
people
will
be
asking.
A
Oh,
why
are
not
doing
this
and
we
would
point
them
have
a
look
at
this
rejected
cap.
We
we
looked
at
it
and
these
are
the
problems
that
we
have
it
that
we've
identified
and
based
on
that
input,
we
decided
that
we
are
not
going
to
pursue
this
so
that
that's
a
comment
that
I
left
for
doug.
We
could
probably
just
complete
that
one
and
then
make
sure
that
does
that
make
sense.
I
I'm
not
sure
if
I
haven't
seen
any
follow
up
from
stuff
based
on
that,
but
I.
A
I
think
that
the
majority
of
the
concerns
is
around
the
tooling
around
the
not
even
close,
but
second
level
and
third
party,
the
third
level
tooling.
That
is
built
on
the
assumptions
where
we
have
key
conflicts.
A
And
a
lot
of
them
are
just
assuming
by
hard
coding
knowing
the
past
history,
so
we
we
could
theoretically
introduce
environment
variables
and
whatnot,
but
at
the
end
of
the
day,
there's
already
this
many
people
are
using
it.
Probably
if
we
could
come
up
with
this
at
a
sooner
time
frame,
it
will
be
possible,
but
given
this
is
out
in
the
wild,
for
I
don't
know
like
five
six
years
by
now,
it's
the
spread
will
be
big
and
the
change
might
be
too
costly.
A
C
Yeah,
I
think
there
is
a
basically
mirror
this
issue
opened
on
the
customized
repo.
It
looks
really
familiar,
there's
a
problem
with
updating
gogaml
that
is
speculated
to
solve
this
and
other
issues,
but
it
also
contains
a
breaking
change
to
formatting.
So
that's
there's
an
open
pr
to
do
the
upgrade,
but
it
needs
more
tests
to
prove
that
it
actually
fixed
this
to
justify
the
major
change,
because
it
would
be
a
major
version
bump
on
the
cli.
B
B
A
A
That
seems
reasonable.
I'm
pretty
sure
that
this
was
addressed
for
sorry
different
yeah.
So
something
like
that
was
done
for
a
set
image,
so
it
seems
it
seems
reasonable
to
add
this
to
other
commands
and
it
could
be
a
rather
simple
thing
to
add.
So
I
would
leave
it
open
and
you
know
mark
it
as
a
good
first
time
issue.
A
B
B
D
Yeah,
I
I'm
waiting
this
one,
because
I
did
some
tests
and
didn't
get
this
error,
but
it's
we
need
information,
so
I
don't
know
if
we
have
like
a
deadline
for
for
needs.
Information
as
well.
B
B
Okay,
create
secret
tls.
I
think
you're
looking
at
this
one
as
well.
D
A
B
E
E
E
So,
in
a
nutshell,
what
what
is
clayton
looking
for
here,
the
api
groups
are
doesn't
exist.
Api
versions
does.
A
I
think
there
are
a
couple
issues
which
are
in
general
about
the
cube
cutoff
usage,
since
this
is
from
four
years
there
are.
It
requires
a
little
bit
of
time
to
go
through
into
every
single
item
verify
whether
we
have
it's
possible
that
we
are
addressing
at
least
half
of
the
things
that
he's
mentioning.
C
C
It
reminds
me
of
I,
I
used
to
maintain
a
diploid
tool,
and
one
of
the
things
I
wanted
to
know
was
what
resources
exist,
whether
or
not
they
were
namespaced,
and
what
group
was
the
most
current
group
for
them,
and
I
remember
that
code
being
difficult
to
write.
So
it's
probably
at
least
a
couple
years
ago.
It
was
still
true.
E
Cool
yeah,
so
we
have,
we
have
api
versions
and
then
it
looks
like
he
also
wanted
api
groups,
and
then
they
get
all
name.
Spaced
was
to
to
get
the
cluster
scope
versus
namespace
scoped.
E
Yeah,
just
I
was
just
curious
since
you
know
clayton's
one
of
the
original
guys.
Just
you
know
what
he
thinks
things
should
look
like.
B
I
wonder
anything
we
do
like
this
would
have
to
be
done
server
side
now,
right,
I
don't!
Even
we
didn't
even
do
that
much
server
side
when
you
open
this
issue.
E
A
A
Only
when
we,
when
you
try
to
get
something,
for
example,
get
something
that
we
don't
know.
We
will
invalidate
the
cache
and
read
it
again,
because
maybe
something
got
created
that
you
just
don't
know
so
all
the
information
about
all
namespace,
whether
it
is
or
not,
namespace,
whether
it's
a
crd
or
built-in,
all
those
information
are
the
they
explain
that
he's
talking
about.
A
We
fix
a
little
bit
of
explain
because
beforehand,
if
you
have
two
resources
with
a
register
for
two
different
api
groups,
because
openshift
had
that
problem,
we
have
one
particular
resource,
a
resource
name
which
was
registered
in
in
two
different
resources.
We
would
only
pick
one
and
there
was
no
option
for
you
to
pick
the
other.
That's
how
the
the
the
mapper
that
was
created
was
written
incorrectly,
so
that's
fixed.
Apparently,
you
can
actually.
E
I
have
a
quick
question
manche
sorry
to
interrupt,
because
I
just
ran
into
something
where
the
discovery
once
you
throw
a
you
know,
you
apply
a
crd
now,
your
rest
mapper
that
you're
using
is
out
of
sync
and
you
have
to
re
reset
it
and
that
reset
it.
It
was
really
I've
seen
really
ugly
code
where
it
had
to
like
downcast
to
the
deferred
discovery
in
order
to
to
call
the
reset
on
that,
so
that
it
would
reset
the
the
discovery
is.
Is
there
a
better
way
to
do
that,
or
is
that?
E
Do
you
know
what
I'm
talking
about?
I
know
I'm
going
down
into
the
weeds
here,
but.
A
So
basically,
what
we
do
is
there's
one
pr
that
I
have
which
is
improving
the
situation,
because
I
think
we
have
some
wrong
things
going
on.
I
can't
remember
where
the
exactly
are.
I
would
have
to
look
it
up,
but
basically,
yes,
you're
correct.
We
are
trying
to
match
a
particular
resource
against
unknown
mappings
that
we
read
from
the
server.
A
If
we
figure
out
that
we
don't
know
it,
we
will
invalidate
the
cache
and
read
everything
from
the
server
once
again,
and
only
then
we
can,
you
know,
tell
the
user
that
oh
it's
actually
something
that
you
don't
know,
but
I
have
a
pr
this
one,
but
I
96
771
for
a
shortcut.
A
Can
we
try?
I
can't
remember
what
the
original
idea
was,
but
I
it's
still
something
that
people
would
want
to
see
happening.
E
A
Fine,
I
mean
it
was
related
with
with
the
short
things
specifically
because
yeah
that
they're.
I
think
there
is
a
description.
B
The
last
thing
that
jumps
on
this
is
since
we're
trying
to
move
away
from
any
sort
of
get
all
in
the
first
place.
I
would
say
that
this
should
be
pushed
out
to
a
plug-in
right.
This
specific
functionality.
A
I
I
think
what
he
wants
to
do
here
is
not
necessarily
get
all
namespace
and
I'm
probably
trying
not
to
do
it,
but
he's
more
leading
towards
getting
all
the
resources
that
are
namespace,
but
I
I'm
not
envisioning
us
doing
something.
Crazy,
ass,
get
all
the
resources
that
are
namespace
everywhere.
Oh
that.
A
A
They're
very
hard
to
figure
out
what's
going
on
and
we
are
trying
not
to
have
them
hard
cut
it
like
it
used
to
be
with
that
keep
gotta,
but
there
are
questions
and
suggestions
about
having
the
client-side
aliases
for
stuff.
But
then
again,
if
you
want
to
go
with
something
like
that,
probably
all
you
need
to
have
is
a
fasciality
escort.
C
A
Right
come
on,
I
would
have
to
pick
up.
I
would
have
the
api
to
know
what
the
field
name
is
yeah.
There
is
something
like
that.
A
And
I
remember
adding
something
for
for
the
web
hook.
A
A
I
will
leave
it
open
for
now.
If
someone
is
interested
in
going
through
all
of
the
things
that
clinton
mentions
it's
fine
and
we
can
comment
on
it.
B
What's
this
one
auto
complete
okay,
so
that
brings
up.
We
have
to
go
through
mark's
pr,
because
that
thing's
been
open
so
long
and
he
put
a
ton
of
work
into.
B
It
I
remember
that
one
and
it's
it's:
it's
really.
I
started
playing
with
it
locally
and
it's
really
cool
what
it
gives
us
the
power
to
do.
We
can
define
custom
functionality
for
completion
inside
go
code,
so
we
don't
need
to
mess
around
with
any
janky
bash.
It
basically
pulls
out
all
of
the
bash.
That's
in
there.
B
A
I
I
have
it
on
my
list,
but
I'm
inclined
to
say
that
I'll
push
this
after
the
enhancement
freeze,
which
is
in
on
may
13th.
I
I
have
a
couple
texts
that
I
need
to
push
through
first
before
I'll
get
to
this,
but
yeah.
It
will
be
good
to
get
something
going
for
122.
With
this.
B
F
Yeah
I'll
take
a
look
at
this
too,
but
it
does.
It
actually
helps
us
too.
With
some
of
the
I
can't
remember,
who
was
that
that
was
saying
they
were
trying
to
do
completions
with
with
a
ton
of
resources,
and
it
was
like
timing
out
during
the
completions
and
things
so
yeah.
E
F
F
A
A
I
mean
eddie
and
brian
if
you
feel
comfortable
and
you'll
be
okay
with
it.
Just
don't
wait
for
me
I'll
get
to
this
pr
and
we
can
address
any
comments
after
it
gets
merged.
There's
nothing
stopping
you
if
you
will
be.
If
both
of
you
are
fine
with
this,
I
know
it's
big,
just
get
it
merged
and
we
can
cover
whatever
will
be
needed
afterwards.
B
In
yeah
there's
a
couple
to
do's
we
have
to
address.
I
think
what
I
think,
what
I'll
do
is
I'll,
actually
I'll,
send
an
email
to
kdev
and
I'll
ask
people
to
take
a
look
at
it
and
play
with
it
and
maybe
do
a
build
and
see
if
they
find
any
bugs.
E
F
I
was
just
not
to
say
I
think
I
might
have
been
responsible
for
that
too,
because
I
I
was
concerned
about
getting
having
two
different
ways
of
doing
completions
and
that's
that
if
we
didn't
get
rid
of
all
the
bash
stuff
that
it
would
live
forever
and
we
would
have
multiple
ways
but
yeah.
I
know
it's
kind
of
a.
It
is
a
big
thing,
though,.
A
Yeah,
that
would
be
my
word
as
well
that
we
would
actually
end
up
in
a
health-baked
solution.
To
maybe
you
know,
maybe
it's
good.
Maybe
it
will
be
just
better
to
merge
it
as
soon
as
possible,
and
then
we
have
sufficient
amount
till,
if
I
remember
correctly
july,
8th
or
something
like
that
is
the
feature
for
the
code
phrase
for
122
that's
more
than
two
months,
so
that
will
have
sufficient
coverage
and
e2es
and
and
and
people
working
with,
with
cube
on
a
daily
basis.
B
Yeah
he's
got
a
few
fixed
v's
and
to
do's
in
here
so,
okay,
brian
and
I
can
take
a
look
at
it.
We
can
sink
offline
on
that
brian.
B
A
B
So
triage
party
is
just
that
tool
that
I
always
go
through.
It's
on
my
list
to
move
it
to
kubernetes
infrastructure,
so
it'll
lose
my
cool
vanity
domain
name,
but
that's
okay.
B
Right
now,
the
config
I'm
using
is
the
stock
default
config
out
of
here
and
it's
extremely
customizable
and
flexible.
So
it's
all
just
yaml
filters
that
we
can
apply
different
rules
to
all
of
these
sections.
Here's
the
release,
engineering
one
and
they
do
actually
a
really
good
job
of
keeping
it
clean.
Apparently,
so
it's
not
a
good
demo.
B
I
think
some
other
folks
have
moved
to
triage
party
as
well
lori's,
also
a
big
driving
force
behind
this,
and
I
know
she
helped
other
folks
do
it.
So,
basically,
what
I'd
like
to
get
together
is
what
kinds
of
sections
would
be
valuable
for
people.
B
C
C
A
That's
one
thing
and
the
other
I'm
even
inclined
to
have
a
separate
section
for
the
upvotes
with
regards
to
the
email
that
I
sent
earlier
today,
but
the
other
thing
is,
we
can
probably
just
have
a
generic
six
cli
and
then,
after
one
or
two
bugs
crops
we'll
see
how
many
noise
we
have
versus
how
much
we
want
to
have,
and
then
we
can
figure
out
what
we
just
need
to
eliminate.
A
I've
noticed
that
api
machinery
is
very
active
in
getting
rid
of
the
sega
cam
machinery
from
those
issues
where
they
are
not
they're,
so
federico's,
so
good
at
that
yeah
I
know
so
maybe
we
should
be
doing
exactly
the
same
thing.
Oh
quickly,
look,
oh!
It's
not
60
like
just
drop
65
from
there
and
it's
gone.
Maybe
initial
one
or
two
meetings.
It'll
take
us
a
little
bit
longer
time
to
go
through
those
and
remove
the
actual
noise
and
over
time,
we'll
just
reach
a
point
where
this
won't
be
a
noise
anymore.
A
So
probably
just
start
simple,
and
then
we
can
always
iterate
because
trying
to
come
up
with
some
fancy.
Initially,
I
just
I
don't
know,
I'm
worried
that
we
will
end
up
in
some
weird
funky
situation
that
we
will
be
actually
missing
stuff.
B
A
Probably
initial,
three
sections
that
I
can
think
of
is
the
most
popular
issues
per
the
email
so
that
we
know
we
want
to
push
forward
and
then
two
sections
for
books,
and
I
don't
know
something
that
just
needs
to
follow
up
and
the
question
is
whether
we
want
to
throw
together
cube
cuddle
and
kubernetes
issues
together
or
we
want
to
have
separate
them
out.
I
I'm
fine
with
both
approaches
as
well.
B
B
C
Yeah,
I
think
this
would
be
very
helpful
for
customize.
That's
just
my
take
on
it,
but
customize
has
a
really
high
issue
volume
as
well.
So
it
looks
very
useful.
B
D
Yes,
okay,
this
is
how
they
break
it
out.
Sorry,
go
ahead!
I
I
didn't
know
that
you
wanted
to
move
that
to
the
triple
a
cluster.
I
was
going
to
to
ask
you
if
you
want
some
help
before
you're
showing
this.
So
if
you
need
some
help
to
to
move
to
to
closer
to
infrastructure
closer,
I
can
help
you
with
that.
D
Yeah
he's
the
chair
of
the
of
of
infra,
but
I
participate
in
that
working
group
as
well,
so
it's
gonna
be
pretty
straightforward
to
open
these
and
ask
them
to
apply
this
to
the
cluster
we
can
just
discuss.
If
you
want
to
send
me
the
current
config
map,
and
then
we
can
actually,
this
is
not.
This
is
not
auto
automated
yet,
and
I've
been
discussing
discussing
with
them
also
like.
Why
can't
we
use
some
teraform
or
something
like
that
as
well?
D
D
I
yeah
I
will.
I
will
probably
write
in
the
next
two
months.
Another
cap
about
the
the
localization
thing
I
was
discussing
with
eddie
and
we
turned
to
iron
as
well,
but
I
am
with
a
pretty
low
bandwidth
to
do
all
of
those
things.
So
this
is,
I
don't
think
it's
a
it's
a
high
priority,
but
yeah
just
to
make
sure
that
everyone
is
on
the
same
page.
D
I
can
show
you
what
I
am
thinking
about,
but
that's
not
like
a
closed
proposal
because
it's
yeah
it
might
get
a
lot
of
changes
as
well
into
the
code
and
some
email
files
with
the
translations,
because
translating
nowadays
is
like
I
I've
been
reviewing
the
portuguese
one.
It's
really
a
pain.
It's.
A
There's
this
topic
that
since
we've
mentioned
about
the
top
level
of
the
most
voted
issues
and
addie
and
myself
and
laurie,
were
talking
about
the
major
teams
for
for
122,
with
enhancement
freeze
in
two
weeks
from
tomorrow.
Basically
for
the
next
meeting,
we
would
like
to
bring
the
most
important
issues
to
the
slack
within
the
next
couple
of
days,
so
people
can
commend
the
most
important
ones
and
we
have
the
caps
pretty
much
identified.
A
A
I
mean
cap
wise.
The
exit
codes
is
definitely
something
that
we
will
be
pushing
forward,
but
it's
not
something
that
we
want
to
do
for
the
122..
A
Even
if
we
would
start
some
implementation,
I
would
probably
wouldn't
want
to
have
that
yet
maybe
we
can
keep
it
in
a
separate
branch
for
now
and
the
cap
problem.
A
E
A
A
A
The
only
important
thing
will
be
the
date
for,
if
you
want
to
bump
this
to
beta,
you
will
want
to
update
the
cap
and
get
it
merged
within
the
next
two
weeks
and
obviously
the
smarter,
the
better.
A
I
need
to
think
talyn
about
it.
I
have
a
as
long
list
as
my
to-do
topic
that
I
need
to
spun
clayton
about,
but.
A
Okay,
we
need
to
have
a
look
at
it.
The
options
will
be
probably
something
that
we
could
start
similar
to
the
exit
codes.
A
A
Theoretically,
there
is
already
a
preference.
If
I
remember
correctly,
the
the
conflict
field,
the
config
type,
has
something
like
preferences:
we've
never
used
it.
Maybe
it's
about
time.
So
one
thing
is
looking
at
the
mechanics
of
implementation
and
how
much
this
would
affect
the
actual
work.
Probably
ansi.
Colors
is
a
good
starting
point.
B
A
Oh
I'm
curious.
Let's
find
a
final
question.
That's
the
goal.
Completion
will
solve
the
third
and
fourth
most
voted
issue
about
powershell
and
fish
completion.
A
B
A
That
would
probably
be
okay,
we'll
try
to
sing
offline
and
put
this
as
threats,
and
we
will
continue
talking
about
it
next
week
on
wednesday.
Okay,
thank
you
very
much.
All.