►
From YouTube: Kubernetes SIG CLI 20230215 Kustomize 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
Hello
and
welcome
again
to
today's
60
live
meeting,
which
is
customized
bug
scrub.
We
are
going
to
go
through
issues
that
need
triage
on
a
customized
repo
and
hopefully
find
some
issues
that
folks
might
be
interested
in
fixing
before
we
get
started.
Can
would
I
like
to
make
some
time
for
anybody
who
has
brought
an
issue
today
that
they
would
like
us
to
discuss?
A
B
Yeah
I've
got
a
PR
on
4928
customized
fails
to
apply
overlay
if
Kian
Bay
CM
was
numeric
and
missing.
We.
A
A
So
this
is
a
key
and
base.
Yaml
is
an
American
missing,
interesting.
A
It
seems
related
to
the
issue
where
numericans
in
general.
Don't
work
super
well,
there's
a
sort
of
an
umbrella
issue
somewhere
about
that.
A
B
I
had
I
had
an
initial
PR
that
just
made
everything
a
string
and
realized
that
wasn't
good.
So
there's
a
second
one
where
the
intent
is
to
honor
the
whatever
the
style
is
of
the
incoming
patch,
so
that
doesn't
guarantee
you'll
get
something
that
works
if
you've
got
a
misalignment,
but
hopefully
that
honors,
the
intent.
A
B
A
Yeah
I'll
have
to
re-familiarize
myself
with
this
code
and
give
it
a
proper
look.
If
that's
okay,
the
approach
seems
reasonable
on
the
face
of
it,
but
whenever
I
go
into
the
weeds
of
cupid
sorry,
K
yaml
I
have
to
sort
of
dig
around
to
remind
myself
of
all
the
edge
cases
and
and
considerations
no
I
I
get
it
I'm,
not
I'm,
not
asking
to
have
it
resolved
now.
B
Just
figured
I'd
highlight
it
and
see
if
there
were
any
questions
off
the
off
the
cuff.
If
not,
we
can
just
move
on.
A
Anyone
else
have
something
for
us
to
look
at
upfront.
A
All
right
so
I
I
guess
we'll
just
take
a
look
at
the
triage
board,
so
at
the
top
here,
there's
someone
assigned
so
I
wonder
if
this
is
just
a
missing
tag
on
this
one
or
perhaps
this
was
a
previous
bug
scrub
where
Natasha
was
going
to
respond.
Does
anyone
remember
I,
wasn't
able
to
to
make
the
last
bug
scrub
whether
this
was
something
that
Natasha
was
already
talking
about.
A
So
they're
talking
about
the
Json
patch
operation,
it
doesn't
hurt
to
read
through
it
here.
Can
everyone
see
this
by
the
way
or
should
I
increase
the
font
size
all.
E
A
A
So
what
occurs
to
me
is
that,
like
we
don't
actually
dictate
what
is
possible
in
Json
formats
patches,
we
sometimes
get
requests
to
handle
a
different
operation
or
behave
differently
than
we
do
today,
with
with
an
operation
that
exists
and
there's
not
actually
anything
that
we
can
do
about
those,
because
this
is
actually
following
us
back
we're
using
a
third
party
library
for
it,
because
it's
not
it's
not
actually
our
feature
and
I
believe
Jason
patch
has
a
test
operation.
A
Let's
see,
add,
remove
replace
copy
move
test,
so
yeah
test
test
that
the
specified
value
is
set
in
the
document.
If
the
test
fails,
then
the
patch
as
a
whole
should
not
apply.
A
Does
anyone
on
the
call
have
experience
with
Jason
patch
in
customize.
A
A
So
I
haven't
done
this
before,
but
I
have
like
a
sort
of
a
scratch
directory
that
just
has
random
customized
stuff
in
it
that
I
repopulate
with
the
test
cases
when
I'm
trying
to
figure
out
if
something
works
or
not.
So
what
I
would
do
in
this
case
is
we
have
our
patches,
and
this
is
looking
like
they're
using
the
patches
field
rather
than
patches,
Json,
60.,
902
or
so.
What
we
can
do
is.
F
A
Not
and
we're
going
to
need
a
resource
to
patch,
so
I
have
some
random
resources
in
here
from
previous
experiment.
A
Unknown
field
up:
okay,
that's
an
authorized
syntax.
Let's
double
check
the
docs
here
so
go
to
reference
customization
patches.
Here
we
go.
Okay,
maybe
maybe
wrong,
and
it
is
attaches
just
in
1602,
syntax
yeah.
It
is.
C
A
A
E
A
E
A
A
So
what
is
the
issue
filer
concerned
about?
It's
not.
That
was
not
a
different
result
right.
That
was
an
error
when
we
didn't
have
the
the
labels
part
of
the
path.
A
A
No,
it
successfully
operates
whether
or
not
that
specific
key
existed,
which
makes
sense,
and
then,
if
we
get
rid
of
this
again,
we
have
an
error
because
the
path
doesn't
exist
so
I'm
not
sure.
Actually,
after
that,
what
what
they
mean
by
it's
having
a
different
result
like
there's,
no
result,
if
the
key
didn't
exist,
does
anyone
have
an
idea
about
what
they
might
be
talking
about.
C
C
A
So
that
that
tells
me
the
test
operation
probably
doesn't
help
them
because
they're
looking
for
a
specific
path,
unless
they
wanted
to
see,
if
life
does,
we
do
labels
Foo
got
you
our
and
if
we
rebuild
that
I'd
say
actually.
E
F
F
Or
something
they
can
do
and
inspect
and
and
do
some
sort
of
check
to
say,
hey
I'm,
expecting
that
I
shouldn't
have
a
bunch
of
changes
and
in
these
certain
changes
I
was
expecting
or
something
that
or
at
least
failing
I
guess
but
I
don't
know
if
anything's
right,
but
maybe
some
way
for
them
to
inspect
what
you're
doing
I
think
it
starts
with
like.
Can
you
give
us
a
better
Repro?
So
that's
my
thoughts.
A
Yeah
thanks
I
agree
with
that,
because,
especially
because,
like
I,
just
I
just
tried
to
do
a
Repro
with
it
has
a
different
result
based
on
the
patch
that
they
supplied,
which
is
an
ad
operation
and
I,
don't
get
a
different
result.
I
do
get
an
error
which
sounded
like
what
they
wanted.
So
it's
a
good
good
call.
So,
let's
see
triage
needs
information.
B
I
took
a
moment
to
go
back
to
the
previous
meeting,
where
this
was
discussed.
The
discussion
boiled
down
to
basically
If
This
Were
pursued
for
this.
B
It
sets
a
precedent
for
basically
all
Transformers
and
then
a
conversation
about
potentially
using
logging
as
a
mechanism
for
tracking,
and
then,
if
the
you
know,
if
the
logging
can
be
then
used
to
do
the
validation
that
they're
looking
for.
If
the
log
messages
are
clear,.
A
A
Yeah,
having
verbose
logging
is
probably
pretty
useful
in
general,
it's
not
something
that's
wired
into
customized
at
all.
Right
now,.
A
Interesting
another
another
thing
that
comes
to
mind
when
you
say
that
is
that
they're
saying
what
they
have
to
do
is
run
a
build
with
without
it
and
compare
the
output
by
doing
diff,
and
that
is
actually
like
what
customize
wants
you
to
do,
because
the
guarantee
that
we
want
to
provide
is
that
and
I
guess:
that's
why
this
is
really
the
trying
to
reproduce
my
attention
here
is
that
their
claim
that
we
have
a
different
result
of
path
existed
or
not
like
that,
does
sound
like
a
bad
experience.
A
We
do.
Generally
speaking,
they
are
modifying
underlying
configuration
to
create
that
scenario.
But,
generally
speaking,
we
want
our
build
out
to
bit
output
to
be
as
stable
as
possible,
and
if
we,
my
intuitive
sense
of
it,
is
if
that
path
doesn't
exist,
then
that
should
be
an
error.
A
So
it
was
kind
of
surprising
we
don't
like
I
was
saying
earlier
to
control
what
Jason
patch
does
as
a
spec,
but
customized
does,
generally
speaking,
want
things
like
operations
that
can't
be
carried
out
to
result
in
errors
because
you're
declaring
an
intent
to
do
a
thing,
and
then
we
can't
do
the
thing
right.
A
A
A
Attempt
to
apply
a
match
on
a
path
that
does
not
exist.
I
get
an
error.
C
E
A
This
one
surprise:
that's
triaged
this
one
I'm
familiar
with.
We
have
a
linked,
PR
open.
It's
an
example.
That's
in
our
repo,
that
is
part
of
a
tutorial,
and
the
tutorial
tells
you
to
create
some
files,
but
somebody
is
going
in
and
trying
to
just
build
it,
as
is
so.
The
question
on
the
link
PR
is
well.
Do
we
say
that's
on
purpose?
You
are
supposed
to
create
the
files.
A
This
is
a
tutorial
or
you
say
like
well
be
helpful
if
it
just
also
worked
and
served
as
an
example
and
people
can
follow
the
tutorial
to
add
modify
files
as
they
go,
but
in
any
case
like
I,
think
we
can
accept
it,
because
there
are
also
some
errors
in
the
files
that
the
pr
is
fixing
so
I'm
just
gonna
go
ahead.
A
A
F
A
A
Oh
there
we
go
that's
the
difference.
They
actually
are
providing
a
duplicate
there
so
that
they
are
able
to
Target
exactly
that
information
and
just
copied
over.
That's
what's
going
on,
I
think
all
right,
that's
fine,
like
so
they're
saying
I
could
duplicate
this
information
in
two
different
places,
one
of
which
is
suitable
for
extraction
into
the
into
the
other
resource
as
is,
and
then
that
would
work,
but
I
would
rather
not
duplicate
that
information.
A
You
can
also
write
a
custom
Transformer,
which
is
very
true,
so
Natasha
is
the
architect
of
The
Replacements
feature,
so
I
usually
like
to
get
her
opinion
on
on
those
but
real,
quick.
Let's
take
a
look
at
the
current
meaning
of
the
delimiter
field.
We
have
some
formatting
issues
in
the
docs
here.
A
A
F
A
Yeah
and
there
there
is
at
least
to
some
extent
here
like
it.
What
they're
saying
the
definition
of
used
to
split
join
the
field.
I
was
just
introducing
a
semantic
for
Target
that
isn't
supported
today.
I
agree.
It
sounds
reasonable
to
me,
I
think
I'll.
Take
it
under
consideration
and
get
Natasha's
opinion
on
it.
She's
the
lead
for
replacements.
A
E
A
E
A
Looks
really
incomplete
actually
and
then
our
resources.
A
Than
it
has
to
be,
presumably
that
doesn't
affect
it
all
right.
It's
useful
to
have
a
I,
always
find
it
difficult
with
these
multi
multi-base
samples
when
they're
just
written
out
in
line
to
like
visualize
the
structure.
So
I
always
appreciate
when,
when
people
provide
a
repo
right,
so
they're
saying
that
they
are
expecting
the
config
map
to
get
a
suffix,
and
in
fact
it's
not,
that
it
isn't
suffixed
I
misinterpreted
that
it's
that
the
ref
doesn't
end
up
getting
updated
for
some
reason.
A
A
I
wonder
if
we
can
simplify
this
case
a
little
bit.
It's
like
the
ref
isn't
even
in
the
deployment
to
begin
with,
it
is
being
introduced
in
this
patch.
A
So
I
definitely
want
to
try
to
reproduce
first
with
5.0
before
going
any
further,
and
it
might
take
a
little
bit
of
time
to
set
the
structure
and
try
to
simplify
it
and
see
what
can
still
reproduce
their
issue,
assuming
that
it's
reproducible
in
5.0.
So
it
seems
like
some
of
these
factors.
A
Be
relevant
and
it
would
be
good
to
know
what
the
minimal
case
actually
is
is
somebody
would
somebody
be
interested
in
trying
to
reproduce
this
and,
if
not,
would
folks
find
it
more
useful
for
me
to
be
trying
to
do
it
live
here
or
to
move
on
to
another
bug.
A
I
actually
have
just
a
related
question
for
the
group,
since
we're
we're
12
minutes
out
here.
I
feel
like
these
bug
scrub
meetings.
It's
a
lot
of
me
talking
and
sort
of
fumbling
through
through
the
issues
and
I
was
wondering
if
there's
a
way
to
make
this
more
interesting
or
interactive
for
folks
like
today,
as
trying
to
show
my
my
screen
with
with
some
of
the
bug
reproduction
process,
but
that's
still
not
very
interactive.
A
So
a
question
for
the
group:
if
we
perhaps
didn't
record
it
to
sort
of
reduce
the
pressure,
would
folks
who
attend
to
be
interested
in
maybe
leading
individual
issues
say
if,
if
we
we
have,
a
the
triage
party
tool
can
actually
split
up
the
issues
among
the
group
and
if
we
had,
if
attendees
were
interested,
we
could
say
split
up
at
the
beginning
of
the
meeting,
and
everyone
gets
an
issue
and
has
five
minutes
ten
minutes
to
dig
into
it.
A
And
then
we
could
rotate
around
presenting
And
discussing
each
issues,
as
everyone
would
have
a
chance
to
sort
of
be
more
involved.
But
if
folks
think
that's
too
intimidating
they're,
like
oh,
really
hate.
That
idea,
then
we
can
keep
it,
as
is
so
I
just
question
for
the
group.
A
What
is
useful
and
interesting
to
you
and
about
this
meeting
and
related
to
my
question
about
right
now?
Would
you
like
me
to
continue
with
more
issues,
discussion
or
dive
into
this
one
for
the
reproduction
case.
E
I
think
splitting
up
issues
seems
fun
and
it'd
be
a
good
way
for
people
to
be
forced
assigned
a
little
bit
because
once
you've
reproduced,
if
it
turns
out
to
be
a
bug,
you're
like
halfway
through
the
work.
So
it
might
be
good
for
getting
some
momentum.
F
I'm,
okay,
either
way,
I
I
think
it's
interesting
enough.
Listening
to
you
talk
so,
but
I
think
it's
helpful
lost
expectations
for
this
meeting.
Yes,
I
just
figured
that
we
go
through
and
triage
what
we
can
and
sign
off
the
things
that
we
that
need
more
time.
F
B
Well,
chime
in
just
to
say,
I'm
I'm,
not
Savvy
enough
with
customized
to
feel
like
I
really
can
be
overly
useful,
in
my
hope,
with
these
meetings,
right
now
is
to
sit
and
listen
mostly
so
that
I
can
be
useful
and
hopefully
not
too
long.
A
Yeah
and
that's
a
great
reason
to
come
so
for
you
would
you
would
you
still
come
if
you
had
a
a
role
in
like
trying
to
understand
a
particular
issue?
I
guess
I
should
clarify,
like
the
expectation
would
be
that
you
know
the
answer
and
you
triage
it
yourself.
A
It's
more
like
you
take
the
time
to
do
that,
like
initial
pass
on
the
issue,
understanding
what
the
person's
trying
to
say,
seeing
if
their
samples
are
actually
valid
and
then
sort
of
just
bring
that
that
to
the
group
when
it's
your
turn
afterwards
to
say,
like
I,
did
reproduce
I
didn't
reproduce.
This
is
the
part
I
don't
understand
and
like,
as
you
saw
today,
I
sometimes
I
forget
stuff
about
how
customized
works.
A
I
forget
what
the
syntax
is
and
I
fumble
around
too
so
I
I
think
there
is
definitely
a
place
for
even
folks
who
aren't
familiar
to
be
very
valuable
in
in
just
trying
to
understand
what
people
are
are
trying
to
tell
us.
B
Yeah
I
mean
I,
think
that
makes
good
sense
I,
where
I
would
be
hesitant,
would
be
sort
of
you
know,
driving
the
discussion.
If
the
you
know,
role
is
really
excuse
me,
take
it
offline
track
down
as
much
as
you
can
with
the
acknowledgment
that
that
may
not
be
very
far
I'm
perfectly
comfortable
with
that.
A
Yeah
I
think
what
I
was
thinking
was
that
it
would
still
be
online.
It
would
be
maybe
at
the
beginning
of
the
meeting.
We
would
split
up
the
trash
party
in
the
trash
party
tool
and
each
person
would
get
one
issue
to
look
into
and
then
we
would
just
kind
of
be
silent
on
the
call
for
the
first
five
ten
minutes
or
something
like
that.
A
I
mean
you
can
ask
questions
as
you
sort
of
read
through
the
issue
and
then
like
go
around
and
when
it's
your
turn,
you
would
just
explain
the
issue.
You
don't
necessarily
need
to
lead
the
conversation
in
a
sense
of
like
getting
the
result
more
like
just
explain
it
to
the
group
like
what's
going
on
here,
and
what
did
you
find
out
about
it
in
your
five
ten
minutes?
Whatever
we
decided
was,
was
reasonable.
A
It
and
and
then
like
I,
will
still
write
the
triage
comment
if
you're
not
comfortable
or
if
we
don't
receive
a
conclusion
and
I
can
also
do
something
in
the
middle
like.
If
we
try
this
next
time
say
we
can
try,
not
recording
it
to
make
it
so
that's
less
intimidating
and
then
take
like
volunteers
to
do
that
that
role.
So
it's
not!
Everyone
like
you
can
still
show
up
and
not
have
to
do
that.
A
You
could
just
listen
in
but
folks
who
want
to
be
like
who
are
looking
to
get
more
involved
and
customize.
Can
can
step
up
to
take
that
more
active
role
in
in
this.
In
this
meeting
and
like
you
know
so,
there'll
be
a
few
of
us
that
that
do
that
and
it's
a
way
to
to
sort
of
take
that
next
step
in
and
try
to
figure
things
out.
A
And
I
see
you
nodding,
would
you
would
you
be
a
volunteer
if,
if
we
did
that.
D
I,
don't
I
don't
have
any
strong
opinions,
but
I
feel
like
if
we're.
If
we're
open
to
the
idea,
we
can
experiment
right
and
take
it
from
there.
Yeah
and
I.
Think
I
think
it
was
really
nice
that
Ed
started
with
a
PR
question,
because
I
can
imagine
if
you
don't
have
ready
access
to
the
customized
maintainers.
Sometimes
PR
reviews
there's
a
lot
of
back
and
forth
and
it
might
be
nice
for
for
people
to
bring
up
their
questions
and
just
have
a
quick
live
discussion
here.
A
Yeah
for
the
pr
review
thing
that
actually
makes
me
think
you
know
if
everyone
is,
if
somebody
brought
a
PR,
that
they
wanted
some
feedback
on
and
what
the
group
was
doing
was
having
individuals
go
and
look
into
specific
issues
during
that
time.
A
If
there
are
maintainers
on
the
call,
I
was
like
yeah,
whoever
is
a
maintainer
on.
The
call
would
also
be
able
to
use
that
time
to
sort
of
dig
into
whatever
question
the
person
had
without
doing
this
sort
of
just
hot
take
live
just
have
that
longer
time
to
to
give
quality
feedback
still
within
the
scope
of
the
meeting
good
work.
A
All
right
thanks
everyone
for,
for
your
opinion,
us
on
that,
and
if
you
have
any
further
thoughts
about
whether
or
not
to
do
that
experiment
or
if
you
have
another
idea
on
how
to
make
this
meeting
more
useful
to
the
project
or
to
you
as
a
contributor,
please
share
it
in
the
customized
Channel.
A
A
So
coming
back
to
this
issue,
we're
trying
to
see
now
if
this
issue
here
might
be
related
to
this
other
one,
which
has
been
accepted
from
last
November.
So,
let's
see
what
was
this
one
about.
F
I'm
not
sure
if
it
is
related,
actually
it
just.
It
was
amazing,
but
I
think
it's
a
different
name
of
issue.
Actually
now
that
I
looked
at
it
again,.
A
C
A
That's
carry
on
KML,
it
sounds
like
so
that
one
is
more
Niche
not
about
customized,
build
itself.
A
This
one
is
related
to
the
deprecation
that
we
did
in
customize
5.0,
so
this
this
is
pretty
interesting
to
look
at
I.
Think
we'd
like
to
when
we
cut
a
a
big
release,
particular
attention
to
issues
that
had
to
do
with
that
release,
so
we
can
give
them
attention
of
their
regressions.
This
one
I
suspect
on
the
face
of
it
here.
I'm
gonna
paste
this
one,
because
I
think
we
should
triage
it
this
one
on
the
face
of
it.
A
I
suspect,
is
not
like
a
new
problem
to
customize
5.0,
so
much
as
a
difference
between
an
old
syntax
and
a
new
syntax,
but
our
conversion
tooling
isn't
handling,
but
let's
actually
look
at
it
properly.
A
A
The
problem
is,
if
I'm
understanding
correctly,
not
that
they
that
their
sample
can't
work,
but
that
the
new
syntax,
as
opposed
to
the
old,
let's
take
a
look
here
so
say
they
had
passion,
strategic
merch.
That
was
strategic,
merge
rate.
A
Secretive
passionate,
yes
strategic
version
so
before
you
could
have
multiple
patches
in
the
same
file,
it
doesn't
even
say
so,
but
you
could,
because
the
targets
were
implicit
and
in
the
new
syntax
the
target
is
or
can
be
at
least
explicit,
so
having
multiple
patches
per
filed
doesn't
necessarily
make
so
much
sense.
A
I
wonder
if
we
actually
stay
in
the
docs
that
you
can
only
have
one
per
file.
Let's.
C
A
So
I
want
to
confirm
one
thing
before
I
tell
them
this:
let's
just
copy
their
sample
in
here,
so
we
have
customization.yaml.
A
A
C
A
So
you
get
the
same
errors,
they
do.
That
makes
sense,
and
then
what
I
want
to
try
is
having
a
patch
two
just
to
make
sure
I
give
them
good
advice
here.
A
So
if
we
take
this-
and
we
put
that
in
there
and
then
we
duplicate
this
and
we
do
patch
two
then
does
it
build
yes,
so
the
problem
isn't
actually
with
the
patch
Transformer.
The
problem
was
with
our
instructions
for
the
migration
and
with
customized
edit
fix
customized.
A
Edit
fix
should
detect
that
there
was
a
file
that
had
multiple
multiple
entries
in
multiple
documents
and
should
split
them
up
and
write
multiple
path,
patch
entries,
so
this
one
I
think
is
is
a
pretty
interesting
starter
issue
for
one,
because
any
change
to
customize
edit
or
any
basically
any
command
other
than
customized
build
is
lower
pressure.
A
Those
commands
are
used
for
out
of
band
customized.
Build,
has
tons
of
constraints
on
it
that
we
have
to
be
really
careful
about,
but
the
other
commands
are
smaller
and
more
approachable.
So
this
one
in
particular,
yeah
there's
a
fairly
straightforward
thing.
You
need
to
change
the
command
to
C,
to
see
if
the
file
that's
being
targeted,
has
multiple
documents
in
it
and
if
it
does
split
it
into
multiple
files
and
write
the
appropriate
just
path
file.
One
file,
two,
five,
three:
five,
four
into
the
transformed
customization.
C
A
E
Yeah
I
tried
to
reproduce
it
and
didn't
get
super
far.
I
also
comb
through,
like
all
the
changes
in
the
release
to
see
if
anything
stood
out
that
might
have
done
this,
but
I
haven't
I,
haven't
done
it
since
you
and
I
talked
about
it
like
next
steps
for
reproducing.
C
A
Yeah,
this
just
needs
to
be
tried
on
the
Tacoma
line,
essentially
redirect
standard
error,
and
this
should
not
appear
if
you've
redirected,
student
or
somewhere
else
like
this
should
this
should
be.
We
need
to
prove
that
this
is
actually
still
going
to
Standard
air,
hopefully
yeah
since
you're
already
looking
into
this
Caitlyn
I'll
leave
it
to
you
we're
already
paying
attention
to
this.
One
I'd
be
very
surprised
if
that
changed,
but
if
it
did,
we
need
to
fix
it.
So.
A
That
is
a
script
this
one
or
example.
Soon
that
is
asking
for
docs.
A
That
looks
old,
oh,
this
is
coming
up
because
of
the
new
triage
retriage
rules
that
were
just
implemented
in
the
community.
A
Okay,
so
I
have
a
couple
ones
that
I
need
to
take
away
to
look
into,
but
I
don't
think
we
have
time
to
actually
go
through
another
one
in
full.
At
this
point,.
D
Katrina
I
was
actually
looking
at
one
before
I
just
wanted
to
get
your
opinion.
I
already
wrote
up
your
response.
If
we
don't
have
time
to
look
at
another
big
one,
could
we
look
at
this
one
absolutely.
E
A
A
I,
don't
see
a
problem
with
ad
replacement
just
to
add
the
path
the
complication
I
see
with
Replacements.
Is
that
it's
not
a
simple
like
there's
a
lot
of
configuration
that
goes
into
their
placement,
Transformer
the.
A
If
they've
done
that
all
already
in
a
path,
then
that's
cool,
like
that's
very
simple,
for
us
to
add,
with
an
edit
command
what
I
so
I
would
be
inclined
to
accept
this
with
that
caveat
that
that's
what
it
will
do
if
they,
if
they're
responding,
that,
that
is
useful,
what
I
would
be
hesitant
to
accept
is
that
ad
replacement
command
that
needs
a
ton
of
flags
to
be
able
to
do
all
the
complicated
stuff
that
you
can
do
within
the
replacement
Transformer,
because
I
think
that
would
not
be
worthwhile
to
maintain
right.
D
A
Yep
that
that
works
for
me
do
you
want
to
accept
it.
E
A
All
right,
I
thought
we
are
at
time.
Thank
you.
Everyone
for
coming
today
to
this
customized
bug,
scrub,
I,
hope
it
was
useful
to
you
and
please
give
me
your
feedback
by
DM
or
in
the
customized
Channel
about
how
we
can
make
this
meeting
better
for
everyone
and
for
your
goals
going
forward.