►
From YouTube: WG API Expression Bi-Weekly Meeting for 20201013
Description
WG API Expression Bi-Weekly Meeting for 20201013
A
Welcome
to
the
api
expression
meeting
it's
the
13th
of
october
and
we
have
something
on
our
agenda
antoine.
Do
you
wanna
talk
about
it.
B
I
was
hoping
jordan
and
stefan
could
join
the
meeting,
but
jordan
is
in
another
meeting
and
so
stefan's
probably
not
coming.
We,
I
have
a
meeting
later
today
with
the
two
of
them
to
discuss
that.
B
A
Go
into
the
details
was
it
about
the
annotation
in
the
types
to
generate
defaults
or.
B
Yeah
yeah
yeah,
yes,
and
we
don't
know
exactly
what
to
do
for
the
nine
panel
fields,
because
built-in
types
can
differentiate
between
an
unset
field
a
and
a
zero
value
field.
When
it's
another
bino
and
clds
can,
and
they
don't
behave
the
same
way
and
from
the
open
api.
We
don't
know
if
it's
a
9.0
or
not.
B
B
A
A
Well,
I,
while
I
have
it
here
from
from
last
music,
we
discussed
how
to
continue
with
telescoping
and
I
I
started
a
document
and
I'm
still
trying
to
to
figure
out
what
I
did
over
the
last
half
year
and
what
we
decided
was
bad
and
why
we
got
stuck
on
those
I'm
pulling
it
apart.
I'm
doing
a
document
right
now
that
I
explained
the
problem
again
and
I'm
I'm
pulling
apart
the
I
think
four
pr's.
A
By
now
to
find
the
problems
we
had
put
them
in
there
and
then
we
can
discuss
them.
C
A
Yeah,
I
I
think
everybody
saw
it.
I
created
this
document
and
there's
not
much
in
it
right
now,
just
a
few
things
we
we
discussed,
I
don't
know
if
we
should
go
over
it
now
and
add
stuff
we
discussed
or
if
we
should
make
a
dedicated
meeting
for
it,
whatever
you
prefer
like
how
we
or
we
just
update
it
as
synchronously
everybody
looks
at
it.
I
think
you
had
some
talks
on
the
side
about
it.
B
I
mean
yeah,
I
don't
know
exactly
what
is
the
timeline
to
get
these
ready?
I
I'm
sure
we
could
all
start
by
doing
some
of
these
asynchronously
and
then
maybe
probably
have
a
meeting
to
discuss
about
the
items
but
yeah.
We
should
probably
elaborate
on
these.
A
Yeah
I
mean
the
most
stuff.
Currently
is
our
day-to-day
stuff
that
we
have
in
the
work
item
see.
I
wouldn't
move
that
out
of
here,
I
think,
would
only
split
stuff
like
we
have.
We.
E
I
would
just
have
everybody:
could
do
a
review
of
the
of
the
cap
and
start
to
kind
of
fill
it
out.
Yeah
I'll,
do
a
pass.
A
A
A
A
B
Phone
call
to
take,
but
yes,
it's
still
blocked
on
the
okay,
so
I'll
make
a
status
for
the
defaulting.
So
we
have
these
discussions
going
with
the
cap.
B
I
started
the
implementation,
so
I
have
a
pr
out
which
I'm
going
to
add
to
this
thing,
I
have
a
pl
how
to
add
the
default
marker
or
field
in
the
open
api
if
it's
present
in
the
built-in
time.
So
that's
good,
so
we
can
already
populate
the
the
open
api
with
the
default.
B
I
have
another
pr
that
removes
the
null
values
when
they
have
to
be
removed.
Oh
it
actually
doesn't
remove
them.
It
defaults
them
when
they
have
a
default
and
the
pi
is
mostly
ready.
I
need
to
add
more
tests,
but
it's
mostly
ready.
B
I
am
currently
looking
at
jenny's
pr
to
revive
it
and
update
it
so
that
we
can
look
at
the
default
fields
from
the
open
api
and
use
them
when
we,
when
they're
on
an
associative
key
and
what's
left
to
do
after
that
is
merely
the
part
where
we
actually
use
the
defaults
and
generate
the
code
to
apply
these
defaults
for
built-in
types.
A
Yeah
yeah,
I
I
mentioned
that
I'm
pulling
what
I
know
apart
tracking
sub
resources,
andrea,
what
happened.
C
Hey,
yes,
I
did
some
progress
there
right
now.
I'm
gonna
try
to
solve
this
with
the
default
field
manager
and
using
the
field
manager
of
the
main
resource
to
try
and
change
the
match
field,
but
yeah
there
might
be
other
solution
possible
like
this
is
just
one
of
them
so
happy
to
get
start
a
conversation
there.
B
Thank
you
so
much
for
doing
that.
Andrea
is.
We
need
to
look
at
how
is
the
field
manager
world
and
how
is
the
subresource
created
so
that
we
can
wire
the
proper
field
manager
in
there,
because
I
don't
think
we'll
ever
need
to
use
the
scale
the
scale
manager,
because
it's
pointless,
we
don't
actually
operate
on
the
scale
objects.
B
So
I
suspect
what
you
need
is
the
main
resources
manager
at
this
point.
So
let's
look
at
the
wiring
and
see
if
we
can
create
the
resource
with
the
pop-up
manager,
I'll
try
and
take
a
look
or
ask
maybe
julian
if
he
has
some
time
to
take
a
look.
B
Yes,
yeah,
I
think
it's
very
useful
that
you
post
these
code
links
and
we
can
look
at
it
and
comment
on
these.
I
think
it's
working
very
well
and
you
don't
have
to
make
the
change
if
you
suspect
you
want
to
make
a
specific
change,
feel
free
to
just
send
a
link
and
say
I'm
planning
to
change
that.
So
I
think
this
is
where
we
need
to
make
the
change.
We
can
also
comment
on
that
and
I
think
it's
very
useful.
Thank
you.
C
Okay,
cool
I'm
going
to
update
you
all
in
the
next
days.
A
A
F
B
B
If
you
know
nothing
or
if
you
don't
have
some
time
to
look
at
it,
I
don't
think
it's
super
hard,
but
yeah.
It's
it's
somewhat
involved.
A
A
We
scaled
for
121,
I
don't
know
if,
if
you've
been
working
on
student
here.
A
A
E
Yeah,
oh,
hey,
so
I've
got
a
working
prototype
of
the
generated
client
in
kidsgates
and
I
think
when
I
I
wanted
to
do
a
couple
other
detailing
on
it,
and
then
I
I
think
it's
probably
makes
sense
for
me
to
go
sync
up
with
go
to
like
a
coup
builder
or
controller
runtime
meeting
and
start
to
show
it
to
people.
I
also
think
I
should
just
take
one
of
the
small
controllers
and
just
go
over
to
show
that
happening.
E
So
I'm
going
to
keep
working
on
that
you'll
probably
hear
more
from
me
about.
It
was
kind
of
heads
down
just
keeping
it
working
initially,
but
I
think
I'll
transition
over
and
start
making
a
little
more
noise.
I
actually
need
to
start
writing
a
cap
at
some
point.
We're
obviously
not
going
to
do
it
for
this
cycle,
but
for
the
next
cycle.
I
think
I'd
like
to
come
out.
B
F
Let's
see
if
we
find
someone
have
you
have
you
shared
a
link
about
it.
E
Already,
I
don't
know,
let
me
let
me
send
and
what
I've
got
now
is
just
my
own
branch.
Let
me
let
me
get
that
organized
and
I'll
send
something
out
I'll
put
it
in
slack
or
in
an
email
or
maybe
both.
A
Cool,
I
think
nothing's
happening
here
right
now
and
it's
not
good.
A
B
E
There
was
another
issue
that
was
open,
trying
to
make
some
kind
of
change
that
could
have
picked
this
up,
so
I
made
sure
that
they
knew
that
this
was
available
now
yeah,
but
that's
kind
of
I'm
going
to
kind
of
let
them
on
their
own.
At
this
point
they
weren't
very
responsive
to
stuff
on
their
pr.
So
if
they
drop
their
pr,
then.
A
Okay,
no
change
apply,
still
persists.
We
discussed
that
on
select
a
little.
B
B
A
Oh
yeah,
that's
that's
the
one
look
and
we
talked
a
bit
on
slack.
A
A
That's
changing,
but.
B
F
I
don't
know
why,
but
you
you
did
test
it
with
your
fix
to
do
the
time.
B
Later
no,
but
my
fix
is
different.
My
fix
is
that
I
mean
possibly
it
would
fix
that,
but
it's
kind
of
odd,
but
my
fix
would
change
if
you
have
a
quantity,
for
example,
which
is
canonicalized
in
a
different
form,
so
the
600
would
become
600
m
or
something,
and
it
would
become
a
string
because
of
the
quantity
becoming
a
string
when
it's
canonicalized,
and
here
it
it
doesn't
even
do
that
the
number
stays
a
number
and
when
you
get
the
results
from
the
api
server,
you
have
the
exact
same
thing.
A
B
A
Yeah,
the
yeah,
the
generation
increases
for
a
custom
resource,
if
anything
but
metadata
changes
and
that
gets
again
status
if
it's
defined
as
a
sub
resource
and
not
just
as
a
field
named
status.
It
also
doesn't
shouldn't
trigger
a
generation
increment.
So
something
is.
A
Yeah,
I'm
I'm
also
very
interested
I'll,
dig
it
into
it.
If
I
have
time,
I'm
still
on
holiday
technically
so
but
yeah,
that's
I
I
looked
at
it.
I
read
through
it.
It's
really
weird.
A
Okay,
quantity
fails
type,
checking,
isn't
that
there's
nothing
much.
G
Is
yeah
go
ahead?
Sully!
Is
this
strictly
true,
according
to
the
code
of
so
like
this
came
up
in
controller
runtime
or
on
controller
tools
rather,
and
it
was
pointed
out
that
quantity
is
not
in
fact,
an
entry
string.
Quantity
is
an
inter
string
or
float
like
code
for
parsing
quantity
accepts
float
as
well,
and
then
normalizes
the
float
before
it
gets
parsed,
which
is
the
only
way
that
it
can
be
valid
asterisk.
G
Ins
yeah,
I
mean
that's
like
that's
fine
this.
This
is
not.
This
is
insufficient
like
this.
This
rejects
things
that
are
currently
accepted,
or
at
least
that's
that's
what
was
pointed
out
to
us
in
in
controller
tools
and
according
to
my
interpretation
of
the
code,
it
seems
like
that's
the
case.
B
G
Yeah
not
that
I
wouldn't
rather
improve
this,
and
I
think
this
is
a
good,
potentially
good
step
in
the
right
direction,
but
like
either.
We
need
to
make
a
decision
on
the
floatness
of
quantity
or,
like
figure
out
a
way,
maybe
like
add
a
new
type
to
to
that
to
structural
schema,
that's
into
our
string
or
float
or
or
make
quantity,
a
valid
format
option
in
and
of
itself
or
something.
B
Yeah,
so
this
is
just
changing
the
format
and
not
the
type.
What
is
the
actual
type
for
these?
Do
we
know?
Can
we
do?
I
remember?
Oh
it
says
string.
B
B
Because
I
I
wrote
the
client-side
validation
from
foreign
and
I've
done
my
own
sauce.
I
mean
I
interpreted
the
stand
out
the
way
I
wanted
at
this
point
and
in
very
wrong
ways,
and
so
I
that's.
There
are
two
things
right:
the
clients
that
actually
look
at
the
open
api
and
use
the
right
validation
and
there
is
cubecode
who
does
it
something.
B
E
Yeah
yeah,
and
maybe
what
we
should
be
doing
is
at
the
very
least,
even
if
it's
a
step
forward
is
like
put
a
comment
in
the
description
about
what
we
know
for
posterity.
So
if
somebody
else
looks
it
up
and
then
make
sure
that's
a
good
point
and
probably.
A
A
Yeah,
we
still
have
help
needed
there,
tombstoning
joe
any
anything
new
on
that.
E
I
think
we
deleted
this
one.
I
I
would
just
put
a
marker
here
that
we've
shelved
this
until.
E
Postgame,
oh
when
I,
when
I
update
the
the
ga
cap
I'll
make
sure
that
we
know
this
properly,
so
that
if
anybody
asks
that,
we
have
a
clear
explanation
of
what
we're
doing.