►
From YouTube: 2022-04-29-Node.js Node-API Team meeting
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
A
I
think
this
is
the
right
one.
Hopefully,
and
so
first
of
all,
let's
see
drive
towards
full
coverage.
I
don't
think
we
have
jack
who
usually
jumps
in
there,
but
so
I
don't
know
if
anybody
anybody
up
else
has
updates
in
terms
of
like
adding
coverage
on
some
tests
or
anything
like
that.
A
Next,
the
stale
issues
in
the
node
add-on
api
to
discuss.
So,
let's
see
what
we
got
there.
C
A
D
So
thankfully,
yes,
we
kind
of
discussed
last
time
and
what
what
I
found
that
previously,
let's
kind
of
repeat
what
we
discussed
last
time,
a
little
bit
so
they're
using
a
num
they're
using
object,
template
and
we
don't
have
any
node
api
around.
This
object
template,
but
we
discussed
last
time
so
thankfully
javascript
has
proxy
objects
and
proxy
object.
D
Pretty
much
does
almost
the
same,
maybe
not
exactly,
but
always
the
same
and
the
last
time
I
promised
that
I
going
to
implement
some
tests,
so
I
worked
on
it
and
I
didn't
completely
com
completely
complete,
but
I
do
have
something
working
end-to-end
for
very
simple
case
like
you
have
some
simple
object
and
this
simple
object
has
add
method
and
it
works.
If
you
look
on
test.js,
it's
a
looks
a
little
bit
simple.
D
D
Yeah
so
think
at
this
point
we
can
create
this
object
and
we
can
add
there
and
if
you
look
in
the
in
the
c
plus
plus
code.
Yes,
I
have
to
write
some
infrastructure.
But
thankfully
speaking,
we
can
have
this
as
a
part
of
our
c
plus
plus
header
files
and.
D
C
D
A
D
Yes,
it's
a
test
only
so
the
idea
that
how
we
can
achieve
exactly
the
same
goals
without
adding
new
node
api
node
apis
can
we
just
show
how
to
use
it,
how
to
use
normal
proxy
javascript
object
from
node
api
to
achieve
the
same
results.
A
A
So
your
plan
is
like
okay
get,
you
know,
add
this
test
in.
It
adds
good,
it
adds
test
coverage,
so
that's
good
for
core
and
then
possibly
add
something
to
node
add-on
api,
that
in
terms
of
c
plus
code
or
just
you
know,
point
to
this
as
an
example.
D
So
what
I
would
like
to
do
I'd
like
to
have
first,
this
just
plain,
vanilla,
node,
api
test.
Then
I
would
like
to
add
another
test
in
c
plus,
plus
node
api
and
probably
as
a
part
of
this
node
api.
We
can,
after
the
decide
if
we
want
to
add
additional
functionality
of
to
simplify
transition
from
none
to
node
api
right.
A
Okay,
so
it
sounds
like
there's
good
progress
on
that.
That's
good,
we'll
flip
back
to
here.
This
is.
A
Yeah,
no,
that's
so
I
think,
that's
all
that
we
have
on
the
stale
issue.
So
that's
good
tracking
issues
for
modules
that
are
reported.
A
C
D
A
A
Right
like
in
in
node
core,
we
actually
run
some
tests
with
debug
enabled
like
the
debug
option.
So
there's
like
a
build
option
where
you
can
build
for
debug
and
I
think
there's
some
cases
where
you
know
that
that
catches,
that
does
some
like
has
extra
asserts
and
can
catch
some
extra
extra
things.
So
it
would
be
nice
to
actually
have
some
of
our
testing
run
with
that
under
okay.
A
A
G
No,
I
plan
I
I
I
will
do
in
the
weekend
so
tomorrow
or
sunday,
but
and
okay,
the
the
new
release
will
be
four
four
zero
and
I
need
to
remove
the
support
for
12
on
on
on
readme
and
the
first
lts
that
we
support
will
be
14.
Is
it
right.
A
B
Right
we
did
say
that,
and
there
was
also
the
discussion
about
the
it
was
that
old,
pull
request
that
we
had
for
migrating.
The
regular
thread
safe
function
to
the
to
be
backed
by
the
type
safe
type
threat,
safe
function,
and
we
couldn't
land
that
because
of
the
gcc
compiler
requirements,
and
we
had
discussed
that
if
we
wanted
to
wait
for
it
to
be
a
semver
major
change
to
node,
add
on
api
so
that
it
would
not
break
people's
compilation.
A
Vaguely
is
it
that,
but
but
is
it
that
we
were
waiting
for
it
to
like
the
an
older
version
of
node
that
used
that
version
of
the
compiler
to
drop
off?
Yes
right,
okay,
so
that
that
would
make
sense
that,
like
you
know,
we
make
it
v5
and
then
in
v5,
if
it
doesn't
support
12
and
14
supports
the
newer
compiler.
Then
we're
okay
with.
A
A
I
think
everything's
good
good.
Okay,
so,
like
I
think,
it'd
be
good.
If
we
cut
five
and
then
we
can
then
I'd
say
the
next
thing
we
should
definitely
land.
Is
this
format
all
code
so
that
we
don't
land
other
things
to
to
conflict
with
that
which
is
quite
possible
so
yeah?
You
know
once
five
goes
out
I'll
look
forward
or
somebody
else's
time.
We
can
look
and
then
land
the
format,
I'll
call
code.
One
is
a
priority.
A
Sounds
good,
okay,
any
other
issues
here
that
we
should
talk
about.
A
A
B
There
was
a
couple
of
prs
that
vladimir
had
on
core
that
we
tried
to.
We
just
pointed
out
that
they
existed.
B
I
have
not
had
any
time
to
go
through
them,
but
there
was
one
point
that
I
did
want
to
discuss
and
I
don't
remember
what
the
pr
was,
but
it
was
the
one
where
you
were
adding
the
the
calling
standard
as
a
as
a
as
a
macro.
B
Yes,
that
one
I
just
had
a,
I
didn't,
make
any
comment
on
it,
but
maybe
we
can
discuss
if
you
go
to
the
files
changed,
you
introduced
a
macro
for
the
all
uppercase
napi
status.
B
I
killed
it.
I
killed
it
in
the
latest
generation.
Sorry,
okay,
okay,
yeah,
because
my
concern
that
I
had
was,
I
I
think
we
should
separate
the
a
a
macro
for
the
return
value
from
the
calling
convention,
because
I
could
see
maybe
in
like
somebody
would
have
tried
to
use
a
variable
to
declare
a
variable
using
the
type
as
the
whole
upper
score
uppercase
and
api
status,
and
I
think
there
would
have
been
a
little
bit
of
confusion
with
that.
D
Yeah
I
kind
of
looked
at
that
last
this
week
and
I
also
kind
of
had
a
bad
taste,
especially
because
this
microstarter
the
same
marker
was
applied
not
only
to
header
file
but
also
implementation
file,
and
I
didn't
like
it
so.
Thankfully
I
killed
this
background.
The
only
micro
new
macro
kind
of
is
there
is
an
api
c
deco
which
either
cdec
or
none
nothing
right.
So,
let's
change.
B
Yes,
all
right,
so
then
that
since
that
resolves
my
concerns,
I
don't
have
any
other
concerns
with
with
this
pr.
D
So
ideas
that
michael
the
problem
was,
we
kind
of
claimed
that
node
api
is
api,
safe
api.
But
if
you
on
windows,
x86
platform,
if
you
didn't
explicitly
specify
what
your
calling
conventions
are,
it
may
cause
issues
because
some
dlls
so
executables
they
may
change
default
coding
conventions
like,
for
example,
in
microsoft.
D
We
often
by
default
using
standard
coding
conventions
and
while
node
api
assumes
it
should
be
cdec
and-
and
we
had
I
had
personally,
we
had
to
work
on
some
bugs
in
our
code
to
address
this
issue,
and
I
decided
like
so
rather
than
me-
kind
of
patching
all
these
files
locally
for
ourselves.
What
if
we
do
it
in
node
api
itself,
sps5
is
called
configurations.
D
Right
so
I
think
default
is
cdeco.
So,
okay.
F
D
Didn't
see
the
echo,
but
I
do
I
had
added
this
discussion
point.
Ideally
it
should
be
a
stood
call
because
it's
my
it's
producing
much
more
compact
code
on
the
calling
side.
D
Right
so
yes,
if
we
change
it
to
stood
co,
it
may
affect
affect
some
cool
sites.
But
if
we're
doing
justice
vehicle
we're,
just
simply
stating
explicitly
what
already
been
assumed
implicitly.
A
D
Like
in
my
case,
the
issue
was
like
that
I
I
put
cdf
on
some
functions,
but
I
forgot
to
put
in
function
pointers
and
because
of
this
we
changing
calling
conventions
on
our
site.
It
was.
B
D
For
review,
because
right
now
I
kind
of
simplified
scope
down
the
whole
thing.
Just
one
specific
marker,
it's
very
kind
of
scoped
down
right.
G
G
So
in
these
podcasts.
G
In
nappy,
in
the
no
dabbing
symbol,
four
nappy
version
was
marked
as
not
the
version
no
right.
Yeah
and
dire
adams
changed
with
nappy
version
eight,
but
I
explained
that
nothing
no
dappy
sample
four
is
an
experimental
feature.
So
it's
not
part
of
any
node
api
version
right.
G
Okay,
so
and
I
checked
okay,
I
tried
to
explain
that
and
I
checked
all
the
function
that
we
are
under
the
experimental
flag
and
I
I
found
some
some
errors
because,
okay
in
some
of
this
function,
we
don't
mark
the
the
function
as
experimental
so
yeah.
G
I
I
tried
to
to
to
explain
this
and
she
asked
us
to
explain
in
some
part
of
the
how
we
increment,
you
know
the
version
of
of
right
and
what
is
nappy
rational,
represent
right,
so
yeah.
I
think
that
we
can
do
this
in
another
pr,
but
I
won't
ask
you
your
thinking
about
this.
A
Yeah,
I'm
just
wondering
if
we
have
somewhere
that
we
have
I'm
just
wondering
if
there's
anywhere,
we've
documented,
like
here's.
B
B
A
A
G
Do
you
think
that
we
need
to
update
this
document
or
the
ndpi
node
api
documentation.
A
I
think
it
would
be
good
to
have
it
this
in
core
like
unless
there's
like
the
it
might
also
be
good
in
the
documentation,
though,
to
explain
to
people
like
this
is
we
I
think
I
could
see
like
something
in
this
doc
to
explain
to
collaborators.
A
A
E
A
G
Yeah,
because
we
have
some
explanation
for
when
all
the
node
api
was
experimental
right
and
but
we
don't,
we
don't
have
the
maybe
the
right
explanation
for
all
the
new
future
that
we
add
as
experimental
in
node
api,
okay,
yeah.
Okay,
if
you
want
I
I
can
try
to
to
do
this.
G
Okay-
and
I
you
can
comment,
maybe
on
the
paracas
these
or
maybe
I
I
comment
later.
G
Yeah
yeah.
G
Yeah
okay,
but
for
me
in
okay
for
the
the
purpose
of
the
of
the
the
the
pr
is
to
fix
these
these.
This.
G
Error
and
it's
okay,
but
I
iphoned
other
errors
right
who
could
I
ask
to
her
to
fix
or
I
we
need
to
to
create
a
new
pr.
I
I
would
say.
A
D
To
let
me
ask
you
about
this
specific
change
because
it
says
on
online
below
it's
still
experimental
and
here
we
say
nappy
version
8.
I've
seen
like
all
experimental
they
don't
they
typically
don't
don't
have
any
version
at
all.
B
A
B
Yeah,
if
you
go
to
the
conversations
quickly,
I
think
that's
what
he
pointed
out
and
then
the
last
one
is
where
he
looked
at
all
of
the
ones
that
need
to
be
need
to
be
fixed
and
basically
says
how
it
needs
to
look
okay,
this
one
here
where
it
it's
where
it
just
has
the
added
and
then
the
stability.
G
A
D
Quite
so
difficult
to
land
with,
and
I
get
all
property
names
fixed,
yeah.
A
I
know
just
very
stubborn.
D
A
A
D
D
And
yes,
also
a
couple
of
outstanding
prs
like
I,
I
want
team
to
look
at
one
of
them.
Remember
we
saw
finalizer
stuff,
so
I
think
we
were
saying
that
I
still
gonna
work
on
this
finalizers,
it's
more
like
I
I'm
kind
of
if
you,
if
somebody
wants
to
look
and
see
if
they
see
any
issues
with
this
approach,
I'm
still
working
on
it.
C
C
A
A
A
E
Is
there
an
issue
or
a
pr?
It
was
pr.
C
E
D
Now
we
can
find
blue
request
number.
E
A
D
So
this
one
should
be
ready
to
go,
I
updated
documentation
and
tests
and
stuff,
so
it
includes
as
soon
as
somebody
signs
off
on
it.
I
mean
I
know,
but
I
really
want
to
have
no.
E
A
A
A
If
not,
I
guess,
thanks
for
everybody,
for
taking
the
time
this
week
and
thanks
nick
for
taking
care
of
that
next
release
and
we'll
see
everybody
next
next
week.