►
From YouTube: 2021-02-05-Node.js N-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
Yes,
okay,
so
first
one
is
rename
primary
matched
branch.
To
main,
I
think
the
last
thing
that
that
was
left
was
updating
the
references
and
I
think
jim
yeah.
You
submitted
a
pr,
and
I
landed
that
yesterday,
so.
A
A
A
A
A
D
A
A
C
C
C
C
D
A
That'd
be
great
yep.
Thank
you
very
much,
so
I
think
anything
else
on
that
one
before
we
move
on.
C
There's
a
there's,
no,
I
nope.
I
was
confused
on
a
different
issue.
A
Okay,
okay,
so
that's
great
thanks,
jim
for
picking
that
up.
Let's
move
on
to
the
next
one
then,
which
is
renaming
an
api
to
something
more
appropriate
and
descriptive.
A
So
there
was,
I
know,
gabriel,
has
started
making
some
changes
and
stuff.
We
did
have
a
short
discussion
at
the
tsc
meeting.
There
was
some
concern
over
the
node.
Api
is
the
name,
but
I
think
the
key
thing
we
agreed
there
was
that
we're
going
to
try
we're
going
to
work
on
changing
the
the
the
like
the
docs
and
the
sort
of
surrounding
stuff,
but
not
necessarily
change
the
symbols
themselves
or
the
pat.
A
You
know
if
there's
paths
that
have
an
api
in
them
and
and
the
thinking
along
the
lines
which
I
was
kind
of
you
know
raising
was
if
we
clarify
enough
other
places
that
napi
is
node
api,
you
know
and
given
that
nappy,
the
association
suggested
is
not
as
strong
as
some
of
the
others.
That
might
be
good
enough
and-
and
nobody
disagreed
with
that
in
in
terms
of
the
technical
steering
committee
there.
So
we
figured
we'd
make
those
changes
first
and
then
see
if
anybody
felt
we
needed
to
go
further.
C
Okay,
that
that's
a
relief,
because
I
had
been
looking
at
some
of
the
tooling
I
had
done,
and
the
that
string
and
a
api
as
a
symbol
is
like
pervasive
and
right.
It
would
be
a
lot
of
work.
A
A
C
A
A
A
A
C
Yeah,
so,
while
I'm
well
while
I
well,
I've
got
the
examples
open
should
I
should
I
look
at
renaming
the
the
text
in
there
michael.
E
E
A
If
somebody
comes
in
and
starts
a
conversation
on
one
versus
the
other,
but
yeah
like
if
you
know
and
and
what
I
had
in
mind-
was
in
the
text
like.
Obviously,
if
the
symbols
aren't
changing,
we
can't
change
the
references
there.
But
in
the
text
where
we
refer
to
like
an
api,
we
can
just
change
that
to
node
api
right.
A
A
C
C
So,
michael
before
we
talk
about
this
I'll,
just
give
some
background
on
my
thinking
here.
So
there's
like
an
entire
universe
of
documentation
and
posts
and
blogs,
and
so
forth
that
are
like
the
wider
community
and
so
like
we
have
what
we've
referred
to
as
the
nipi
resource,
for
example,
which
which
I
I
will
rename
yeah
do
we
want
to
okay?
The
first
question
is:
should
we
mention
that
this
was
previously
known
as
napi,
so
that
when
people
are
first
coming
into
it,
they
they
they
realize?
B
C
For,
for
example,
I
I
will
update
like
the
tooling
that
I've
done,
there's
document
documentation
on
the
readme
that
so
I
should
at
least
update
the
documentation
renaming
it
from
n
api
to
node
api,
but
I
feel
it
would
be
important
to
put
a
node
in
there.
That
said,
this
was
previous.
This
is
the
package
previously
known
as
an
api.
A
C
C
Okay,
maybe
maybe
it's
not
necessary
to
refer
back
then
in
that
when
people
read
through,
they
start
seeing
the
symbols.
They'll
see
the
napi,
so
they
then
they
might.
B
B
C
Okay,
so
so
really
we're
not
gonna
we're
not
gonna.
Do
outreach
to
the
wider
community.
A
A
C
A
D
D
A
Okay:
okay,
that's
good
anything
more
on
that
one
before
we
move
over
move
on.
C
C
C
Yeah,
I
liked
your
edits
on
that,
michael
that
okay,
those
are
really
good
now.
My
next.
My
next
step
is
to
make
the
submission,
and
so
typically
the
way
it
works.
Is
I
put
it
in
and
then
I
can.
I
can
mention
that
there
will
be
co-presenters,
so
I
thought
it
would
be
good
to
have
an
outline
or
work
on
an
outline
and
then
see
if
there
are
team
members
who
would
like
to
volunteer
for
particular
sections,
then
when
I
know
who's
going
to
be.
D
C
C
A
That
is
a
good
question
like
I
don't
know
who's,
it's
kind
of
comes
like
it's.
I,
like
your
like
your
approach
of
whoever
volunteered
for
sections
right.
C
F
C
A
It's
better,
if
I'm
not
on
there
in
terms
of
who's
here
is
there.
You
know
people
saying
hey
like
to
be
on
the
submission.
A
A
A
F
A
A
Would
be
a
good,
like
you
know,
short
short
intro,
just
as
the
context
in
terms
of
like
hey,
if
you
know
how
to
do
native
add-ons
and
it
can
kind
of
be
end.
The
way
that's
recommended
now
is
is
node
api
right
and
you
got.
A
F
A
C
D
A
Cover
so
I
think
the
hardest
ones
are
the
building
and
the
multi-threaded
ones,
which
you
two
are
well
positioned
to
cover
and
the
others.
I
think
the
rest
of
us
could
help
yeah
get
a
couple
slides
that
you
know
either
either
we
can
have
other
people
jump
in
to
cover
those
or
even
you
know,
help
help
you
to
get
comfortable.
So
I
don't
be
too
bad.
C
Okay,
the
the
this
mission
is
due
the
15th,
so
we
will
have
one
more
team
meeting
between
now
and
the
15th.
A
C
A
This
running
node
thread.
I
know
we
had.
A
E
E
A
Okay,
we'll
just
flip
to
that
in
one
second,
just
going
to
quickly
look,
we
see
the
the
upper
trend
is
still
progressing
this
little
dip
here
is,
inter
I
mean
that
was
like
across
all
modules.
So
I
don't
know
quite
what
went
on
there,
but
it's
good
to
see
the
upward
trend
is
still
going
yeah
and
actually
that
yeah
sorry
go
ahead.
A
A
Okay,
so
let's
flip
back,
there
was
one
in
here
this
one
which
is
related
to
that.
So
I
think
we
should
so
legend
cast
will
just
come
back
to
yours
in
in
one
sec.
A
A
A
E
C
F
A
D
F
Okay
right
so
listen
to
the
cast,
which
issue
was
it
that
you
thought
we
should
do?
Is.
A
E
E
Yeah,
it
only
happens
on
finalization
and
occasionally,
but
it's
crashing
for
the
double
free
of
via
example,
rough
base.
So
it
only
happens
if
the
let's
say
not
p,
wrapped
object
or
an
appearance.
Reference
object,
pin
gc
after
the
add-on
instance
data
or
let's
say
the
nappy
emv
of
the
module
have
been
destroyed
so
right.
The
case
is
very
obscure
to
reproduce,
but
I
managed
to
make
the
nikkei
reproduce
to
force
the
program
to
crash
in
the
case
and
it's
all
valid
to
use
in
with
no
non-api.
A
E
A
Okay,
so
it's
then
it's
at
on
shutdown.
E
A
A
A
That's
good
because
there
have
been
some
intermittent
failures
in
the
regular
ci.
Like
you
know,
kevin
kevin
commented
on
one
change
that
he
made
recently.
There
was
in
the
lint
in
the
github
actions.
There
was
a
failure
and
it
was
kind
of
a
crash
after
the
tests
were
run,
so
it
could
actually
be
related
to
this.
E
Yeah,
I
have
noticed
some
early
crash,
and
so
I'm
I
think
that
might
be
really
it's
hard
to
track
down,
since
this
crash
is
not
immediately
happens
and
right.
E
Have
luckily
to
find
out
the
reproduce
right.
D
A
A
That's
causing
problems
related
to
like
nappy,
ref
and
nappy
ffi,
which
are
some
modules
that
use
node,
api
or
yeah,
but
I'm
not
sure,
there's
anything
we're
going
to
be
able
to
do
about
it,
but
I'm
just
trying
to
understand
a
little
bit
better.
What's
going
on
there,
it
has
to
do
with
how
v8
handles
buffers
and
it
seems
like
in
the
past.
A
You
could
have
two
buffers
pointing
to
the
same
like
memory
ad
like
physical
memory,
but
today,
if
you
do
that
it
throws
in
a
search,
saying:
hey,
wait:
a
sec,
you
can't
have
two
backing
stores
pointing
to
the
same
thing,
and
then
that
has
some
follow-ons,
because
it's
almost
like
it's
it's.
You
know
if,
if
something
uses
that
memory
but
then
gets
freed,
but
the
gc
or
hasn't
figured
out
that
it's
freed
yet
and
then
you
do
another
allocation
and
get
the
same
address.
A
It
can
run
into
problems,
so
that's
I
don't
understand
it
well
enough
to
really
fully
explain
it,
but
that's
what
I'm
looking
at,
because
it
sounds
like
quite
a
few
people
are
or
more
than
a
few
people
are
having
the
challenge
with
that
and
and
in
fact
one
of
our
internal
teams
is
having
a
challenge.
So
I'm
take
trying
to
understand
a
little
bit
better.
What's
going
on
on
that
front.
A
Otherwise,
that's
that's.
On
the
plane,
node
side
on
the
node
add-on,
api
side.
B
E
A
That,
like
I,
I
think
it
may
be
that,
like
we've
already
stopped
using
it,
and
it's
just
that
the
gc
so
like
the
memory
has
actually
been
freed,
but
the
gc
just
hasn't
figured
that
out
yet
kind
of
thing.
It
sounded
to
me
it
sounded
to
me
like
it
had
to
do
with,
possibly
with
the
what
callbacks
we
have
in
terms
of
doing
finalization
and
that
kind
of
stuff.
A
A
D
D
A
D
A
F
A
I
think
yeah
that
one's
basically
ready
to
land
whenever
I
get
around
to
doing
or
somebody
gets
around
to
doing
the
next
set
of
landing.
We
can
just
land
that
one
yep,
so
thanks
for
diving
into
that,
otherwise
don't
see
too
many
new
ones,
any
other
ones
that
people
want
to
sort
of
call
out
that
we
should
discuss.
D
A
Okay,
so
I
think
that's
this
one,
sorry
right!
This
is
the
one
I
just
created,
so
I'm
gonna
move
that
to
the
beginning.
I
think
that's
what
we
had
on
the
agenda
to
talk
about
anything
else.
We
should
talk
about
before
we
close
out
for
today.
C
A
Okay:
okay!
Well,
if
not
thanks
for
everybody's
time,
and
thanks
and
we'll
talk
to
everybody
next
next
week,.