►
From YouTube: Node.js Tooling Group 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
All
right,
I
think
we're
live
yeah,
hey
everyone!
This
is
the
tooling
meeting
for
what
is
the
date
january
21st.
A
It's
just
a
couple
of
us
here
today,
so
probably
keep
it
kind
of
short.
Let's
see,
oh
actually
so
first
thing
on
the
agenda
here.
Actually,
is
these
test
failures
in
the
rmdir
recursive.
A
A
So
yeah,
that's,
let's
sort
it
out.
What
else
do
we
have
here?
A
Recursive
reader,
we
had
a
bunch
of
people
on
the
call
last
time,
including
ethan
barwood
from
microsoft.
I
think
he's
vercell.
Now
everyone
just
probably
just
assume
everyone
works
at
purcell.
Now.
A
Anyways,
it
looks
like
he
put
up
a
pr
to
implement
the
recursive
reader
functionality
in
in
node
core.
So
that's
good.
There's
some
progress
there.
I
don't
really
think
I've
had
a
chance
to
look
at
it.
Yet
I
don't
know
if
anyone
has
any
comments
on
that.
B
A
And
then
I
alert
end
users
to
features
that
were
added
to
no
core.
That
was
one
that
I
was
I've
been
on
my
list
for
a
while,
but
I
haven't
done
anything
with
that
recursive
copy
we
have
on
here.
I
don't
actually
know.
A
Smartly,
because
this
is
done,
it's
it's
still
flagging
the
experimental
I
know
ben
wants
to
get
it
unflagged.
But
since
he's
not
here,
I
don't
really
know
what
the
status
of
that
is.
So
unless
anyone
wants
to
talk
about
that,
we
can,
we
can
move
on
to
everyone's
favorite
topic
of
discussion
argument.
B
A
I
saw
the
baton.
Oh
sorry,
go
ahead,
darcy!
No!
No!
Okay!
I
was
just
gonna
say
I
saw
that
ben
ben
put
up
pr.
I
think,
like
last
week,
to
use
the
primordials
in
that
library.
B
A
He
wasn't
100
sure
why
they
were
there
so
now,
they're
back
so
yeah
I
mean
that's
good.
Some
progress.
Anyways,
let's
see
what
else
is.
B
Yeah
john's
been
adding
a
ton
of
great
like
issues
john
g
shadow
spawn.
So
if
you
haven't
followed
along
in
there,
I've
given
like
a
ton
of
feedback,
a
ton
of
sort
of
back
and
forth
between
us
two.
I
think
the
biggest
thing
that
changed,
I
think
in
the
last
six
months,
since
we
kind
of
initially
kicked
off
what
I'm
calling.
I
guess,
v2
of
this
whole
farsag's
implementation
is
just
I
I
think,
and
to
wrap.
Maybe
summarize
it
here.
B
I
think
john,
like
something
shifted
between
the
way
it
was
specked
originally
and
the
way
that
we're
expecting
flags
to
return.
Now,
I'm
not
sure
if
you
either,
you
read
through
some
of
these
discussions,
but
I
think
john
misinterpreted
kind
of
like
I
think
the
intent
of.
B
Always
setting
boolean
values
within
flags
which
was
previously
called
args,
as
if
those
were
intended
as
like
a
argument
type
almost
like
a
flag,
it
was
actually
supposed
to
be
the
existence
of
the
flag.
So
originally
values
was
not
supposed
to
be
like
options
with
values
and
flat.
The
flags
returned
would
not
just
be
like
only
boolean
values.
They
weren't
supposed
to
be
like
two
different
types
of
option
types:
it
wasn't.
It
was
that
one
was
supposed
to
be
one
return.
B
Value
was
supposed
to
be
just
the
existence
of
that
flag
and
the
other
was
supposed
to
be
like
any
values
that
would
have
been
set
and
because
of
that,
what's
happening
is
like
we're
now
moving
options
from
one
set
to
another,
sometimes
with
or
without
the
with
value
option,
and
so
I
can't
meaning.
I
can't
look
at
a
single
s
like
a
single
return
value
to
check
the
existence
of
a
flag
or
not
like
consistently
so
anyways
I've.
A
B
Anything
yeah,
so
the
idea
with
shorts,
like
even
in
their
initial,
I
think
proposal
was
that
they
would
just
be
aliases
so,
like
a
short,
is
no
different
than
a
flag.
In
my
opinion,
like
it's
just
a
single
character,
a
flag
and
the
only
association
like
the
only
way
to
map
that
then
so
it
doesn't
dedupe
is
to
then
say
it's
an
alias
or
it's
a
it's
a
short
right,
a
short
for
a
different
value
or
a
different
flag.
Sorry
right.
A
Right
yeah,
that
makes
sense:
oh
yeah
yeah.
I
actually
realize
that
I'm
behind
on
these
discussions.
I
just
realized
that
I
didn't
have
notifications
on
for
this
repo,
so
turn
them.
A
A
Okay,
good
to
know
I'll,
try
and
find
some
time
this
weekend
to
get
caught
up
on
some
of
this
stuff.
I
still
do
want
to
work
on
the
strict
implementation
as
well.
B
B
Just
that
I
think
we
had
a
call
and
we
were
like
hey,
let's
yeah
like
we
should
have
a
strict
mode
and
then
that
that
would
kind
of
like
everybody
would
be
happy,
because
the
people
that
don't
want
this
to
ever
fail,
it
won't
fail
by
default,
and
then
people
that
want
this
to
throw
that
makes
sense
where
I'm
concerned
about
implementing
a
strict
is
just
like
what
what
do
we
consider
to
be
a
like
throwable
condition
like
what
is
the
like?
B
What's
an
air
case
really,
especially
because
we
are
discussing
about
in
one
of
those
other
threads
other
than
the
strict
thread,
the
default
parsing
of
like
foo,
equal
bar,
equal
bar
like
like,
if
we
de-sugar
that
to
be
the
same
as
with
value
foo,
pretty
much.
B
That
then,
is
there
really,
then
it
kind
of
like
introduces,
I
don't
know
like
with
value,
then
becomes
less
less
of
like
a
less
valuable.
In
my
opinion,
like
I
think
I
even
said.
As
much
and
in
my
feedback
to
that
idea,
because
with
strict,
if
you
have
strict
mode,
the
only
way
that
you
would
know
that
about
at
flag's
existence
is
by
setting
it
with
by
setting
with
value
and
so
like.
That's
the
only
way
that
those
two
things
could
work
together
to
actually
throw
an
error.
B
In
my
opinion,
like
that's
the
only
way
that
you
would
like
that
would
happen
right.
Like
you've
seen,
we've
been
able
to
see
that
there's
extra
arguments
that
were
passed
that
aren't
listed
in
with
value,
and
so
we
must
throw
like
that's
the
condition
that
I
imagined
would
be
possible,
but
if
we
default
parse
out
foo
equal
bar
and
make
that
assumption
that
we
can
do
that,
then
it
kind
of
throws
strict
mode
up
the
window.
I.
B
B
A
A
Yeah,
okay,
all
right
I'll,
yeah
I'll
have
to
read
up
on
that
to
see
kind
of
what
you
have
been
talking
about.
A
What
else
needs
to
we
did
create
a
milestone
here.
I
guess
for
what
we
need
to
do
to
get
this
merged
into
node.
A
A
B
Because
it's
kind
of
like
the
thing
that
I
I
harken
back
to
is
like
the
bring
your
own
validation
kind
of
concept.
I
originally
pitched,
like
quote:
unquote,
bring
your
own
validation
like
we
don't
want
to
have
to
do
any
like
type
coercion.
We
don't
have
to
do
any
like
type
checking.
We
don't
have
to
do
any
kind
of
like
inference.
B
We
don't
want
that
kind
of
stuff.
We
want
to
give
you
like
a
minimal
config
maximum
value
like
kind
of
feature
set
here
that,
like
let
you
quickly
scaffold
a
cli
tool
or
or
let
you
then
utilize
the
return
values
to
do
something
more
complex,
like
okay,
aliasing
or
like
building
like
a
yards
or
commander.
On
top
of
this,
like
really
nice
primitive
set
of
functionality,
so
I
think
shorts
are
great.
B
I
think,
and
the
way
that
we've
kind
of
defined
those
I
think
works
for
me
now,
I'm
on
board,
with
that,
I
had
like
a
knee-jerk
jerk
reaction.
The
last
time
we
talked
about
it
because
I
was
just
like
touching
up,
but
I
think
I'm
I'm
on
board
with
like
kind
of
like
what
was
said
out
there,
but
yeah.
Definitely
like
the
the
flags.
B
I
didn't
understand
what
really
why
it
was
being
changed
before
so
I
was
kind
of
like
okay,
that's
totally
fine,
and
I
know
ben
brought
it
up
in
our
last
call
like
oh,
is
that
okay,
are
we
all
plus
one
for
like
the
flag's
name
change
and
everybody
was
like
yeah,
but
I
didn't
really
realize
that
the
reason
why
it
was
it
got
changed
was
because
certain
values
were
going
to
start
to
be
omitted
from
that
from
that
set
from
that
array
so
like
and
only
exist
in
the
values.
B
B
The
args
object
would
always
have
a
reference
to
the
the
option
or
argument
if
we
were
able
to
parse
it
or
if
we
knew
it
was
expected
right
so
similar
to
like
you
trying
to
write
a
strict
implementation,
you
could
use
that
that
object,
knowing
that
something
was
supposed
to
be
provided
to
you
and
seeing
that
the
boolean
was
actually
false
right.
So
the
the
boolean
associated
with
the
the
key
was
actually
the
existence
of
that
thing.
B
So
it
would
always
be
true
if
we
we
for
for
each
one
of
those
keys,
that
we
parsed,
except
when
you
provide
when,
when
you
define
that
you
expected
a
certain
argument
in
the
with
value
option
and
then
and
then
that
wasn't
passed
to
the
the
program
right
and
then
and
then
that
would
be
false,
and
so
that
would
be
give
you
the
ability
to
essentially
check
for
missing.
B
You
know
flags
that
you
expected
and
then
you
could
throw
on
that
or
or
you
know
you
know
you
could
throw
separately
as
well
to
have
like.
I
have
an
example
in
one
of
these
threads,
where
I
show
you
how
you
can
throw
on
like
a
strict
mode
where,
like
you,
have
known
options,
and
if
you
see
anything
but
that
in
that
parsed
field,
then
you
would
then
you
can
throw
yourself
right.
You
can
just
do
your
own
kind
of
like
strict
mode
there.
B
B
But
I
don't
know
I'm
I'm
okay,
with
keeping
it
as
flags
for
now,
but
so
long
as
like,
we
don't
lose
that
that
so
long
as
we
don't
lose
like
that
understanding
of
like
the
scene
or
existed
like
existence,
because
otherwise
I
have
to
go
looking
at
two
different
objects
to
actually
figure
out.
If
a
flag
was
passed,
I
have
to
check
both
flags
and
or
values
to
see
if
it
if
there
was
like
a
flag
was
and
and
in
terms
of
quickly
scaffolding.
B
A
program
like
all
I
want
is
like
a
force
flag
right
like
or
something
like
that.
You
know
half
the
time.
You
only
want
one
or
two
like
actual
arguments
to
to
see
a
lie
program
and
in
in
those
quick
prototyping
cases
and
like
I
feel
like
if
I
have
to
check
two
different
places
for
existence,
it's
not
gonna.
Be
yeah
fun.
A
Yeah,
I
see
what
you
mean.
This
could
be
like
error
prone
to.
If
people
don't
really
know
that
they're
supposed
to
check
in
both
places
or
whatever
yeah,
I'm
kind
of
in
favor
of
there
being
one
like
you
should
get
yeah
one
list
of
all
of
the
options.
C
Now,
if
I,
if
I
recall
this
correctly-
that's
been
a
while
now,
but
I
think
these
two
things
might
be
tied
to
original
flashbacks
in
the
in
the
original
pr,
the
chris
open
long
ago,
right
that
didn't
land.
First
time
we
tried.
So
I
believe
there
was
some
pushback
with
regards
to
flags
to
the
specific
case
where
you
just
define
a
flag
equals
false
and
again.
How
could
you
tell
that
without
extra
confidence?
C
A
A
B
Yeah,
so
I
do
remember
like
the
reason
why
I
came
up
with
this,
like
change
is
because
I
like
read
through
thoroughly
the
feedback,
and
we,
like
I
tried
to
thread
essentially
the
needle
of
everybody's,
wants
and
needs
there
that,
like
definitely,
I
think,
you're
right
roy
that,
like
some
people,
wanted
the
throw,
but
I
think
it
was
because
it
was
ambiguous
whether
it
should
or
should
it
be.
If
we
say
explicitly,
we
don't
want
to
throw
like.
I
think
that's
that's
fine
and
the
reason
is
because,
like
if
you
imagine.
B
I
I
think
it's
going
to
step
too
far
to
to
assume
that
we
should
throw
when
we
see
things
we
weren't
explicitly
like
like
like
expecting
to
see.
I
don't
know.
A
Yeah
I
mean
that
kind
of
makes
sense
sounds
familiar.
I
forgot
that
you
were
the
one
that
that
yeah
did
go
through
all
that
feedback
and
and
write
up
this
new
proposal.
B
So
I
felt
really
bad,
like
chris
said
spent
so
much
time
on
on
it
and
then
it
kind
of
you
know
it
seemed
like
it
was
coming
to
an
end,
and
there
was
obviously
you
know
a
lot
of
energy
at
the
time
that
we
had
put
into
it,
and
we
wanted
to
like
circle
back
on
on
this
and
find
a
way
to
make
it
happen,
and
it
seemed
like
separating
out
again,
like
the
eureka
moment
for
me,
was
suffering
out
of
like
the
actual
values
from
the
exit,
like
the
actual
value
of
the
options
versus
the
existence
of
them,
was
kind
of
like
the
eureka
moment
for
me,
like
putting
them
into
these,
like
like
three
different
return,
values
and
then
saying
that
we
don't
and
then
also
like
being
a
staunch
advocate
for
saying
we
don't
try
to
coerce
anything.
B
So,
like
I
think,
roy
mentioned,
that,
like
foo,
equal,
false
or
something
like
one
of
the
biggest
pieces
of
feedback
was
like
okay.
Well,
do
we
because
what
was
happening
before
was
like
some
type
coercion,
where
we
would
actually
set
a
boolean
for
for
flags
and
people.
I
think
the
original
implementation
would
take
like
false
and
actually
return
a
boolean
like
a
proper
like
yeah
boolean,
instead
of
string.
So
now
we
we
say:
no,
we
don't
do
that
at
all.
B
We
don't
deal
with
any
types
of
objects
except
strings,
like
all
we're
going
to
return
is
a
string,
no
matter
what
string
string
all
the
way
down.
So
I
feel
like
saying
that,
like
is
going
to
that,
that
is.
A
goal
is
going
to
appease
a
lot
of
people,
because
it's
like
okay,
well,
you're,
you're,
doing
just
string,
manipulation,
there's
no
way
for
you
like
there's
no
way
for
us
to,
like,
I
don't
know,
make
the
wrong
choice
here,
because
we're
just
going
to
return
you
back
like
what
we
sliced
up
right
like.
A
Right
yeah,
which
kind
of
makes
sense
in
the
context
of
like
strict
mode,
then
like
what
would
the
error
be?
You
know
if
we're
just
yeah,
just
parsing
the
string
and
returning
using
strings
exactly.
A
I
do
like
the
idea
of,
like
I
mean
realistically
you're.
Probably
gonna,
build
your
own
validation,
where
you
maybe
would
turn
something
like
the
string
false
into
a
boolean,
false
and
check.
You
know
for
options
that
are
maybe
mutually
exclusive
or
an
option
that
needs
a
value
that
wasn't
provided
so,
maybe
just
leaving
that
all
kind
of
after
the
user
is
is
kind
of
the
way
to
go.
A
Okay,
well
yeah,
I'm
definitely.
I
had
not
seen
the
flurry
of
discussion
this
week,
so
I'll
definitely
get
some
reading
to
do
this
weekend.
A
A
Excuse
me:
okay,
does
anyone?
Is
there
anything
else
we
wanna
talk
about
here
on
the
call
today,
or
I
mean
yeah,
it
sounds
like
you
know,
for
my
part,
anyways.
I
need
to
get
caught
up
on
these
discussions.
B
No
just
kind
of
get
this
over
the
finish
line.
I
should
have
wrote
the
like
initial
implementation
and
just
like,
I
just
never
had
time
at
the
at
that
point
in
time.
So
I
know
that
would
have
helped
with.
I
think,
a
lot
of
like
a
lot
of
the
confusion
is
like
right.
Now
we
have
like
the
code
is
like
half
done
so
like
like
you're,
seeing
like
some
people
are
trying
it
out
and
like
oh
wait.
A
A
Yeah,
I'd
like
to
I'd
like
to
spend
some
time
on
this
this
weekend
as
well.
So
I'll
start
by
reading
all
these
issues
and
then
see
if
there's
something
that
I
can
work
on.
B
Because
the
I
would
deal
with
your
strict
mode
just
and
I'm
trying
to
let
me
look
back
at
the
proposal
for
strict
so
boolean
on
whether
or
not
to
throw
an
error
when
unknown
arcs
are
encountered,
so
the
the
unknown
errors
are
our
arcs
are
encountered
would
be.
B
So
all
three,
if
I,
if
I
define
anything
in
any
one
of
those
three,
then
that
becomes
the
list
that
the
like
the
exemplist
or
something
I
don't
know.
A
B
Well,
just
like
the,
I
think
the
example
of
like
how.
How
would
you
do
that
without
a
strict
mode
would
be
literally
you
set
define
the
known
like
it's
a
one-liner
like
you
define
the
array
of
known
values
like
in
a
variable,
and
then
you
pass
that
and
then
you
check,
if,
like
there's
anything
that
isn't
that
set
past
passed
back
in
the
flags
all
right
right.
B
B
A
A
Okay,
yeah
well,
like
I
said,
I'm
going
to
hopefully
get
caught
up
on
this
this
weekend
and
if
there's
some
work
that
I
think
I
can
do
there,
I
will
we'll
do
that.
Maybe
I'll
I'll
post
on
slack
or
something
before,
if
I
afford,
like
taking
anything
on
just
make
sure
everyone's
on
the
same
page.
B
B
Yeah
apologies
like
I
definitely
was
making
up
for
lost
time.
So
you've
got
like
a
book
to
read
there
so.
A
B
Yeah
yeah,
one
of
the
sticking
points
you'll
see,
is
also
like
whether
or
not
we
should
be
passing
back
undefined
versus
an
empty
string,
so
it's
in
in
the
cases
where
let's
say,
equals,
and
and
then
no
this
like
value
suffixing
or
like
ending
it.
What
gets
returned
there
and
like
originally,
I
thought
we
were
going
to
pass
back
undefined
in
those
in
those
cases,
but
then
also
like
what
would
you
do
with
like
dash
foo
without
the
equal
sign?
B
Is
that
the
same
thing,
and
then,
if
you
did
that
jeff
foo
equals
and
then
like
actual
like
parens
and
not
friends,
but
like
like
an
empty
string
and
and
then
that
those
two
actually
in
shell,
like
d
sugar,
to
the
same
thing
anyway,
so
I
was
just
like
okay.
Well,
are
these
all
then
just
empty
strings?
Should
we
always
just
return
an
empty
string
for
all
three
of
those
cases,
jordan
had
some
like
thoughts.
B
He
wanted
undefined
john
g,
like
was
okay.
I
think,
with
like
empty
strings
for
all
three
of
those
cases,
so.
B
You
just
escape
it,
you
just
you
escape
it
yeah,
so
you
do
like
yeah
back
backslash
quote:
backslash
quote.
A
Okay,
cool
well
yeah.
It
sounds
like
we
got
a
bit
of
a
plan
at
the
moment.
Anyways
anyone
have
anything
else.
They
want
to
talk
about
about
parsogs
or
anything
else.
A
Yeah
yeah,
that's
it
yeah,
I'm
starting
to
starting
to
roll
and
em
out
in
in
a
bunch
of
projects,
including
ones
that
use
overrides
and
so
far
so
good.
A
Okay,
that's
good
to
hear!
Okay!
I
gave
I
gave
luke
some
other
feedback,
so
yeah,
I'm
not
sure
if
you
shared
that
with
you,
but
no
I
didn't
get
that
all
right.
I
mean
nothing,
nothing
too
important
that
we
need
to
talk
about
now,
but
yeah.
A
Maybe
we
can
talk
on
on
slack
or
twitter
or
something
later
all
right.
Well,
yeah,
we'll
make
it
make
a
short
meeting
today,
thanks
for
thanks
for
joining-
and
I
will
talk
to
you
guys
later
cheers.