►
From YouTube: JupyterLab Team Meeting - 5 July 2023
Description
A meeting to share and discuss features, ideas, issues, and pull requests in JupyterLab and other Jupyter frontends. This meeting is open to anyone and everyone.
Join future calls via the Jupyter community calendar: https://docs.jupyter.org/en/latest/community/content-community.html#jupyter-community-meetings
Notes for upcoming meetings can be found on the agenda: https://hackmd.io/Y7fBMQPSQ1C08SDGI-fwtg
Past notes can be found on the JupyterLab team compass: https://github.com/jupyterlab/team-compass/issues/205
A
A
Oh,
hey
there
folks
I'm
Tony
fast
and
it's
July
5th,
and
this
is
our
weekly
Jupiter
lab
development
call
we
have.
Currently
we
have
nine
folks
on
the
agenda
and
yeah.
There
are
a
few,
a
few
agenda
items.
So,
let's
get
started
with
Jason.
B
Hey
good
morning,
good
evening,
everyone
so
I
am
curious
about
adapting
extensions
to
support
both
Jupiter
Lab,
3
and
Jupiter
lab
four.
This
is
something
that
I've
been
working
on.
My
team
has
Jupiter
Ai
and
Jupiter
scheduler,
both
of
which
are
pinned
to
Jupiter
lab
three
dependencies.
B
I've
tried
to
upgrade
to
Jupiter
lab
4,
and
the
problem
is,
is
that
there
are
certain
aspects
like
in
the
file
browser.
The
iterator
result
uses
has
a
different
schema
in
Jupiter
lab
four,
but
the
code
is
expecting
the
Jupiter
Lab
3
schema
and
it's
really
tricky
I've
seen
people
have
been
trying
to
upload
upgrade
other
things,
but
I'm
curious
about
what
the
the
proper
way
forward
is
like.
Should
we
be
using
the
same
code
to
support
both,
in
which
case
we
have
to
basically
support
multiple
schemas?
B
We
have
to
kind
of
ignore
certain
type
guard
rails
that
typescript
puts
in,
or
do
we
just
Fork
every
incompatible
extension
and
have
a
Jupiter
Lab
3
and
a
Jupiter
lab
4
variant
of
each?
What
do
folks
think
about
the
way
forward
on
this.
B
It's
already
happened
to
people
on
36x
well,
I'm
not
experienced
with
the
with
the
the
compatibility
problems.
36X
I
I
have
seen
that
the
major
upgrade
broke
schemas
for
things
like
the
iterator,
like
the
iterator
result
that
there's
just
another
accessor
that
I
have
to
call.
B
C
B
Right
so
like
for
Jupiter
scheduler,
we
have
a
1X
and
a
2X
yeah
Nick
is
is
saying
the
same
thing
in
the
chat
and
then
we
would
have
to
like
maintain
both
versions
with
bug,
fixes
or
any
later
patches.
B
Yeah
I
I
can
see
that
and
git
doesn't
make
that
particularly
painless,
but
it
is
certainly
possible.
We
do
that
for
Notebook
and
Jupiter
lab
right
now.
Glpm
won't
work
the
same
way
between
them.
What
does
that
mean?
Nick.
C
That
different
versions
of
yarn
and
they
they
are
conflicting.
That
would
be
very
hard
to
maintain
two
versions,
but
you
can
always
choose
to
use
npm
or
fix
a
different
version
of
yarn.
It
will
be
just
quite
painful
if
you
wanted
to
maintain
versions
compatible
with
3.x
and
4.x,
just
because
the
two
link
changed
so
much
so
yarn
different
versions
of
typescript
yeah.
B
Okay,
this
is
something
that
I
want
to
discuss
with
my
team,
because
it's
going
to
be
a
tough
sell,
because
we've
already
got
two
extensions.
We're
maintaining
this
effectively
doubles
that
to
two
extensions
crossed
with
two
Jupiter
lab
major
versions,
but
if
that
that
seems
to
be
so
far
the
least
hacky
solution
so
far,.
A
Anybody
else
have
any
comments.
C
Yeah
I
will
be
speaking
about
one
one.
Is
it's
my
turn,
but
also
I
think
at
least
six
extensions
that
reported.
So
if,
if
there
is
an
issue
tracking,
all
the
ported
extensions
I
will
link
that
in
the
chat.
A
Well,
why
don't
we
just
use
this
time
to
Trend
to
switch
over
to
you
Mike
if
you're
ready,
and
we
can
just
continue
along
the
same
thread.
A
No
problem
I'll
go
I'll,
I'll,
I'll
jump
in
and
and
do
my
update,
then
cool
so
I
wanted
to
talk
about
keyboard
issues
and
parody
with
notebook
classic.
A
So
one
of
the
things
that's
kind
of
been
seems
to
be
a
little
bit
of
a
challenge
from
triaging.
Like
organizational
point
of
view,
is
collecting
the
keyboard.
Yet
the
lack
of
like
keyboard
parity
between
the
old
version
of
MB
classic
and
Jupiter
Lab
3,
with
the
new
version,
that's
using
the
Jupiter
lab
for
base,
so
I
spent
a
very
short
period
of
time.
A
I
did
not
spend
much
time
collecting
some
issues
from
Jupiter
lab
I
was
wondering
if
any
folks
could
take
like
a
10
minutes
themselves
or
if
they
had
any
eyes
on
any
issues
related
to
keyboard
shortcuts.
We
kind
of
collect
them
in
one
place,
because
we
did
promise
some
keyboard
parody
and
right
now
things
are
sort
of
like
fractured
and
broken
up
across
places.
So
it
really
help
us
in
organizing
everything.
A
If
yeah
we
get
a
few
eyes
on
this
yeah,
there
are
some
things
that
are
are
weird
and
sort
of
uncomfortable.
If
you're
an
experienced
notebook
user
that
would
be
good
to
sort
out
before
we
do
a
big
release,
otherwise
I
think
folks
might
feel
a
little
bit
awkward
in
their
editing
experience,
Isabella.
D
I'm
curious:
are
you
asking
here
for
people
to
collect
other
issues
that
already
exist
for
this
topic?
Are
you
asking
people
to
do
their
own
review?
I
thought
it
was
like
kind
of
their
own
review
last
time
and
I
wasn't
sure.
A
No
just
issues
that
exist
just
like
just
just
so
we
can
like
a
keyboard.
Milestone
would
be
really
helpful
in
having
some
perspective
on
that,
and
we
heard
the
last
meeting
that
folks
are
working
on
some
of
these
things.
So
the
more
we
can
help
with
the
entropy.
Maybe
we
can
get
more
fixes
in.
A
Okay,
so
the
next
thing
is
on
keyboard
stuff
right
now,
I
think
a
lot
of
things
have
been
tagged
that
are
keyboard.
Issues
have
been
tagged,
accessibility,
which
is
the
case
sometimes
but
I
think
we
need
to
think
about
a
way
to
triage
keyboard,
shortcut,
navigation,
editing,
sort
of
issues
a
little
bit
differently.
A
They
are
hard
ux
to
write
issues
for
and
treating
them.
I
don't
know
if
there's
a
way
that
we
can
treat
them
differently
than
we
are
now.
Yes,
sometimes
they
are
accessibility
issues,
but
sometimes
they
are
just
downright
user
experience
issues.
So
I
don't
know
if
any
folks
have
any
suggestions
about
ways
that
we
might
be
able
to
like
organize
these
things.
The
keyboard
issues
a
little
better.
A
Yeah
yeah
yeah,
so
I,
yeah,
I,
didn't
even
think
about
the
international
I,
don't
even
know
if
it's
internationalization
or
localization
at
that
point
it's
one
or
the
other,
it's
some
numer
in
them,
but
yeah,
that's
a
good
point
and
maybe
there's
like
a
ux
keyboard
tag
or
something
we
could
have.
B
Yeah
I'm,
okay,
with
creating
a
label
for
like
keyboard
compatibility
and
because
I
mean
you
know,
Nick
points
out
that
not
all
keyboards
are
tied
to
a
nation.
Even
within
a
nation
you
have
like,
there
are
Mac
keyboards.
There
are
windows
keyboards.
We
even
had
a
bug
where
presumably
US
keyboard
user
had
one
of
those
trendy
60
keyboards
that
lacks
certain
unshifted
Keys
and
Jupiter
lab
4
broke
his
user
experience,
so
I
I
think
this
trans.
This
goes
beyond
just
internationalization,
localization
to
like
user.
B
Like
usability
with
a
specific
eye
on
on
keyboards,
so
should
I,
add
I,
don't
think
we
have
a
label
for
keyboard.
No,
we
don't
I
can
add
that
right
now,
cool.
A
A
Alrighty
anybody
have
any
thing
any
comments
or
questions
on
keyboard
related
things.
D
E
D
A
Oh
I,
gotta,
yeah
I,
will
add
the
link
to
the
issue
there
and
great.
A
A
Thanks
for
creating
that
keyboard
label
that'll
be
useful
when
we
go
through
triaging
cool.
A
All
right:
well,
that's
all
I
have
maybe
it's
a
good
time
to
move
over
to
Mike.
It
seems
like
he
has
lots
to
talk
about
I'm,
so
excited
it's.
Your
turn.
Buddy.
C
All
right
excuse
sound
quality.
If
it's
about
I,
I
personally,
say
I
had
to
skip
admitting
so
yeah.
The
first
one
is.
There
is
a
very
cool
request
on
NB
time,
which
improves
the
performance
of
The
Thing
by
up
to
150
times
when
there
is
a
large
notebook
and
I.
C
Think
vdr
is
not
on
this
call,
but
I'm
I'm,
just
very
excited
about
it
and
likes
to
see
it
merged
and
I
also
know
that
many
people
are
waiting
on
it
because
gitcup
uses
it
to
render
notebooks
and
any
large
notebooks
which
cannot
be
tipped
on
the
user
interface
on
GitHub
right
now,
because
it's
just
too
slow
and
and
that
requested
to
fix
that.
So
if
the
the
issue
is
that
it
changes
the
behavior
of
the
thing
slightly.
C
So
if
there
is
a
cell
which
gets
deleted
and
then
a
new
cell
is
added,
if
their
contents
is
the
same,
they
will
no
longer
be
identified
as
unchanged.
They
will
be
they
they
will
appear
as
if
one
was
deleted
and
the
other
AD
that's,
as
was
the
case
rather
than
to
the
same
cell,
so
yeah.
C
If
anyone
is
interested
or
knows
how
push
that
forward,
please
check
out
the
request,
and
there
is
also
a
second
small
pull
request:
adding
a
user
interface
change
to
expose
the
cell
ID
make
the
data
to
users.
If
anyone
adds
opinions,
I'm
pretty
excited
to
hear
you,
for
example,
this
foreign.
C
C
C
C
So
my
next
point
is
just
to
highlight
that
spiritual,
correct
station
is
out
and
it's
it
was
a
bit
of
a
challenge
to
rewrite
it
with
code,
mirror
6,
but
I'm
very
satisfied
with
the
result,
because
it
allows
us
to
finally
spell
tracking
code
commands
and
strings,
and
it's
so
much
nicer
to
use
right
now.
There
will
be
a
much
release
coming
today
with
more
user
experience
improvements
if
something
broke,
please
open
an
issue
because
it
was
a
substantial
rewrite.
C
Okay,
under
the
next
Point,
there
is
a
discussion
about
whether
the
computer
provider
should
be
able
to
advertise
their
rank
so
for
context.
Complete
a
computer
computer
provider
provides
completions.
So
when
you
press
stop,
it
can
add
more
suggestions.
C
Natural
conflicts
and
yeah
I'm
curious
with
another
one
has
been
used.
Please
leave
your
thoughts
on
the
issue.
E
C
Second,
to
last
point
is
about
adding
CSS
validator
as
a
very
nice
extension
for
plugin
for
styling,
and
this
one
found
a
number
of
instances
where
our
values,
the
CSS,
were
invalid
and
I
actually
didn't
know
that
styling
was
not
tracking
whether
the
values
are
correct.
C
Unless
you
know
better
Alternatives,
then
please
do
comment
on
the
issues
or
pull
requests
and
finally,
just
wanted
to
look
back
in
into
extension
compatibility
table.
There
is
an
issue
1459
or
please
have
that
in.
If
you
have
rights
to
edit
just
edit
the
comments
that
will
help
us
to
see
how
the
work
is
going
and
whether
it
will
also
help
users
to
understand
what
they
can
already
migrate
to
4.0,
because
whether
the
accessions
they
are
depending
on
are
broadly
important
or
not
so
yeah.
That's
all
from
me.
A
Thanks
so
much
for
all
those
updates,
we
really
appreciate
all
that
work.
Man,
any
comments
or
questions
for
Mike.
A
All
right:
well,
if
there
are
no
comments
or
questions,
there's
nothing
else
on
the
agenda.
Is
there
any
less
things
that
folks
might
like
to
say
before
we
stop
the
recording.
D
Nothing
big,
but
are
we
still
trying
to
keep
scheduling
different
hosts?
Do
we
have
the
issue
for
that
or
should
I
go
looking
for
it.
A
We
do
have
a
issue.
Good
good
looks.
It
seems
that
okay,
we
do
not
have
anybody
for
the
12th,
so
we
are
looking
for
a
host
for
next
week
and
if
you
want
to
volunteer
yourself,
there
is
the
issue:
oh
yep,
thanks
for
posting
that
Jason
yeah.
So
if
anybody
who
wants
to
try
their
hand
to
host
in
one
of
these
lab
meetings,
yeah
sign
up
over
there.
Thank
you.
E
How
do
y'all
ran
into
a
few
things
working
on
that
mermaid
PR?
It's
looking
better.
E
E
So
previously
Vega
you
run
Vega,
it
does
canvas
junk.
It
also
would
put
that
the
fully
rendered
PNG
back
into
the
notebook,
and
it
doesn't
do
that
anymore.
E
So
I
got
a
PR
App,
but
I,
don't
know,
I
mean
I.
Can't
I
haven't
really
reasoned
about
what
that's
going
to
do
in
a
multi-user
setting
if
the
inputs
are
not
deterministic.
If
there's
some
re,
you
know
if
you're
running
a
layout,
folder
or
something
inside
your
browser,
so
I
don't
know,
there's
a
PR
open
for
it.
It's
linked
from
the
mermaid
PR.
E
Also,
we've
started
getting
bit
by
pedantic
going
to
2.0.
That
was
only
a
matter
of
time,
specifically
copier
isn't
ready
for
pedantic
or
pedantic.
Two,
so
I
think
we've
handled
that
in
a
couple
places,
but
there's
no
Upstream
fix
yet
and
then
one
weird
little
thing
on
the
docks
we
check.
E
Github
usernames
and
someone
changed
their
GitHub,
username
and
GitHub
doesn't
redirect
from
old
usernames,
which
means
that
any
of
us
on
here
that
has
committed
anything
ever
that
goes
into
the
changelog
can
break
our
ability
to
do
releases
because
all
that's
very
tightly
bound
and
Link
checked
and
all
that
so
I
I,
don't
know,
I,
don't
know
what
the
right
thing
is
to
do
about
that.
A
Cool
any
comments
or
questions
for
Nick.
A
I
shared
a
link
to
his
mermaid
PR
in
the
Zoom
chat
ends
in
the
agenda.
Be
cool
folks
could
check
it
out.
I'm
excited
for
it.
E
Yeah
I'll
call
that
the
most
recent
changes
so
Paul
on
the
on
steering
committee,
lovely
person
raised
the
very
valid
concern
that
affects
anybody
that
does
anything
outside
of
basically
the
core
mermaid
or
the
core
markdown
spec.
E
You
know
adding
something
even
across
two
renderers
doesn't
mean
that
something's
going
to
render
the
same
way
in
the
future.
So
if
we
were
serializing
the
full
render
into
a
notebook
at
least
then
we'd
have
that
in
the
future
of
sales
call,
we've
talked
about
markdown
cells,
doing
that
where
they
would
serialize
to
text
HTML
and
store
that
as
an
output.
So
if
you
run
a
notebook,
you
will
get
exactly
the
same
Dom
out,
no
matter
how
many
times
you
open
it
until
you
re-run
it
or
re-render
it.
E
E
Bearing
on
Magics
and
stuff
like
that
too,
so
that's
about
that's
just
the
backstory
down
there.
A
Alrighty,
any
more
any
comments
or
questions
or
last
minute
updates
that
y'all
feel
compelled
to
share.
A
A
Well,
I
think
that
that
means
I'm
going
to
stop
the
recording.
If
there's
any
for
anybody,
say
anything
last
minute,
so
I
guess
as
Darian
always
says,
brace
yourself.