►
From YouTube: Open Telemetry's Personal Meeting Room
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
B
A
B
Yeah
I
think
it's
a
conflict
he
might
be
joining
later.
So.
D
Yeah,
let's
wait
for
one
more
minute
and
we
will
start.
D
All
right,
I
think
it's
it's
just
on
this
group.
Let
me
share
my
screen.
D
Yeah
yeah
written
did
you
get
the
first
one.
B
B
I'll
be
pretty
much
out
until
beginning
of
next
year,
so
we'll
be
missing.
The
next
three
six.
D
I
I
will
not
be
available,
the
from
the
11
to
18.
and
and
and
following
that,
the
following
week
from.
Let
me
see:
yeah
11
to
18
and
20
24
to
30.
A
D
Yeah
I
think
we're
also
like,
like
due
for
our
like
I
think
before
we
leave.
If
we
should
probably
do
another
release,
it's
almost
one
month
in
the
last
release,
I
think
if
you
do
it,
you
know
this
week
or
next
week.
We
don't
have
to
worry
about,
like
whatever
the
changes
so
far,
we
will
do
a
release
and
then,
when
we
are
back
after
the
holidays,
we
will
I
think
we
will
be.
D
It
will
be
again
like
another
release
after
that,
so
anything
like
it
will
be
like
throughout
the
December,
I,
I,
hope,
I
guess
there
will
not
be
much
activity.
D
D
B
Oh
I,
pretty
much
kind
of
starting
off
I'll
be
I'll,
be
off
starting
tomorrow.
So.
D
I
will
probably
do
like
I
I'll
need
the
scriptures,
but
yeah
I.
Think
planning
like
should
do
some.
We
should
do
release
next
week,
probably
sure.
D
I,
don't
will
you
be
available
like
I
think
what
what
will
be
your
schedule?
Will
you
be
available
or
will
you
be
I
think
you
can
also
add
your.
C
Yeah,
you,
you
said
Aaron
right,
yeah,.
D
C
Yeah
I
mean
I,
have
you
know
like
holidays
off,
but
I
haven't
I'm,
not
sure
what
additional
time
I'm
taking
so
I'll
I'll
get
back
to
get
back
to
you.
D
Yeah
cool
thanks,
yeah
I
think
that
the
value
people
who
are
doing
the
reviewing
work
I
think
if
there's
nothing
else,
we
can
move
on
to
next
topic:
DC
review,
metrics.
Let
me
open.
C
Yeah
I
added
this
one,
so
Josh
got
around
to
doing
the
review
versus
the
spec
Matrix.
E
C
Only
open
these
two
issues,
I
think
Leighton,
you
already
charged
them,
but
I
think
they're,
pretty
straightforward
in
the
issues
describe
them
they're
both
related
to
like
this
duplicate
registration,
how
it
interacts
with
views,
but
there's
examples
and
issues
that
should
outline
how
to
fix
it.
So
yeah
I
think
the
overall
feedback
also
was.
C
It
would
be
good
to
have
more
like
integration
tests
for
specifically
these
kinds
of
things,
because
it's
hard
to
tell
what
the
actual
behavior
is
from
like
the
unit
tests
that
are
testing
the
just
the
really
local
Behavior.
So
if
we
had
some,
you
know
create
a
top
level
meter
provider
and
check
that
it
logs
or
check
that
it
fails
check
the
output
with
the
in-memory
metric
reader
thing,
I
think
that
would
yeah.
That
would
be
a
good
addition
to
like
solving
these
so
other
than
that
I
think
things
are
pretty
good,
so.
D
Yeah
I
I,
don't
I
think
we
have
like
at
least
some
integration
test
for
the
tracing
I.
Don't
think
we
don't
we
have
any
further
metrics.
D
Yeah
this
this
is,
this-
is
looked
into
on
this
okay,
I
think
that's
about
it.
B
Nope,
it's
pretty
awesome
that
he
triaged
it
I
mean
that
he
looked
through
the
whole
thing.
D
Yeah
we
should
should
thank
him
I'm
going
to
like
if
you
attend,
one
of
the
maintenance
meetings
are
where
he
attends
like
I.
Think
we
can.
We
should
all
right
him
taking
time
and
then
reviewing
it
yeah.
That's
the
topics,
I,
don't
see
any
other.
Does
anyone
else?
Have
anything
look
like
to
bring
it
up
here.
D
B
Just
wondering
if,
like
what
the
kind
of
timeline
is
to
bring
the
specs
to
stable,
I
haven't
personally
been
attending
the
logging
6
either,
but
yeah.
Just
thinking
about
the
effort
behind
that
and
like
what
it
will
take
to
kind
of
move
it
to
stable.
D
Uh-Huh,
so
so
there
is
one
the
other
group
like
a
client
instrument
like
this
one
set
of
people
who
are
working
on
the
like
event,
kind
of
APA
part
where,
like
they
are
there,
there
are
the
ones
who
are
actually
pushing
like
okay,
so
the
greater
model
fee
is
fixed.
D
What
should
be
when
you're
doing
with
the
you
know,
real-time
user
monitoring
are
where,
like
there,
the
concept
of
event
I
think
they
are
doing
some
previously
work,
and
you
know
they
will
share
like
okay,
so
once
they're
done
with
that
previously
and
then
how
they
want
the
that
to
be
like
that,
it
even
taper
to
be
model.
D
You
will
probably
have
some
idea,
okay,
so
when,
when
is
this
going
to
be
like?
What's
the
timeline
for
that?
But
but
currently
there
is
no
like
there's
no
strict
discussion
around
like.
When
do
we
want
to
complete
this.
D
D
Yeah
so
I
I,
I'm,
I,
yeah
I,
don't
have
a
like
anymore.
Oh.
D
Anyone
else
have
anything
to
add
or
any
other
topic.
They
want
to
talk
about.
D
Yeah
we'll
come
to
that.
Like
any
other
topics,
anyone
else
have.
D
All
right,
let's,
let's
move
to
the
issues.
F
Yeah
so,
regarding
this,
I
have
I
have
looked
into
the
other
six
implementation,
so
they
have
kind
of
implemented
the
feature,
but
it's
not
available
as
part
of
our
transformation
is
only
available
in
a
manual
instrumentation
using
SDK.
So.
A
F
To
look
into
notice,
then
the
dotnet
and
Java
and
Co,
so
all
of
them
have
implemented
this
feature,
but
only
in
a
manual
instrumentation.
There
is
no
it's
not
available
in
Auto
instrumentation
at
all.
Just
yeah.
D
But
I
realized
that
that
there
is
no
way
for
you
to
specify
it
in
Auto
instrumentation.
F
A
D
D
Yeah
I
brought
this
up.
I
asked
like
I,
create
this
issue
that
you
know
so
that,
like
similarly,
how
we
support
the
our
export
to
selection
with
the
EnV
I
wanted
to
do
such
similar
capability
with
the
resource
detectors,
but
but
really
said
that
they're
like
they're
putting
it
on
a
hold
like
they're,
not
going
to
add
any
more
any
more.
D
Even
we
were
able
to
specification
like
I'm,
not
sure
when
they
will
unfreeze
it
or
what's
going
to
be
the
plan
ahead,
but
so
the
the
TC
like
looks
like
they
have
decided
that
they
put
a
related
changes
until
the
formula
is
configuration
mechanism
in
place
and
the
one
which
we
are
suggesting
it.
It
falls
under
the
non-primitive
type
like
it's,
not
a
regular,
primitive
type,
and
so
they
don't
want
to
accept
any
more
envisa
like
SDK
Envy
configuration
which
are
non-timitive
even
even
the
Primitive
ones.
D
They
are
going
to
evaluate
Case
by
case
spaces,
so
I
guess
I,
guess
I,
don't
think
we
will
be
able
to
until
like
until
that
gets
resolved.
I
don't
think
we
can
implement
the
this
EnV.
E
D
You're
saying
no,
there
is
no
entry
like
there
is
no
EMA
variable
I
so
before
we
implement
it
in
the
SDK.
I
wanted
to
formally
get
it
I
accepted
in
the
specification,
but
the
TC
said
they
are
not
going
to
accept
anymore
until
they
have
like
other
formal
configuration
mechanism.
D
B
D
Yeah
something
that
I
mean
I
I
think
you
can
continue
to
work
on
the
adding
the
resource.
Detector
part
users
may
not
be
able
to
use
it
in
the
auto
instrumentation
and
you
won't
be
able
to
add
the
entry
point
currently
I
mean
you
can
add
the
entry
point,
maybe,
but
it
it
will
be
written
if
the
values
change,
but
you
you
don't
have
to
be
waiting
on
that
like
that.
Like
that's.
A
D
B
Hey
sir
Kath
I
was
wondering
or
everyone
else
what
everyone
thinks
about
like.
So
like
the
mechanism
we
use
for
auto
instrumentation
like
it's
like,
you
know
something
that
we
came
up
with,
and
it's
like
python
specific
and
not
really
specked
out
or
supported
anywhere
was
wondering
like
what
our
kind
of
policy
would
be
is
for
like
supporting
features
that
are
specific
to
Auto,
instrumentation
and
obviously
like
it
is
important,
because
a
lot
of
our
users
are
already
using
Auto
instrumentation.
B
But
technically,
it's
like
not
specced
out
anywhere
like
what
what
would
be
our
like
product
story.
Moving
forward.
D
B
Different
Stakes
have
taken
different
routes
for
like
their
their
Auto
instrumentation
capabilities,
right
functionalities
and
I.
Think
I,
guess
what
I
mean
is
like
we
expose
Auto
instrumentation
as
like,
like
if
a
feature
of
the
Python
SDK.
B
B
B
Like
should
Auto
instrumentation
be
like
a
top
level
like
first
level
kind
of
concern
for
us,
it
seems
like
so
far,
we've
been
taking
the
approach
of
like
let's
implement
this
for
manual
instrumentation
and
like
just
hope
that
it
works
for
auto
instrumentation
and
then,
if
people
have
issues
they
just
like,
you
know
it
just
comes
up,
and
they
just
report
the
book.
That's
kind
of
like
the.
How
to
the
approach
that
we've
been
taking
right.
D
D
Yeah
I
I
think
it's
it's
like.
We
don't
have
any
clear
strategy
and
like
currently
like,
if
you
take,
there
are
lots
of
things
that
we
support
through
the
with
the
manual
instrumentation,
but
there
is
no
way
for
want
to
achieve
the
same
with
other
instrumentation,
and
if
you
want
to
do
that
it,
it
will
either
be
something
that
can
be
generalized
and
added
to
the
spec
that
plug
either
again.
We
are
something
like
that,
or
it
might
be
something
specific
to
python.
D
I
think
I
think
it's
a
good
part
good
that
you
bring
it
up.
I
think
this
needs
a
like
a
broad
discussion.
A
B
I
think
yeah
I
think
it's
worth
to
look
into
my
thoughts
like
also
like
a
few
of
every
anyone
feel
free
to
try
me
here.
It's
like.
If
we
decide
to
make
Auto
instrumentation
like
a
full-fledged
kind
of
feature
of
python.
B
We
either
need
to
make
it
in
like
log
step
with
their
manual
instrumentation,
so
include
like,
like
testing
end
to
end
for,
like
every
feature
that
we
add
for
both
manual
and
auto
instrumentation,
or
we
kind
of
maintain,
like
a
list
of
capabilities
that
Auto
instrumentation
is
like
limited
in
I.
Think
either
approaches
is
fine,
like
I
think
at
least
we
can
like
have
a
place
to
like
expose
the
customers
or
users
that
like
hey.
B
This
is
like
the
functionality
we
have,
if
you
don't
have
any
code
or
sorry,
if
you
don't
manually
instrument,
so
now
we
can
start
taking
like
Auto
instrumentation,
actually
like
not
like
seriously
but,
like
you
know,
as
like
a
first
level
thing
we
do
have
like
documentation
on
it.
I
think
people
are
like
a
lot
of
people
are
actually
using
it
and
we're
getting
a
lot
of
you
know:
feedback
on
the
GitHub
issues.
B
Regarding,
like
our
you
know,
otlp
collector
examples,
but
it's
not
like
exposed
as
like.
Oh
python,
SDK
also
supports
Auto
instrumentation,
like
yo
right
in
your
face.
Yeah.
B
E
Point
is
like
it's
not
just
that
we
want
features
that
work
in
manual
instrumentation
to
also
be
available
in
automation.
It's
also
like
I
think
we
should
make
sure
that
features
from
manual
instrumentation,
don't
break
Auto
instrumentation
so
like
right.
It
was
that
issue
I
covered
in
pavitica,
where,
like
the
the
way,
the
Django
instrumentation
was
right.
Auto
instrumentation,
if
you
have
it
installed
so
yeah.
G
B
Yeah
so
I
think
yeah
I,
just
just
wanted
to
like
bring
that
up.
I.
Think
more
discussions
definitely
needed.
B
Let
me
let
me
I'm,
probably
gonna,
create
an
issue
about
this
to
discuss
this,
but
does
anyone
else
have
any
thoughts
about
it
and
like
is
it?
Is
this
making
sense?
Does
anyone
have
any
confusion
about
like
what
we're
trying
to
do.
G
G
Question
if,
if
I
may
oh
go
ahead
yeah,
how
do
we?
How
do
we
expect
the
majority
of
users
to
sort
of
onboard
to
open,
Telemetry
and
python?
Like
do
we
see
auto
instrumentation,
the
UC
I'm,
not
sure
I'm
we
yet,
but
do
you
see
most
people
wanting
to
start
out
and
just
wrap
it
in
and
auto
instrumentation
script
and
see
what
comes
out
or
that
feels
like
the
the
easiest
on-ramp
right.
B
Oh
yeah,
we
don't
have
any
like
kind
of
explicit
like
usage
data,
but
from
the
kind
of
like
feedback
we're
getting
from
like
GitHub
issues
and,
like
slack,
a
majority
of
people
are
saying,
like
oh
I'm,
trying
to
set
up
open
Telemetry
with
autonomous
rotation
or
I'm
trying
to
use
the
collector.
It's
like
a
number
of
people
have
been
saying
this,
so
yeah
I
think
I
agree
with
like
what
you're
saying
the
primary
like
onboarding
mechanism
that
we've
seen
so
far
is
like
you
know,
people
hit
the
getting
started
page.
G
B
B
Is
it
is
limited
in
if
you
I
guess
like
in
terms
of
like
configuration,
if
you
really
want
to
like
get
something
up
and
running
quickly,
it's
like
the
most
convenient
mechanism,
yeah,
but
yeah
I.
Think
you're
right
in
terms
of
like
it'll
get
you
far
but,
like
you
know,
for
like
the
nitty-gritty,
like
very
specific,
like
let's
say
you
have
like
really
complex
view
configurations
for
your
metrics.
B
That
could
be
only
achieved
with
like
manual
instrumentation,
probably
so
yeah
so,
but
it
is.
It
is
a
Persona
of
python
users
that
is
pretty
significant,
so
I
think
it'd
be
worth
like
taking
Auto
instrumentation
quote
unquote
more
seriously
in
the
terms
that
we
won't.
We
aren't
like
retroactively,
fixing
bugs
and
finding
issues
when
they
come
up
and
more
like
making
sure
that
it's
a
very
kind
of
robust
product
in
itself.
D
Also
I
want
to
add
that,
like
there's
also
usually
like
some
other
personas,
where,
like
people
from
the
platform
teams,
are
operations
teams,
they
do
the
like
they
try
to
provide
like
used
auto
instrumentation
so
that
what
the
value
it
provides
and
then
that's
the
people
to
instrument
their
code
but
yeah.
So
they
are
like
they
start
with
the
auto
instrumentation.
And
if
there
is
something
very
specific,
that's
not
you
know
it
supported
in
our
instrumentation.
D
D
Then
they,
you
know,
yeah
man.
B
Yeah
and
then
like,
we
want
to
be
able
to
address
those
kind
of
personas.
B
C
Sorry,
were
you
saying
something:
oh
yeah,
my
bad
yeah
I
was
just
gonna.
I
was
gonna,
agree,
I,
think
it's
more
important
than
the
amount
of
effort
we're
able
to
put
into
it
right
now.
Also,
if
you
look
at
like
the
Java,
Auto,
instrumentation
I
think
they
take
it.
C
So
I
think
what
I
was
gonna
say
is
I
think
they
target
more
than
just
like
the
person
starting
out
like
there's.
There
are
actually
people
who
want
to
use
Auto,
instrumentation
right,
don't
do
anything
else.
We
also
have
the
open
Telemetry
operator,
which
I
I
tried.
I
tried
it
out
for
Python
and
ran
into
quite
a
few
bugs,
but
it
the
idea
is
that
it
supports
Auto
instrumenting,
your
your
workloads
and
kubernetes.
H
C
I,
don't
know,
would
you
would
you
want
to
summarize
what
you
said
in
like
the
notes
or
an
issue
or
something
and
yeah.
D
I
I
On,
a
related
note
I,
want
to
add
that,
like
there
is
also
a
small
team
people
who
are
working
on
the
the
file
based
configuration
like
because
not
everything
can
be
configured
today
and
we
write
so
there's
also
an
ongoing
effort.
You
know
like
trying
to
come
up
with
a
file
based
configuration
for
the
SDK,
which
can
you
know
like
robust
enough
to
support.
D
You
know
all
your
use
cases
if
you
guys
like,
if
any
one
of
our
instruct
interested
you
can
also
join
I,
think
I,
I
I
know
the
repo,
but
I
don't
know
the
like.
If
that's
yeah,
this
I
can
add
to
the
meeting
notes.
B
D
Let
me
add
to
the
yeah
like
the
file
based
configuration
by
the
SDK.
If,
if
you're
interested
add
your
suggestions,
there.
F
D
F
D
Yeah,
so
the
currently
the
focus
is:
what
should
be
the
schema
for
the
like.
You
know
file
like
when
you
say
like
you
can
go
with.
You
know:
ml
Json,
you
know
Proto
different.
You
can
choose
different
ways
of
representing
it
and
then
within.
If
you
pick
one
of
the
cell
ml
configuration,
not
Json
configuration
schema,
how
do
you
want
to
like
there's
already
a
collector
configuration
file
format
how
the
SDK
configuration
be?
It
should
be
similar
so
that
people
don't
get
confused
or
should
be
different,
there's
an
ongoing
work.
D
You
can
participate
and
you
can
add
your
inputs.
There.
F
D
Yeah
I
I
think
some
people
are
from.
This
is
called
there
already
I'm
I,
don't
remember
the
names
but
yeah
you
can
there's.
C
C
Yeah
I
added
this
one
I
I
just
was
looking
at
some
of
the
backlog
and
this
I
think
Diego
may
have
been
taking
point
out
because
he
may
have
I
think
he
regenerated
the
protocols
for
the
version
that
might
work
for
both
3.x
and
4.x,
but
this
is
affecting
a
lot
of
people,
so
I
think
it's
worth
raising
here.
It's
got
17
yeah.
H
Yeah,
that's
a
that's
a
great
point.
I.
H
Yeah
sorry,
yes,.
A
H
I
tried
fixing
this
issue
opened
the
pr
and
asked
for
the
scissors
to
give
it
a
try.
They
replied,
but
I
just
haven't
had
time
to
take
a
look
into
how
this
ended
up
so
yeah
I
have,
depending.
C
Yeah
I
do
I
think
your
fix,
along
with
this
last
comment,
to
loosen
the
dependency
restriction,
I
think
or
maybe
there's
a
later
comment,
but
I
think
those
two
things
might
work
but
I'm
not
too
familiar
with
the
issue.
There's
also
added
in
the
notes,
but
there's
a
PR
and
a
separate
issue
that
says:
try
to
support
3.x
and
4.x
at
the
same
time
and
somebody
opened
a
PR
that
basically
generates
code
for
both
versions,
which
is,
is
pretty
cool
but
I,
don't
know
about
the
the
maintenance
burden
of
that.
H
H
Just
a
little
bit
up
until
you
find
the
command
of
mine,
no
yeah
scroll
down
scroll
down
scroll
down.
A
H
C
H
The
one
right
there
from
psegen,
if
you
scroll
down
a
little
bit,
who
says
that
then
it
won't
be
necessary
to
do
the.
If
else
yeah.
C
Yeah
exactly
so,
I
commented
that,
on
the
other
on
the
pr
with
the
other
fix,
I
I
mean
I
personally
hope
that
this
this
one
works,
because
it's
a
lot
easier
to
just
you
know,
keep
our
stuff
the
same
and
just
pin
a
version
for
the.
C
Then
it
would
work
for
both
through
X
and
4.
portable
version,
so
I
think
this
is.
This
is
a
pretty
important
issue,
because
you
know
protobuf
is
like
a
you
know:
python
doesn't
allow
multiple
versions,
supporting
multiple
versions
of
packages
and
protobuf
is
a
very
popular
Library.
So.
H
Yeah
I
I
agree.
That's
a
very
important
issue.
I'll
try
to
take
a
look
into
this
as
as
soon
as
you
can,
hopefully
then
at
the
beginning
of
next
week,
but
if
someone
else
can
take
a
look
into
it
sooner
or
find
out
solution
faster,
that's
fantastically!
Please
go
ahead.
C
D
D
There's
a
lot
lot
of
activity.
We
should
address
this.
C
H
Oh
yeah,
that
one's
fine-
okay,
that's
a
huge
PR,
but
it's
pretty
straightforward
so
that
Library
Factory
resources
has
been
deprecated,
it's
slow
and
there's.
There
are
other
things
now,
the
pr
itself,
it's
quite
early,
because
it
has
a
lot
of
ifs
and
else's
because
for
3.7
we
need
to
do
something
for
3.9.
We
need
to
do
something
else
and
for
3.10
we
need
to
do
something
else
too.
There
was
a
breaking
change
in
the
API
of
the
import
Library.
H
But
besides
that,
it's
pretty
straightforward,
it
just
replaces
package
be
package
Resources
with
with
something
else
which
is
the
importantly
library.
Now
there
is
not
an
equivalent
one-to-one
replacement
from
the
functional
impacted
resources
to
the
important
in
portlib.
We
kind
of
need
to
re-implement
the
to
get
the
same
functionality,
but
it's
pretty
much
there.
So
you'll
find
it.
That
is
a
very
big
lots
of
Facebook
analysis,
but
it
pretty
much
repeats
itself
all
over
the
place.
C
H
Some
one
question
that
I
wanted
to
ask
Aaron
Aaron
when
what's
the
number
of
this
PR.
H
H
H
When
I
was
removing
all
these
package
resources
I
noticed
that
there
was
a
a
comment
of
yours
in
dogs
requirements.
Txt
that
says
that
we
need
to
install
the
API
and
the
SDK.
H
Can
you
go
to
the
to
the
main
repo
and
look
into
the
this
file
name?
Docs
requirements?
Exe
it's
right
there,
the
top
folder.
A
H
This
is
that's
the
only
thing
that
I
could
not
replace
docs
requirements,
txt
yeah
right
there,
so
it
says
there
and
I
think
that
comment
is
yours.
It
says
something
that
we
need
to
modify
this
path,
because
that
doesn't
work
for
practice
resources,
so
I
I,
guess
that
those
lines
9
10
and
11
are
something
that
are
related
to
some
work
around
because
packet
resources
didn't
work.
So
maybe
now
that
we
could
get
rid
of
practice
resources,
we
may
not
need
this
work
around
I
tried
removing
this
from
this
file.
H
Try
adding
it
there
in
the
in
the
doc
configuration
file,
didn't
work,
couldn't
reveal
the
dots
environment,
so
I
just
left
it
there,
but
I
I
just
wanted
to
bring
it
to
your
attention.
In
case
you
have
any
more
information
about
what
we
can
do
with
this.
If
not
I
guess
you
can
say
it
should
not
be
a
big
deal.
A
H
Okay,
thanks
all
right
so
yeah,
if
you
can
just
take
a
look
at
it,
it's
very
big
but
you'll
see
that
it's
pretty
much
doing
the
same
thing
over
and
over
again.
Thank.
A
B
Hey
Diego:
do
you
think
you
would
be
able
to
kind
of
summarize
the
version
differences
and
what
you've
did
to
address
them
in
the
description
yeah
I
was
saying
how
like
for
different
versions
of
python.
You
had
to
do
kind
of
different
things.
H
Yeah,
let's
I
can
add
that
into
the
the
description.
But
if
you
take
a
look
at
the
there
you'll
see
a
lot
of
fixed
me
comments
right.
That
said
before
307
disrupt
okay,
then
remove
support
when
support.
First,
we
had
another
3.97.
You
can
just
stop
there.
Please
so
scroll
two
lines
up,
please
just
a
little
bit:
no
up
up.
None
of
the
scroll!
Okay,
sorry,
just
scroll
down
yeah
just
show
the
the
the
green
part
that.
H
Just
show
us
the
line:
142,
please,
okay,
right
there!
Okay,
thank
you!
So
it's
it's
very
easy.
So
there
is
no
importantly
library
for
307,
so
it
needs
an
additional
Library
that
is
being
added
in
the
environment
txt
file.
Now
that
library
has
a
different
API
for
the,
how
you
get
the
entry
points
in
one
of
them
you
they
give
you
a
dictionary,
but
when
you're,
using
3.9
or
less
you're,
not
getting
a
dictionary
you're
getting
something
else,
and
when
you're
using
310,
you
get
something
even
different
right.
H
So
all
this
happens
because
there
was
the
way
you
get.
The
entry
points
is
different,
but
at
the
end
it's
the
same
thing.
You
get
the
same
entry
points.
It's
just
that
for
one
version
you
need
to
query
a
dictionary
and
in
your
case
you
need
a
like
a
list
of
tuples
or
something
like
that.
B
Right
right,
that
makes
sense.
I'm
I'm
just
wondering
if
you
can
add
that
as
to
the
description,
because.
H
Like
hey,
I
will
add
there
too
I
see
that
Aaron
is
mentioned
and
that
maybe
you
can
wrap
this
logic
in
a
private
function.
I
think
we
can
I
just
think
that
it
make.
H
It
makes
sense
to
leave
it
like
this
because
3.7
it's
something
that
we
will
stop
supporting
in
a
year
or
so
and
I
guess
that
when
that
time
comes,
you
can
just
delete
those
lines
right.
D
D
H
Can
do
that?
On
the
other
hand,
there's
a
lot
of
changes
in
the
in
most
of
the
changes
are
in
the
tests,
because
the
tests
need
to
be
needed
to
be
modified
and
I
don't
know
if
we
will
be
able
to
find
a
function
that
perhaps
all
this
code
in
tests
I
I'll
try
to
refactor
this
to
repeat
the
smallest
amount
of
code,
but
I.
Imagine
that
there
will
be
a
lot
of
repeated
coding
tests,
but
but.
D
All
right,
I
think
we
don't
have
any
more
PRS.
H
Last
time,
I
I
have
the
expression,
histogram,
PR,
open,
I,
know
it's
a
big
thing
and
it's
complicated
and
everything
else.
We
just
got
a
review
from
Josh
McDonald,
one
of
the
metrics
experts,
but
if
someone
can
take
a
look
at
it,
I'll
be
very
thankful
if
there's
a
PR
that
you
need
me
to
review
our
happily
Trader,
probably
so
yeah,
because
I
know
that
that
PR
is
going
to
need
a
little
bit
of
help.
Since
it's
a
big
and
kind
of
complicated
topic,
but.
H
You
can
please
take
a
look.
Ask
me
any
questions
if
you
okay,
thank
you.
D
All
right
thanks,
everyone
take
back
your
12
minutes.