►
From YouTube: Meshery Development Meeting (June 29th, 2022)
Description
Meshery Development Meeting (June 29th, 2022)
Agenda:
[Aditya.C] Catalog updates.
[Ashish] Importing helm charts as Meshery application
Request: add column to the “Applications" table that signifies the original form of the App (Docker Compose, K8s Manifest, Helm Chart).
Ask: publish the Istio+SPIRE pattern into the Meshery Catalog (https://meshery.io/catalog)
[Ashish] Istio+Spire design demo and few bugs to point
[Harkirat]docker extension empty state fixed. #5684
[Aaditya Narayan Subedy] SMP Dashboard Updates.
Request: Make "Test Specification" and “Environment Details" feel like tabs. Bring
[Yash] Meshery v0.7.0 Planning
Refactoring Meshery UI: meshery-ui-restructuring | How to contribute
Updates made on the branch.
[Nithish] Refactoring Meshery Controller Handlers and GraphQL susbcriptions
Join the community at https://layer5.io/community
Find Meshery on:
GitHub: https://github.com/meshery
Twitter: https://twitter.com/mesheryio
LinkedIn: https://www.linkedin.com/showcase/meshery/
Docker Hub: https://hub.docker.com/u/layer5/
A
A
A
We
did,
I
did
some
code
refactoring
here
there
was
a
lot
of
redundant
code.
There
still
is
a
lot
of
redundant
code.
There's
an
open
effort,
I
think
other
than
our
ranks
is
working
on
it
to
refactor
some
more
code
over
here.
So
now,
we've
moved
on
to
individual
pages
so
that
it's
easier
for
us
to
share
these
patterns
with
people.
A
Now
we
have
chips
which
also
do
filtering.
So
with
these
chips,
you
can
filter
these
go
back
to
the
and
these
chips
are
also
clickable.
These
are
some
of
the
changes
that
I
made
other
than
that
in
the
last
call,
we
had
decided
to
add
the
date
when
it
is
published
on
and
the
auth
and
the
author's
name
right
now.
All
the
patterns
are
published
by
us,
so
the
author
name
is
leah5
and
it
appears
that
the
date
is
not
working.
The
date
gets
updated
every
day.
So
that's
a
bug
over
here.
A
If
anyone
wants
to
fix
that
or
I'll
fix
it
myself
so
yeah,
these
are
just
some
of
the
updates
other
than
that.
Aditya
is
working
on
a
couple
of
more
things,
also
yeah,
also
in
mobile
view.
The
filters
were
not
visible,
so
I
added
that
in
as
well
so
yeah.
These
are
just
small
updates,
any
thoughts,
any
feedback.
C
E
F
A
C
C
C
C
I
wanted
just
to
make
sure
that
everyone
had
context
for
what's
going
on
here
and
so
you
know,
mesherie
is
available
as
a
set
of
containers
to
deploy.
There's
a
server
any
number
of
service
mesh,
specific
adapters.
There's
a
cli
there's
a
few
components.
There's
we
should
probably
share
a
link
to
meshrey's
architecture.
C
C
C
I
suspect
not
because
that's
kind
of
an
unfair
question,
because
that
wasn't
part
of
the
issue
description,
but
it's
just
feedback
for
you
to
think
about.
If
that's
an
easy
enough
check
to
make
that,
then
you
could
visually
indicate
in
the
ui
that
a
particular
adapter
is
unavailable
is
like
disabled.
C
D
Okay,
assuming
I'm
audible
so
and
also
you
can
see
the
right
screen
so
you're,
seeing
the
application
space
right,
yep.
Okay,
so
I
have
two
items
on
my
gender.
One
of
them
is
the
additional
features
that
I've
added
that
is
importing
helm
charts
as
much
the
applications.
So
just
to
give
a
little
bit
of
context
we
have
mischief.
Applications
are,
is
an
umbrella
term
for
your
communities,
manifest
helm,
chart
and
their
patent
files
and
docker
compose,
so
anything
that
can
describe
your
enterprise
covered
under
application
and
slowly
we
are
expanding
upon.
D
What
are
the?
What
are
the
things
that
we
support
so
currently
I've
added
this.
There
are
some
ui
considerations
which
I
will
talk
about,
which
we
can
talk
about
after
I
demo
the
thing,
but
let
me
give
you
a
demo
so
when
you
import
an
application
currently,
it
allows
you
to
import
a
file
that
can
be
a
remote
file
that
can
be
linked
to
a
github
repo,
a
particular
file,
and
you
and
or
you
can
pass
a
file
from
your
local
system.
D
So
there
are
two
options
available
right
now
and
currently
I
will
demo
by
passing
the
the
download
url
of
the
tar
file
of
that
particular
helm
chart
also,
I
have
added
the
feature
of
basically
importing
your
local
helm
repo,
but
how
that
looks
on
the
ui.
We
can
discuss
it
after
this,
so
just
for
example.
This
is
this:
is
the
prometheus
community
index.camel
of
their
helm
chart,
so
I
have
taken
this
one
alert
manager,
for
example,
and
I'll
try
to
import
this.
D
Basically,
this
is
the
same
link
with
which
you
would
download
and
charge
so
after
that
is
successfully
imported.
You
can
see
this
thing
comes
up
over
here,
and
this
is
nothing,
but
this
gets
converted
into
your
regular
gates
manifest.
So
you
have
your
service,
your
deployment
and
your
service
account,
etc.
Anything
that
is
required
for
deploying
this
alert
manager.
D
Okay,
now
one
other
ui
bug
kind
of
thing
I
wanted
to
point
out
is
okay,
so
first
thing
is
that
when
I'll
deploy
a
model
would
not
come
up
on
the
contrary,
when
I
would
click
deploy
in
the
other
demo
that
I'm
going
to
do
right
now
inside
of
the
meshmap
designer,
the
model
would
come
up.
So
that
is
one
thing
to
note.
Second
thing
is
that,
even
though
the
deployment
would
be
successful,
I
would
not
get
the
snag
bar
over
here.
D
I
won't
get
any
error,
but
still
I
won't
get
this
successful
snack
bar,
so
that
is
two
things
to
keep
in
mind,
so,
like
click
on
deploy
and
as
it
gets
successfully
deployed.
Okay,
so
again
I
did
not
get
a
snack,
but
it
should
have
said
application
successfully
deployed,
so
I've
deployed
my
alert
manager.
Now,
if
I
go
to
my
mesh
map
visualizer,
I
should
be
able
to
see
my
prometheus
alert
manager
so
opening
up
the
visualizer.
If
I
filter
out
my
default
namespace
here,
you
can
see
this.
D
This
is
the
alert
manager
that
I
just
deployed
using
my
chart.
So
I
have,
I
just
use
the
simplest
example,
so
the
that
was
the
first
demo
that
I
wanted
to
do,
and
you
know
simply
you
can
once
you
have
imported
that
as
an
application.
You
can
also
undeploy
that,
so
I
think
first,
we
can
discuss
upon
this
thing
then
upon
the
ui
of
importing
a
local
helicopter,
and
then
I
can
move
on
to
the
my
other
topic,
which
is
the
skew-inspired
design
so
yeah.
I
can
undeploy
like
this.
D
Also
so
yeah
there's
a
model
needed
here
as
well
as
so
so,
for
example,
right
now,
when
I
click
on
import
application,
when,
if
I
upload
a
file,
this
is
not
like
a
job
of
back
end.
Ui
goes
and
downloads
that
file
from
your
local
system
and
that
convert
basically
sends
the
text
the
it
in
plain
text
over
to
the
back
end,
so
that
that
works
with
the
files.
But
when
you
talk
about
directories,
you
cannot
send
it
in
a
text.
D
You
have
to
give
the
path,
and
the
back
end
has
to
have
the
logic
to
ma
to.
You
know
manage
that
path
or
basically
to
use
that
path.
In
order
to
do
its
stuff,
so
how
do
we
want
to
do
that?
How
do
we
want
to
you
know,
allow
users
to
give
the
path
to
their
local
helm
depot
here,
or
do
we
want
to
use
the
same
url
page
just
using
the
file
poll
and
colon
like
this
one,
so
that
do
you
want
to
use
this
method
or
do
we
want
to
have
something
else?
D
I
think
if
it's,
if
it
is
something
related
to
file,
then
of
course
the
the
file
manager
you
know,
is
better
yeah
yeah,
but
I
meant
by
that
that
you
know
I
passed
the
index
dot,
the
basically
location
of
index.aml,
and,
although
that
that
also
doesn't
make
sense
right
now
so
yeah
it
should
be
a
file.
It
should
be
a
folder,
so
it
shouldn't
be
a
file
basically
a
directory.
I
have
to
give
the
path
to
a
directory
to
backend
and
the
logic
that
we
have
right
now.
D
In
back
end,
it
expects
you
to
give
that
path
inside
of
the
path
we
already
have
that
path
key.
So
instead
of
the
path
key,
if
you
just
give
slash
homes
that
something's
not
something,
it
would
go
ahead
and
import
that
from
there
but
yeah
right
now,
we're
not
using
path
field
here
either
we
are
directly.
You
know
when
I
click
on
upload
file.
We
are
directly
extracting
the
manifest
and
sending
it
over
or
we
are
using
a
remote
url.
So
for
what
should
be
a
better
user
experience.
D
D
D
D
A
D
Yeah,
when
you,
you
know
upload
something
from
the
file,
then
it
keeps
you
the
content
inside
by
like
send
the
content
inside
file
to
the
back
end,
yeah
yeah,
that
that
I
already
know
what
I'm
asking
is
that
inside
of
this
file,
name
dialog
box,
do
we
give
the
explicit
path
or
what
is
what?
What
does
this
take
as
an
input?
I
don't
think
it
takes
anything.
Even
if
I
write
anything
to
it,
I
still
have
to
click
on
browse,
so
so,
basically,.
D
D
D
You
know
yeah
just
like
this.
You
can't
open
it.
When
you
will
be,
you
know,
providing
an
extra
parameter
that
can
accept
the
directory
name,
so
you
will
be
able
to
open
any
directory
from
from
that
button,
and
I
I
hope,
like
I
I
haven't
seen
yet,
but
I
hope
it
will
return
back
the
correct
file
location
that
user
just
chose
so
okay,
I
think
so
you
you
just
need
that
right,
yeah
yeah,
the
location
of
the
directory
itself,
so
we
can
put
a
pin
on
that.
D
So
basically
the
backend
supports
it.
I've
demonstrated
from
the
ur
item,
and
this
can
be
a
separate
issue
into
there.
Basically,
two
issues,
one
is
the
enhancement
of
this
particular
model
and
the
other
one
is
the.
When
you
click
on
deploy,
you
should
get
the
same
model
as
you
get
when
you,
when
you
are
on
designer,
so
okay,
so
that
was
importing
your
helm
charts
onto
this
also
you
can
import
docker,
compose
and
other
things.
D
Okay-
and
this
is
on
the
call-
so
there
are
a
few
issues
right
now-
very
minor
issues
that
can
be
that
I
have
made
a
kind
of
hack
over
to
make
the
design
work,
but
you
know
that
so
we
would
we
would
be.
Basically
we
can
fix
that
thing
before
the
weekend.
The
the
there
is
some
problem
with
custom
resource
definition
on
so
I
have
created
a
basic
design.
This
is
the
somehow
I
designed
it
so
recently
it
didn't
save.
So
that's
one
of
the
bug
results.
D
Sometimes
I
design
stuff-
and
I
expect
it
to
save
but
like,
for
example,
it
says
saving.
So
if
I
come
back
to
it
yeah
this
it's
it's
kind
of
in
a
stale
stage:
okay,
whatever
I
can
prettify
it
using
this
thing,
using
let's,
let's
okay,
so
this
is
my
spire
design,
and
I
have
again
I
have
created
this
using
just
by
importing
the
aspire
yaml
that
they
provide
in
their
demo
that
the
basically
this
is
the
quickstart.yaml
you
can
find
on
their
website.
D
So
if
I
wait
a
second,
the
zoom
chart
thing
is
coming
in
front
of
my
web
button,
so
hold
on
a
second
yeah,
so
yeah.
So
there
would
be
a
false
negative
over
here.
The
the
basically
a
snack
bar
would
come.
That
would
say,
there's
some
tcp
connection
error.
I
would
look
into
that.
That's
not
an
error
that
shouldn't
come
but
bear
with
me
for
when
that
happens,.
D
D
And
if
I
go
to
my
spire
namespace
yeah,
so
the
workload
registrar
is
the
spire
server
and
the
other
stuff
that
comes
up
with
the
the
basic
quick
start:
speed,
spires
theo
design,
this
stuff
got
deployed
in
my
aspire
game
space
along
with
things
like
you
know,
demon
set
and
your
config
maps
that
are
required
to
for
different
purposes,
yeah
so
and
you
can.
Similarly,
you
can
undeploy
it
using
you.
Can
you
basically
have
the
same
design?
D
You
know
you
can
undeploy
it,
and
I
again
you'll
get
that,
but
it
falls
error
over
here,
but
again
ignore
that
for
a
while,
so
stuff
is
still
getting
slowly.
Yeah
name
space
gets
destroyed
in
the
last
year,
so
the
name
space
is
no
more
as
well
so
stuff
gotten
deployed.
So
you
can
use
this
right
now.
D
G
C
I
think
I
think
I'm
so
excited
that
I'm
stumbling
over
my
words,
this
is
great.
Do
you
mind
recapping,
for
my
benefit,
for
everybody's
benefit,
the
some
of
how
these
same
capabilities
so
you're,
using
the
ui
to
import
a
home
chart,
get
it
converted
to
a
mesry
app
then
be
able
to
visually
diagram.
It
make
some
changes,
understand
how
it's
you
know
the
particulars
of
it.
C
C
D
Yeah,
so
the
app
on
board
works
just
fine
and
there
are
some
bugs
over
there,
which
is
which
will
be
sold.
There
is
an
open
pair
by
uttarats
that
actually
solves
a
lot
of
stuff
there,
and
the
pattern
apply,
works
as
usual,
of
course,
but
yeah
the
the
app
on
board,
the
bugs
in
the
app
on
onboard
that
even
if
they
were
spacing
like
you,
tried
to
import
some
manifest
right.
So
it
doesn't
matter
if
you
like.
D
C
Nice,
so
do
you
recall,
is
it
a
two-step
procedure?
So
if
let's
say
so,
we've
got
a
helm.
Chart
we'd
like
to
use
the
cli
to
ingest
the
you
know
to
basically
programmatically
or
someone
might
want
to
anyone
can
go
use
the
rest
api
to
perform
the
same
actions
that
the
the
ui
just
performed
good.
So
if
someone
wants
to
write
a
script
or
use
their
favorite
programming
language
to
do
that
they
can
or
they
could
use
the
cli
mastery
ctl
to
which
is
also
using
the
rest
api.
D
D
Okay,
great
perfect
cool,
if
you,
if
you,
if
you
like,
if
you
say
messy
city
app
on
board
and
if
you
give
it
a
specific
name,
for
example,
if
right
now,
if
you'll
do
message,
it'll
app
onboard
test3.eml,
some
name
that
is
already
there
in
the
provider
that
you
are
using,
then
it
would
use
that
particular
file.
Otherwise,
if
you
give
it
a
path
it
would
it
would
whether
it's
a
remote
path
or
a
local
path.
It
would
download
it
on
basically
import
it
here
and
then
try
to
deploy
that
so
yeah.
G
Yeah,
seeing
this
so
sorry
application
more
guidance
on
how
to
how
to
use
patterns.
Let's
say:
let's
say
you
have
you
have
a
an
issue
deployment
already
in
place
and
you
have
to
you
only
need
to
configure
it
to
add
spire
or
the
other
way
around.
You
already
have
spire
and
you
want
to
deploy
an
issue
and
have
them
consuming
spider
identities.
G
D
Yeah,
actually,
it
would
be
recommended
to
for
it
to
be
a
single
design.
So,
for
example,
if
if
it's
the
part
of
the
same
design,
then
for
example
right
now,
if
I
go
over
to
the
design
configurator
you
we
already
whatever
services
that
you
have
inside
of
your
design,
you
can
add
extra
stuff
onto
that.
So,
instead
of
designer,
if
I
open
the
same
design
that
I
had
opened,
you
can
right
now
you
you
won't
be
seeing
some
sql
components
you
could
have,
because
I
don't
have
the
adapter.
D
But
since
you
know
we
on
the
designer
we
do
a
lot
of
referential
stuff
and
in
a
single
design
one
thing
can
reference
another,
and
so
we
are
going
to
use
state
machines
in
order
to
actually,
basically
you
can
some
particular
field
of
one
service
can
be
referred
back
to
another
service.
So
if
you
have
aspiring
history
deployment,
that
can
be
the
case
that
maybe
let's
say
maybe
so,
while
configuring
your
skew,
you
need
to
give
it
something
that
is
dependent
upon
something
that
you
had
already
configured
in
spire.
C
C
C
C
Or
yeah,
or
maybe
they've
got
an
existing
deployment
of
and
they've
got
services
running
they're
going
to
stand
up
another
instance
or
they're
going
to
replicate
that
play
with
it.
They
want
to
retroactively,
apply
it.
They
know
that
they've
got
to
you,
know,
cycle
identities
and
stuff,
but
and
so
yeah.
It
would
be
fantastic
to
have
ultimately
kind
of
two
artifacts
two
two
patterns,
one
full
istio
plus
spire,
another
one,
just
just
the
spire
portion
of
the
the
config
to
to
apply
to
it.
D
Also
inside
of
patterns,
we
have
a
feature
of
depends
on
which,
which
actually
helps
to
create
a
directed
acyclic
graph
as
the
time
of
your
evaluation.
So,
for
example,
right
now,
while
I'm
going
over
this
particular
installation
page,
it
says
that
okay,
so
after
you
have
deployed
aspire
and
everything
is
in
ready
state,
then
you
do
some
this
extra
stuff.
So
in
this
particular
case,
what
you
can
do
is
you
can
combine
all
your
on
the
designer
so
I'll
come
on
the
design
just
to
just
so
that
you
can
have
reference
so
on
the
designer.
D
What
you
can
do
is
all
your
spire
related
stuff
that
needs
to
first
be
in
the
ready
state.
You
can
combine
that
into
a
single
single
yeah
so,
like
you
can
have
that
into
a
single
node
group
and
basically
so
that
you
can
package
all
these
things
together
and
then,
after
that,
let's
say
that
the
documentation
says
you
have
to
deploy
an
sd
operator.
You
can
go
over
to
your
basically,
you
can
go
over
here.
Click
on
sto
select
this
key
operator
and
then
using
the
depends
on
field.
D
You
will
be
able
to
make
sure
that
first
that
node
everything
in
that
node
group
gets
deployed
and
after
that
is
deployed,
then
your
other
stuff
is
deployed.
So
basically
you
can
make
relations
between
separate
stuff
into
the
same
thing
or
you
can
first
deploy
this
thing
and
go
to
the
visualizer
and
check
when
the
things
are
ready
and
then
come
over
here
again:
click
on
skew
which
it's
not
here,
because
again
I
don't
have
the
adapter
right
now,
but
again
you
can
come
over
here
and
then
continue
with
the
process.
D
C
So
that's
kind
of
an
interesting
journey
that
helms
you
start
with
a
helm
chart
in
this
case.
Like
you
start
with
a
helm,
chart
you
import
that
to
become
a
meshery
app,
so
it's
just
kind
of
a
verbatim,
not
a
verbatim,
but
it's
a
direct
translation
from
helm
to
mesri
app
from
there.
You
can
just
you
know,
deploy
to
kubernetes.
C
C
C
It's
like
okay,
you
know
great
what
if
I
want
to
go
back
to
home
like
for
whatever
reason
I
you
know,
I
I
use
the
visualizer
to
do
the
hard
stuff
of
like
redesigning
reconfiguring
something
and
then
but
my
ci
process
does
something
else
or
for
whatever
reason
you
need
your
home
chart
back.
You
want
to
update.
You
want
to
make
an
update
on
artifact
hub.io.
C
C
I
think
the
re-export
back
to
helm
chart
it
would
be
a
feature
request
and
it
would
kind
of
come
with
some
caveats
that
you
might
expect
if
you're
using
meshri
to
if
there's
anything,
you've
included
in
your
design
that
helm
couldn't
describe
or
that
you
couldn't
describe
in
a
chart,
then
there
might
be
an
issue,
but
anyway,
some
kind
of
an
interesting
thing.
Kind
of
worth.
Exploring
one
thing
that
I
want
to
say
out
loud
is
in
the
applications
table,
at
least
for
my
part,
I'd
be
curious
for
feedback
from
everyone
else.
C
A
E
E
And
another
thing
I'd
configure
like
updated
was
to
get
the
responsiveness
of
the
sidebar
like
if
we
move
the
screen.
Smaller,
the
sidebar
automatically
hide
itself,
and
the
last
thing
I
did
was
to
automatically
load
all
the
like.
The
test
results
and
the
like
the
environment,
details
and
test
specification
on
load
of.
When
we
show
the
chart
like
earlier,
we
was
required
to
click
the
buttons.
Then
this
test,
specification
or
environment
will
show
up,
but
now,
as
the
page
load
up,
we
will,
we
will
come
with
test
specification
and
environment
details.
D
D
No,
no,
let's
take
someone
else's
opinion
on
it,
so
that,
whether
the
tab,
you
know
this
tab
thing
should
be
there
or
everything
is
visible.
On
the
same
page,
I
voted
for
everything
because
there
is
not
only
one
test
at
the
at
one
time:
the
time
when
the
tabs
were
implemented,
it
was
because
you
know
we
were
loading,
six
seven
tests
at
the
same
time
and
showing
results
of
all
of
them
at
the
same
time.
So
we
want
to
just
show
smaller
information.
C
I'd
love
to
hear
from
anyone
else.
My
vote
would
be
for
tabs.
I
mean
you
could
optionally
make
it
like.
Actually,
let's
not
even
bother
like
yeah,
just
just
tabs
like
I
was
gonna
say
you
could
optionally
make
it
so
that
they
can
show
all
of
it
at
one
time
but
like
we
need
to
work
on
the
actual
tests
and
different
types.
We
need
to
work
on
other
things
other
than.
D
E
C
Yeah
actually
I'll
say
this
since
nadalia
and
maxi
are
here
and
since
we're
on
the
cusp
of
having
a
convenient
way
to
deploy
istio
with
aspire
configuration
is,
I
think
that
makes
a
perfect.
You
know
unless
others
see
it
differently.
I
think
that
makes
a
perfect
deployment
scenario
to
have
here
and
start
to
have
those
results
from
the
cncf
labs
published.
G
C
Different
deployments
of
different
applications,
different
different
service
meshes
under
different,
different
apps
and
then
different
performance
tests,
and
what
they've
done
is
so
we've
gotten
access
to
the
cncf
labs,
which
is
an
equinix
data
center
somewhere
and
right
now.
I
think
it's
just
one,
but
we
have
access
to
geographically
disperse
data
centers,
so
they've
got
a
couple
of
so
so
what
they've
done
is
they've
taken
and
they're
they're
using
that
github
action
as
a
self-hosted
runner.
C
C
So
all
that
we
would
need
to
do
to
have
if,
if
maxie
or
natalia,
if
you're
running
tests
and
you'd
like
for
them
to
be
shown
here,
pranav
singh
is
on
the
call.
All
we
need
to
do
is
identify
the
user,
the
measure
user,
whose
tests
that
would
like
to
be
you
would
like
to
have
exposed
and
and
then
you
can
self-report
in
that
way.
G
C
F
G
C
C
It
I
mean
other
than
like
other
than
the
bare
metal
provisioning
of
the
operating
system
and
then
the
provisioning
of
kubernetes,
which
is
already
you
know,
which
the
thing
that
we
just
covered
with
the
helm
chart
being
imported
as
an
app
like
meshery.
Will
you
know
mesri
does
the
rest
from
there
with
provisioning
the
service.
H
C
You
just
have
to
stub
your
stub,
your
toe
on
a
couple
of
bugs
you
know
squash
them
as
you
go
and
then
wipe
off
your
shoe,
keep
going,
but
but
yeah,
no!
No,
but
in
all
seriousness
that's
exactly.
This
has
been
like
a
goal
for
the
last
year
and
a
half
is
to
be
able
to
get
to
a
point
where
this
conversation
is
a
real
conversation
and
and
the
plumbing
is
all
there
and
yeah
we.
G
I
So,
like
I
hope
my
screen
is
visible,
so
most
of
you
like
would
know
that
we
are
restructuring
machine
ui
for
incorporating
redux
and
yeah
and
and
ui5
so
like.
If
anyone
want
to
get
started,
here's
some
links,
you
would
love
to
follow
and
updates
on
the
restructuring
branch
where
we
are
now
so
I
made
I
made
some
changes
on
the
dashboard.
I
E
D
Yeah,
while
you're
doing
refactoring
those
things
and
creating
a
new
ui,
make
sure
that
you
are
using
just
functional
components
wherever
possible
and
secondly,
make
smaller
components
like
it
is.
If
you
are
able
to,
you
know,
make
smaller
components
like.
However,
the
number
goes
it's
it's
great
because
you
can
reuse
it
in
future.
I
mean
yeah,
it
may
take
time
and
bit
of
focus
to
do
it,
but
yeah.
That's
the
beauty.
I
B
D
D
I
think
make
style
also
lost
its
priority
in
mui5.
So
when
you
have
a
style
api
as
well
as
make
sense
and
having
the
same
styles,
then
style
api
will
only
get
applied
and
make
styles
will
get
rejected.
So
just
to
get
away
of
the
future
box
from
mua,
v5
or
v6
v7.
So
since
make
sense,
is
deprecating,
so
we
should
not
be
using
it.
It's
just
one
comment
that
I
want
to.
You
know
make.
B
A
A
A
A
So
if
you
want
to
look
at
like
where,
where
a
specific
logic
for
a
particular
thing
is,
then
if
someone
is
new
to
the
code
base,
it
would
be
it'd,
be
very
difficult
for
him
to
navigate
through
the
code
base,
because
because
of
the
way
that
it's
being
written
so
and
also
there
were
other
issues
regarding
graphical
subscriptions
where,
after
a
point
like
it
became
very
hard
for
us
to
manage
the
channels
that
we're
using
to
establish
subscriptions.
So
that
also
become
a
bit
of
a
pain.
A
So
we
introduced
a
broadcaster
which
yeah
which
didn't
go
well,
so
that
didn't
actually
solve
the
issue.
So
so
I've
taken
a
a
bit
of
a
different
approach
here,
rather
than
yeah,
rather
than
trying
to
you
know
juggle
with
the
so
many
channels.
At
the
same
time,
so
yeah
I
won't
like,
I
won't
be
able
to
talk
about
every
single
change
that
I've
made
because
of
the
time
constraints.
So
I
encourage
you
guys
to
look
into
this
pr
if
you
are
interested
yeah.
A
This
is
in
this
is
regarding
the
machine,
server,
pr
and
the
other
ones
in
meshkit.
So
this
essentially
the
main
gist
of
it-
is
that,
like
there's
a
aventador
interface
to
handle
the
controllers
to
so,
basically,
you
know
like
we
have
three
main
controllers
in
sorry,
three
controllers
in
measuring
I'm
actually
operator
machine
broker
and
machining.
So,
as
I
said
like,
we
have
been
like
the
logic
for
like
how
to
deploy
an
operator
and
or
how
to
get
the
status
of
an
operator.
A
Those
were
those
are
not
very
transparent,
so
I
have
basically
implemented
interface
over
here,
which
is
which
is
pretty
generic
it.
So
in
the
future
like
there
are
some
considerations,
and
there
are
some
improvements
that
we
can
obviously
do,
but
the
this
is
sort
of
like
the
first
iteration
of
how
we
would
want
this
to
be
so.
A
It
basically
has
three
these
methods
on
them
and
further,
like
we
actually
have
all
the
three
controllers
different
over
here
and
the
logic
for
them,
as
well
as
to
how
to
deploy
them,
how
to
get
the
status
and
all
those
stuff
were
also
defined
here
as
yes,
and,
as
I
said
like
after
after
having
a
discussion
like
we
concluded
that,
like
there
are
some.
Obviously,
this
would
not
be
the
ideal
thing
and
since
like
in
the
future,
we
would
be
moving
it
to
the
measuring
operator,
but
yeah.