►
From YouTube: Meshery Development Meeting (July 13th, 2022)
Description
Meshery Development Meeting (July 13th, 2022)
Agenda:
[Ruturaj] Meshery Cloud seed content table design
[Yash/Gaurav] Meshery v0.7.0 Planning
Refactoring Meshery UI: meshery-ui-restructuring | How to contribute
Updates made on the branch.
[Nithish] Conventions
[Gaurav] Grid and table view for all configurations (Designs/Applications/Filters)
[Gaurav] Progress on support for source-type on Applications UI
[Nishant] mesheryctl release improvements on homebrew
Migrating mesheryctl some custom tap to homebrew-core
Edge releases for mesheryctl
[Uzair] Handle Expired Session #4956
[Alpha] Status of screenshot logic in docs (with sample example) #5778
[Aaditya Subedy] Docs table sorter
[Aditya] Working for mockups for an extension marketplace in the Meshery.
[Aditya] Open “Help Wanted” issues.
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/
B
D
F
F
G
E
H
Yeah,
okay,
so
just
to
give
you
a
context
on
what
this
is
we
discussed
about
yep,
so
we
discussed
about
this
seed
content
and
we
decided
that
we
we
should
rather
call
it
the
mystery
catalog,
which
is
just
basically
a
pre-loaded
set
of
applications
patterns
and
filters
which
all
users
would
be
able
to
access
from
meshi
and
if
they,
if
they
want
to
you,
know
edit
it
they
can
create
a
copy
for
themselves.
They
can
deploy
it
and
you
know
try
out
so
so
this
was
wait.
H
Yeah,
so
this
is
basically
the
design
spec
that
I
updated
today.
So
what
basically
just
use
user
flow
will
just
be?
There
will
be
like
an
option
for
so
a
provider
could
register
that
as
a
capability
that
they
support,
catalogs
and
then
the
user.
Could
click
on
the
catalogs
have
a
look
at
the
catalogs
if
they
want
they
can
make
copy
for
themselves,
and
it
will
show
up
in
their
measuring
applications
or
designs
or
filters,
so
yeah
and
the
database
design
is
still
we're
discussing
on
it.
H
H
Yeah
yeah,
that's
another
consideration
in
the
in
the
design.
Spec
is
how
different
this
will
be
from
service
mesh
catalog.
So
the
one
clear
difference
is
that
that
catalog
is
not
integrated
with
measuring.
So
you
can't
just
you
know
or
mr
cloud
and
first
and
second
of
all,
it's
just
messy
stuff.
I
mean,
if
some
other
provider
wants
to
give
any
other
example
and
other
sample
content.
H
F
Yeah,
that's
a
great
question
to
sean.
If
I
don't,
oh,
let
me
give
a
little
more
color.
If
I
could
there's
a
couple
of
there's
a
couple
of
things
about
measuring
that
are
well
pleasantly
sophisticated.
Maybe
is
the
right
way
of
characterizing
it,
and
that
is
to
say
that
it's
well
that
it's
an
extensible
platform,
so
mesri
has
a
number
of
things
going
on.
F
One
of
the
things
that
it
has
going
on
is
that
it
has
extension
points
strategically
placed
or
not.
It
has
extension
points
throughout
its
architecture.
One
of
the
extension
points
is
something
that
users
get
an
immediate
choice
on
when
they
go
to
deploy,
measuring
and
it's
the
this
concept
of
a
provider.
F
You
know
an
identity
provider
or
a
provider
of
other
things,
a
provider
being
a
provider
being
this
concept
of
an
extension
point
wherein
measuring
that
that
ships
out
of
the
box
it
comes
with
a
couple
of
providers
that
you
can
connect
to.
One
of
them
is
called
none,
and
it's
just
this
basic,
this
basic
remote
provider
or
I'm
sorry,
local
provider.
This
concept,
these
concepts
are
going
to
be
a
little
bit.
F
These
are
these
are
concepts
that
all
of
you
can
understand,
no
doubt,
but
the
way
that
I'm
presenting
them,
perhaps
not
the
easiest
to
grok
at
the
moment.
The
point
is
measuring,
is
a
tool
that
can
have
users
or
not,
if
you
and
you
can
shoot
so,
you
can
choose
to
sign
into
mesri
and
have
a
user
account
or
not.
F
That
choice
is
made
here
as
you
deploy
mesheri.
If
you
say
that
you
don't
want
to
sign
in
to
a
provider
an
identity
provider,
a
user
account
provider,
then
you
just
say
none
and
you're
anonymous.
You
just
use
using
measure
as
a
tool
that
doesn't
necessarily
have
a
memory
per
se,
which
is
to
say
that
it
doesn't.
It
doesn't
guarantee
persistent
storage
of
your
infrastructure.
F
You
kind
of
remember
necessarily
what
what
you
were
doing
and
what's
going
on,
it
might
come
out
of
the
box
with
some
content
like
some
sample
designs,
sample
patterns,
maybe
some
sample
applications
sample
web
assembly
filters
and
that's
great
and
go
off
and
use
measuring,
be
successful.
Add
things?
Do
you
think
you
know
you
use
it
to
do
performance
management?
Do
all
the
things
if
you
use
it
to
do
performance
management,
you
run
a
test
yeah
and
you
don't
have
a
user
account?
F
F
Anybody
can
grab
a
free
account
and
what
you
get
out
of
the
box
is
well,
you
get
some
persistent
memory,
so
there's
you
get
a
user
account
created
and
if
you
run
performance
tests,
for
example,
this
particular
account
has
run
53
tests,
and
so
they
can
come
back
and
look
at
those
and
look
at
the
results
and
there's
depending
upon
what
provider
you
sign
into
they
might
have.
They
might
offer
different
things,
one
of
those
being
persistence
of
performance
tests,
persistence
of
conformance
tests,
these
or
other
things.
F
Things
like
like
mesh
map
is
a
is
an
extension
that
we
talk
about
fairly
often,
and
so
that's
in
part
what
root
raj
is
alluding
to.
Is
this
concept
of
the
users,
so
this
user
that
just
signed
in
they
have
a
couple
of
other?
They
have
more
content,
that's
been
saved
under
their
account
because
they
signed
up
for
an
account.
So
they
get
to.
You
know,
create
designs
and
save
their
designs.
F
They
get
to
maybe
add
their
own
filters
or
add
their
own
apps
and
have
those
saved,
and
so
that's
what
raj
is
referring
to.
What
he's,
and
sometimes
it's
not
even
necessary
to
make
that
clarify
that
that
distinction,
because
it's
just
kind
of
like
most
people
choose
to
sign
in
and
when
they
do,
they
get
access
to
sample
content.
F
There's
an
initiative
that
nashant
was
just
highlighting
that
a
number
of
you
have
worked
on
and
are
actively
working
on.
This
is
in
like
a
beta
phase,
probably
about
where
it
is,
which
is
that
some
of
you
are
helping,
publish
best
practices
or
common
patterns,
deployment
models
or
just
sample
content,
some
of
the
same
sample
content
that
we
were
just
seeing
and
so
actually
raj.
I
would,
as
you
think,
through
this
and
consider
the
things
that
you're
putting
into
that
design
spec.
F
That
content,
that
you
were
thinking
of
in
the
design
spec
as
one
in
the
same
as
the
content,
that's
here
in
the
catalog,
if
you
weren't
such
that
mesh
reui,
when
someone
is
inside
of
that
ui
that
this,
the
content
that
someone
would
see
when
they
go
to
take
a
look
at
designs
that
maybe
there's
just
a
distinction
between
content,
that
they've
created
a
design
that
they've
created
here
locally
and
then
maybe
another
category.
That's
like
the
rest
of
the
catalog
is
one
way
of.
I
have
a
few
other
thoughts
to
share
there.
H
F
About
right,
yeah,
that's
about
right
and
that
that
doesn't
mean
that
providers
couldn't
bring
their
own
or
that
that
is
part
of
your
design.
Spec
that
you,
you
know
I'll
I'll.
Tell
you
a
little
bit
more
later.
We
can
have
the
chat
absolutely
for
sure
publicly,
I'm
just
I'll.
Just
talk
everybody's
head
off
for
half
an
hour.
If
I
keep
going
okay,
that's
that's
great,
so
actually
yeah.
The
last
thing
that
I
have
to
say
as
like
a
point
of
pride
on
behalf
of
some
of
the
work
that
you
all
have
done.
F
Some
of
you
are
working
with
some
of
the
folks
from
the
hpe
engineers
who
who
have
a
who
focus
a
lot
of
their
time
on
spire,
and
so
there
was
a
recently
a
published
design
for
istio
and
aspire
configured
istio
deployment.
So
that's
a
new
entry
out
in
the
catalog,
so
so
kudos
to
those
that
have
been
working
on
that.
I
So,
as
most
of
you
know
that
we
are
restructuring
our
ui
for
mui5
and
in
an
income
and
redux,
so
I
have
listed
some
of
the
components
updated
the
tracker
for
the
latest
components
and
the
status
on
what
they
are.
There,
also
two
or
three
open
components
that
people
can
work
on,
and
I
will
also
add
more
components
so
on
the
updates.
I've
made
made
the
user
preferences
page.
I
don't
have
it
running
right
now.
I
I
B
B
B
If
you
have
some
questions,
then
we've
put
together
this
document,
which,
which
would
mostly
answer,
answer
the
questions
that
you
would
have
initially
so
do
check
out
this
thing
before
you
yeah,
I
mean
yeah,
do
check
it
out
and
if
you
scroll
down
like
I've
added
two
more
sections
over
here
yeah,
they
have
conventions
over
here,
yeah
and
they're
like
yeah.
These
are
so
so
one
more
thing
like
one
more
thing:
that's
actually
important
with
this
refactoring
process
that
you
know
how
to
is
reviewing
the
prs.
B
So
it's
as
equally
important
important,
as
you
know,
creating
the
prs
because
yeah
it
helps
it
helps
us
improve
the
quality
of
code
and
stuff.
Like
that
I
mean
I
don't
have
to
go
through
the
benefits
of
doing
that,
but
yeah.
So
so,
when
you
actually
want
to
review
the
pr
someone
has
made
or
if
you
want
to
suggest
some
something
better
then
yeah,
you
should
prefer
to
do
that.
But
when
you
are
actually
reviewing,
there
are
some
things
that
we
try
to
follow
in
this.
B
B
If
you
could
like,
we
will
be
using
models
everywhere
in
our
ui
code
base
and
then,
like,
let's
say,
like
let's
say
like
everywhere:
we
are
using
a
model,
we're
actually
creating
a
new
component
and
like
they
have
their
own
set
of
styles
and
stuff
like
that,
then
it
becomes
very
hard
for
us
to.
I
mean,
let's
say
that
you
you
just
have
to
change.
The
structure
of
our
model
looks
like
you
know
like.
B
Instead
of
instead
of
having
a
green
car,
you
want
to
change
it
to
black
or
something
so
in
that
case,
you'll
actually
go
have
to
go
to
each
and
every
component
in
this
team
change
the
styles.
But
when
you
have
a
when
you
have
a
global
component,
that's
defined
like
it.
When
it
is
more
generalized
it
it
becomes
easier
for
it
to,
and
then
it
actually,
you
move
closer
to
the
more
concrete
set
of
components.
Yeah.
B
When
you
actually
follow
this,
when
you
actually
follow
this
convention,
then
it
becomes
easier
for
us
to
refactor
code,
or
I
mean
the
ability
of
ability
of
changing
something
in
the
code.
Base
becomes
actually
much
easier,
which
is
which
is
generally
proportional
to
yeah,
which
is
actually
proportional
to
the
quality
of
the
code
piece
so
yeah.
There
are
some
things
that
has
been
added,
such
as
we
try
to
always
use
the
mua
style.
B
Dpi
mu
actually
provides
some
other
alternatives
as
well,
but
then,
like
they'll,
be
dropping
the
other
ones
and
then
they'll
be
sticking
with
nua
styled
ones.
So
it's
better
if
we
stick
with
using
one
thing
and
that
will
be
nua
styled
api
and,
as
I
said
like
we'll
have
to
see
to
that,
we
actually
reuse
the
components
as
much
as
possible
unless,
unless
they're,
I'm
not
saying
like
you
should
actually
completely,
I
mean
take
it
to
the
like.
B
You
will
know
like
you'll
know
like
when
you
actually
have
to
create
a
new
component.
So
unless
like
there
is
something
that
cannot
be
derived
from
the
existing
ones,
you
don't
have
to
do
that
so
yeah
and
one
more
thing
that
that
is
not
actually
spoken.
Much
is
that
it
would
be
better
if
we
start
writing
tests,
the
components
that
we
are
generating.
B
So
sorry,
tests
of
the
components
that
you're
actually
creating
so,
as
you
know,
most
of
the
backend
integration
stuff
are
actually
done
so
the
readex
part-
and
you
know
observables
and
stuff,
like
that-
those
are
actually
already
hooked
up.
So
most
of
the
things
that
that
you
guys
will
be
working
on
would
be
related
to
the
ui
component
creation
and
stuff
like
that.
So
there
is,
there
is
a
component
which
has
tests
as
well.
B
So
if
you
are
creating
components
and
if
you,
if
you
would
like
to
start
writing
tests
for
them,
then
it's
yeah.
It
would
be
good.
So
we
haven't.
We
are
not
forcing
to
do
that
because,
like
it's,
not
everyone,
yeah
yeah,
it's
like
you'll,
have
to
you'll
have
to
learn
that
right.
So
that's
that's!
Actually
a
process
so
yeah.
If
you'd
like
to
invest
some
time
on
that,
then
it
would
be
great.
I
I
I
So,
according
to
having
a
source
type
shown
on
the
applications
ui,
so
we
will
be
having
a
column
for
the
source
type
and
according
to
the
uploaded
application
file,
we
will
be
displaying
the
docker
compose
gate,
manifest
or
helm
source
type
with
the
icon.
According
to
that,
and
then
we
click
on
this
icon,
so
it
will
download
the
particular
application
file
for
the
ui
from
the
ui.
So
currently,
the
downloading
feature
is
not
working
over
there
and
I
have
to
add
some
more
updates
to
the
pr
so
I'll
push
that
thing.
I
So
for
this
thing
adding
a
support.
I
have
few
queries
if,
like
I
would
like
to
ask
so
currently
we
are
using
this
slash
application.
I
have
the
api,
slash
application,
slash
types
to
get
the
types
of
different
application
using
method
get
and
when
I'm
doing
a
console
over
here,
so
I
should
be
getting
a
null
type
value,
because
the
current
applications
do
not
have
a
source
type
and
source
content
over
there.
So
I'm
not
getting
this
source
content
from
the
console.
I
B
Yeah
previous
okay
yeah,
because,
though,
as
I
had
said
earlier
as
well,
the
only
change
that
we
would
need
in
mystic
cloud
is
inside
of
the
migration
files
we'll
have
to
include
that
column
in
the
table
and
that
we
haven't
done
right
now,
so
yeah.
So
that's
why
the
data
that
we
get
from
mystic
cloud.
It
does
not
have
that
extra
column
so
that
it
doesn't
have
that
field
so
yeah!
B
So,
okay,
so
I
guess
I'll
I'll
create
a
pr
in
mystic
cloud
and
we
can
get
that
must
the,
because
that
could
be
backward
compatible
with
the
correct
meshy
server.
That
will
just
be
adding
a
few
extra
fields.
So
I
guess
that
that
wouldn't
break
anything
and
that
can
help
you
to
get
going
with
this
pr
other
than
okay
yeah.
I
I
would
also
like
to
discuss
this
point
for
having
our
drop
down
support
for
uploading
from
the
import
model
so
like
currently,
I've
added
this
restricted
input
for
all
these
supported
file
formats
over
here
itself.
So,
but
I
have
a
question
like
why
there
is
a
need
for
having
a
drop
down
for
specifying
this
uml
and
like
other
formats,
we
can
just
from
the
ux
perspective.
We
can
have
it
and
the
text
or
tooltip
format.
So
is
there
anything
in.
B
My
head
in
my
head,
when
I
said
about
anything
about
drop
down,
I
meant
inside
of
that
model
only
so
while
uploading
you
can
specify
so
the
model
that
you
get
yeah
this
this
one
in
here
like
somewhere
around
here.
You
could
have
it
on
list
of
available
types,
so
you
can
specify
which
type
it
is
before
uploading
the
file
or
before
putting
something.
B
Yeah,
if
someone
else,
someone
has
some
better
because
I'm
not
really
good
at
ux.
So
if
someone
has
a
better
opinion
regarding
this
thing,
so
that
would
be
helpful,
although
yeah
and
we
want
users
to
specify
which
type
their
file
is
while
they're
added
just
before
they
click
on
import.
So
it
would
be
on
this
model
only.
I
I
I
I
E
C
C
So
here's
the
progress.
So,
firstly,
we
need
to
add
our
custom
formula.
So
the
formula
which
currently
korea
is
the
rates
for
us
is
not
a
valid
formula.
According
to
the
home
viewport
app
like
it
works
in
our
custom
app,
but
for
a
home
group
app,
we
need
to
have
some
extra
fields
like
description
home,
page
and
all
and
also
like.
C
We
can't
just
use
the
binaries
built
by
go
release
it
in
our
in
the
home
depot,
so
we
need
to
so
what
we
need
to
do
is
we
need
to
put
in
instructions
in
this
install
function
which
will
tell
homebrew
how
to
build
our
cli
from
source
so
yeah.
So
I
worked
on
this
formula
and
hopefully,
like
I
think
it
it's
a
valid
homebrew
core
formula.
I've
like
tested
it
on
the
ordered
checks
that
the
home
view
team
recommends
themselves
and
yeah
like
like
feeds,
need
to.
C
C
I
C
C
And
instead
we
need
to
integrate
a
github
action
which
will
automatically
make
a
pull
request
to
from
view
for
repository,
updating
the
the
version
tab
and
the
commit
commit
ss
sha
and
all
and
yeah.
And
the
third
thing
which
I
was
unsure
about
was
whether
we
should
continue
supporting
our
custom
tab.
That
is
this
repository.
But
like
yeah,
I
got
to
know
that
home
view
supports
migration
between
taps.
C
H
C
Yeah,
like
I
think
it
needs
a
config
to
check
for
the
server
url
and
display
this
version
of
the
server
too,
but
I
think
yeah
we
can
like
yeah.
We
can
keep
it
optional
like.
If
there's
a
virgin
config
file,
we
can
show
the
version
of
the
server.
Otherwise
we
can
say
that
it's
not
configured
or
something
like
that.
A
C
J
J
H
K
J
J
Another
thing
with
this
servo
version
details
not
available
part
right,
so
I
have
seen
the
cube
curtail
implementation
right
whenever
our
docker
related
commands
right
it.
If
the
server
is
not
running,
it
just
gives
the
client
information
and
then
says
that
server
is
not
running
or
not
available.
I
think
that
should
be
acceptable.
Instead
of
forcing
the
users
to
depend
on
a
font
file
right.
C
So
like,
if
there
are
no
other
comments
like,
I
would
like
to
like
talk
about
another
task
like
a
related
like
a
similar
task
that
is
like
including
mystery
ctl
in
edge
releases
like
currently.
We
like
the
users,
can
only
install
messages
from
our
stable
releases
like
when
we
do
a
stabilities,
and
there
is.
I
don't
think
there
is
any
way
like
users
can,
like
simply
install
edge
releases
currently,
like
a
user,
can
probably
clone
their
app
and
build
themselves.
But
there
is
no
simple
way
of
doing
that.
C
So
after
looking
a
bit
like
there's
a
nice
way
of
doing
it
in
home
view
so
like,
if
you
see
here
as
we
have
defined
the
head
of
our
repository
in
the
formula,
so
what
the
users
can
do
is
if
they
simply
install
mercedes
etl
with
this
command
like
with
the
head
flap.
Basically
so
it
build,
it
will
build
the
like
current
head,
the
master
branch
of
the
cla
for
them,
and
also
afterwards
they
can
upgrade
the
cli
using
a
similar
command.
C
The
upgrade
measures
cdl
fetch
head
yeah,
so
I
think
it
will
work
like
if
a
user
wants
to
build
the
cli
from
the
master
branch
but
as
far
as
other
other
modes
of
installation
like
screw
or
a
script,
are
concerned.
I
haven't
really
given
a
lot
of
thought
yet,
but
but
something
that,
like
I
had
an
idea,
I
haven't
really
thought
much
on
it.
C
But
here's
what
I
thought
like
we
can
use
the
binaries
built
and
put
them
somewhere
in
a
separate
repository
or
a
release
or
somewhere,
and
we
can
maybe
modify
our
script.
The
custom
script,
which
we
have
written
for
installation
for
installing
mystery,
cdn
and
like
add
a
flag
or
something
for
the
edge
release,
and
it
will
pull
from
there
every
time
everything
when
the
user
wants
to
download
the
edge
release
so
yeah
like
I
just
as
I
said.
It's
not
like
really
well
thought
at
this
point
of
time,
but
yeah.
F
Cool,
that's
great
nishant.
It's
like
you've
been
on
this
call
before
or
something
you've
got
two
different
issues
to
present
a
pull
request
that
changes
around
the
cli
client.
I'm
great,
like
we've,
been
one
of
the
goals
of
the
project.
For
since
its
inception
has
been
to
delight
users
to
do
delightful
things,
and
if
users
can
use
one
command
to
install
the
ctl,
that's
great,
they
don't
have
to
go.
Add
another
add
a
repo
to
their
existing
repos.
F
So
this
is
really
nice
just
to
reflect
on
potential
use
of
brew
ad
mastery
ctl
head
the
attractive
portion
of
this
of
this
possibility
is
well
it's
just
less
maintenance
for
us
collectively
to
add
natively
to
mastery
ctl.
So
it
makes
sense
to
just
you
lean
into
it
might
be
also
something
that
maybe
brew.
Users
are
familiar
with,
or
some
brew
users
might
be
familiar
with
this
convention
and
so
that
that
might
feel
natural
to
them.
F
F
If
you
haven't
spent
a
lot
of
time
with
that
particular
command
meshre
ctl
system
channel
play
with
a
bit
it
lets.
You
update
your
mesh
redeployment
to
unless
you
switch
your
measuring
deployment
from
one
channel
to
the
next
it
today.
It
doesn't
switch
your
measuring
ctl
binary
from
one
channel
to
the
next
in
part
for
for
a
few
different
reasons
in
part,
because
people
may
have
installed
meshry
ctl,
using
brew
or
scoop
or
a
bash
script,
or
just
gotten
it
directly,
and
so
that
can
be
figured
out.
F
E
G
C
So
this
is
about
the
handling
of
expired
session
earlier.
So
what
was
happening
when
the
session
was
expiring
and
the
wish
the
user
should
be
logged
out
from
the
bachelory,
but
this
was
not
happening.
We
were
redirecting,
but
due
to
some
cost
error
or
some
similar
error,
we
were
not
able
to
redirect.
C
C
F
Shazam,
okay,
nice,
that
that
first
one's
been
a
long
awaited
for
redirect
right
when,
if
someone
has
established
a
a
session,
you
know
signed
into
measure
ui
at
some
point
that
session
expires,
it
hasn't
always
been
obvious.
Sometimes
users
will
come
sit
back
down
at
their
desk.
They
start
to
use
messaging
they're
kind
of
wondering
what's
going
on,
why
aren't
they
seeing?
Why
are
they
getting
errors
and
that's
because
their
session
is
expired,
so
that
redirect
is
nice?
I
missed
the
second
thing
and
what
you
just
said,
though,.
C
F
E
K
All
right,
I
hope
the
screen
is
now
visible,
so
for
like
up
to
past
two
days,
I've
been
working
on
an
issue
which
is
like
a
feature
request.
That
is
the
addition
of
the
screenshots
and
the
mystery
ctl
reference
section
such
that
the
users
can
see
what
the
basic
usage
of
the
command
is,
and
they
can
visually
refer.
Okay,
so
this
is
the
basic
usage
of
this
command.
K
Take
a
such
that
the
developers
can
add
the
command
using
this
specific
symbols
which
have
made
them
as
a
rule,
as
you
can
see
here
so
like
I've
made
a
logic
where,
if
there
is
a
prefix
with
this
certain
set
of
characters,
this
is
for
those
who
are
new
to
this
change.
For
those
who
already
know
about
this
yeah
they
already
know
so.
K
I
was
slacking
out
an
example
like
how
this
is
exactly
works
when
it
is
implemented
in
the
documentation,
so
a
sample
has
been
valid
on
the
mastery
cpl
system
check
command.
So
right
now
what
you
are
seeing
here.
This
is
the
comment
section
like
you
can
indicate
the
developers
of
the
measuring
that
this
is
just
a
command.
K
Please
don't
change
the
below
command
like
that,
any
anything
can
be
given
here
as
a
command
and
this
star
star
prefix
is
used
to
denote
the
caption
of
the
image
such
that
the
users
can
know
what
kind
of
image
it
is
and
last,
but
not
the
least,
is
the
hash
prefix,
which
denotes
the
markdown
code
for
the
image.
So
right
now
the
image
has
been
fetched
from
the
twitter.
K
F
F
K
Yes,
which
can
be
implemented
on
the
machine,
cto
golang
files
itself,
so
that
developers
need
not
directly
edit
onto
the
markdown
page
and
then
do
the
changes.
Rather,
they
can
just
add
the
code
here
itself
for
using
the
command
see
itself
in,
as
you
can
see
here
in
the
example
variable,
you
can
just
add
the
code
here
itself
and
it
will
automatically
render
using
the
just
give
me
a
sec.
K
E
A
So
just
a
quick
update.
This
is
regarding
the
table
sorter
in
which
we
are
sorting
the
tables
of
the
past
integration
test
results
and,
according
to
each
field,
on
the
call
according
to
each
table
header.
So
now
it's
complete
like
earlier.
It
was,
and
there
was
an
issue
that
when
the
table
was
sorted,
we
could
not
see
the
details
of
it
directly
below
it.
So
now,
after
like
reading
the
documentation
and
experimenting
a
lot
of
things,
I
am
able
to
do
that
so
yeah.
F
Nice
speaking
of
new
data,
there's
so
is
mr
hamza,
on
secrunder,
he's
been
working
on
a
workflow
for
end-to-end
tests
within
adapters.
That
would
I
don't
know
if
it
sounds
like
you're
aware
of
that,
once
a
console,
she
was
working
on,
console,
yeah
and
then
hopefully
he
would
take.
That
would
be
universal
to
all
of
the
adapters
where
output
of
like
either
whether
it
was
end
to
end
test
whether
it
was
the
sample
apps.
I
think
he
was
just
going
to
augment
some
of
the
return
data,
some
of
the
some
additional
tests.
A
F
A
F
Just
for
my
own
benefit-
and
maybe
everyone
else
is
so
betty-
if
you
don't
mind
if
you
go
back
back
a
page
to
the
compatibility
matrix,
so
we've
got
a
top
level
summary
of
kind
of,
I
think,
sort
of
kubernetes-centric
compatibility
across
the
adapters.
I
wonder
if
it
makes
sense
to
have
that
from
measuring
at
some
point,
ashish
sounds
like
a
maybe
a
good
topic
for
tomorrow's
build
and
release
call.
F
But
and
then
you
have
a
high
level
summary
of
or
I
guess
it's
the
latest-
the
results
from
the
latest
integration
tests
across
the
adapters.
I
don't
see
mesh
reservoir
necessarily
here
either
so
food
for
thought
about.
You
know,
there's
any
number
of
tests
that
run
there
as
well
might
be
good
to
expose
all
right.
B
Actually,
yeah,
that's
a
that's
a
good
point.
The
same
way.
We
test
the
adapters
we
test.
We
are
actually
testing
the
pattern,
ops
capability
of
the
adapters
and
when
we
are
deploying
just
the
kubernetes
components
in
the
machine,
server
itself
acts
as
a
delegator,
so
it
has
the
humidity's
component.
So
I
think
we
should
have
the
same
exact
tests,
these
end-to-end
tests
that
we
have
in
our
adapters.
B
We
should
have
these
tests
inside
of
meshi
server
as
well,
and
there
we
can
test
the
application
core
component
along
with
the
rest
of
kubernetes.
Any
number
of
you
know
pod
service
and
all
we
can
have
yeah
so
the
basically
that
would
that
would
so.
In
that
way,
the
master
component
on
this
matrix
would
be
messy
server,
and
that
would
signify
its
capability
to
deploy
the
communities,
components
and
the
core
components
that
it
carries
with
itself.
B
Yeah
and
right
now
it
isn't
available.
So
the
first
step
is,
of
course,
to
just
like.
If
you
go
to
dot
github,
slash,
workflow
slash,
you
will
find
e2e
text
dot,
yaml
everywhere
in
all
the
adapters.
We
need
to
have
one
of
those
inside
of
machine
server
as
well,
and
there
we
would
use
the
pattern
file
that
we
would
deploy
there
wouldn't
have
any
of
the
adapter
stuff.
It
would
have
the
application
component
and
the
rest
of
the
communities
components.
B
So
the
basically
a
few
basic
things
we
can
test
our
report
service
deployment,
config
map
secrets.
You
know
just
things
that
of
the
top
of
my
head.
These
things
you
can
test
and
put
in
a
single
big
yaml
file,
for
example
the
istio
spy
design.
Actually,
that's
a
that
can
be
used
to
have
tests
in
our
message,
server
and
yeah,
so
it
would
also.
B
But
the
only
thing
that
would
differ
is
that
currently,
the
third
job
inside
of
meshi
adapter
in
all
of
the
tests
is
the
update
for
docs,
where
it
goes
and
creates
a
comma
basically
commits
on
the
messaging
message
repository,
because
in
this
case
the
workflow
would
be
running
in
machine
machine.
So
we
would
not
need
to.
E
Okay,
so
we
have
one
minute
left
so
I'll
just
quickly
go
over
my
topics-
they're,
very
small.
So
currently
we
are
thinking
of
like
he
had
brought
it
up
sometime
before
we
are
thinking
of
having
extensions
marketplace
sort
of
where
we
can
have
the
catalog
mesh
map
and
even
the
adapters
listed
listed
as
an
extension,
and
it
would
be
an
easy
install
from
there.
I
didn't
have
a
mock-up
ready,
but
since
I
had
already
shared
the
screen
there,
this
is
a
small
mock-up
that
I've
been
working
on.
E
I'm
gonna
iterate
more
on
that
it's.
This
is
a
very
yeah
like
I
started
on
it
like
an
hour
back
so
yeah,
so
something
that
we're
going
to
work
on
and
yeah.
I
just
wanted
to.
Let
you
guys
know
if
someone
like
I'll
be
working
on
making
the
mock-up
and
then
maybe
we
can
collaborate
or
something
to
make
this
and
the
other
topic
was.
There
are
a
couple
of
help
wanted
issues.
E
E
E
D
F
F
D
F
Yeah,
oh
got
you
gosh,
you
got
you
yeah.
It
makes
sense
some
on
occasion
as
people
have
work
in
progress.
Sometimes
it's
helpful
to
them
to
submit
like
a
draft
pull
request,
knowing
that
it's
not
time
to
review
it,
but
what
sometimes
the
reason
they
do.
It
is
one
just
to
like
tell
people
to
prove
that,
like
oh,
hey,
they're,
actually
working
on
it,
you
know
like
don't
give
it
to
someone
else,
I'm
doing
it
or
like
hey.