►
From YouTube: wasmCloud: "Good First Issue" Callout, GraphQL Provider Demo - 11/03/2021 Community Call
Description
wasmCloud is a platform for writing portable business logic that can run anywhere from the edge to the cloud, that boasts a secure-by-default, boilerplate-free developer experience with rapid feedback loop.
https://wasmcloud.com
https://wasmcloud.dev
Demo:
https://github.com/StuartHarris/wasmcloud-graphql-provider
A
All
right
welcome
to
wasm
cloud
wednesday
for
wednesday
november,
the
3rd
2021
we're
going
to
start
with
a
review
of
some
of
the
performance
on
the
social
media.
Stuff
we've
been
putting
out
jordan,
you
wanna
maybe
take
control
and
nobody
has
rights
to
present.
It's
like
a
running
joke.
It's
like
every
week
watch
liam
forget
to
give
people
permissions
to
present
all
right.
It's
okay.
B
We
forgive
you
thanks,
jordan,
such
a
great
all
right
there
you
go,
I'm.
B
B
Yeah,
so
no
the
few
things
I
just
wanted
to
share
right
that
ignore
the
downtails
at
the
end
right,
some
of
that's
just
the
day
analytics,
but
over
the
last.
I
think
this
is
scoped
at
you
know.
Four
weeks
we've
had
almost
500
views
on
our
on
our
videos
and
almost
40
hours
worth
and
even
though
that's
a
red
negative,
eight
right
and
11
plus
subscribers.
So
we
are
getting
a
lot
of
traffic
to
these
videos
and
and
which
I
thought
was
really
cool,
because
I
mean
we've
only
just
started.
B
I
think
in
the
last
week
or
two
ensuring
like
they
were
played,
you
know
plopped
on
every
platform
at
the
end,
so
you
know
I
just
wanted
to
show
that
I
mean
one
of
these
was
pretty
cool
yeah.
The
google
search
is
like
you
know,
accountable
for
44
and
a
half
percent
of
our
of
our
traffic
so,
and
that
just
goes
to
show
that
google
is
oh
and
they're
searching
for
webassembly
kubecon,
and
these
are
the
things
so
people
are
finding
us.
B
You
know,
what's
the
word
naturally,
I
know
there's
a
word
for
that,
but
organically,
and
I
just
think
it's
it's.
It's
very
cool
to
show
everyone
that
this
this
work
is
that
we're
doing
is
not
going
unnoticed
and
people
are
finding
us-
and
you
know,
continue
to
drive
your
co-workers
and
friends
to
the
to
the
content,
because
you
know
the
more
people
see
it
we're
seeing.
You
know
upticks
and
you
know
pretty
much
everything
these
big
up,
ticks
in
the
middle
were
kukan,
so
that's
cheating,
it
doesn't
count
but
yeah.
A
Well,
I
really
appreciate
all
the
stuff
that
you're
doing
the
pr
team.
That's
going
to
chat
with
you
about
I'm
writing
up
a
blog
post
each
week
and
they'll
help
you
with
that.
If
you
can
just
give
them
some
bullet
points
and
getting
it
also
linked
on
watsoncloud.com,
which
is
something
we
probably
should
go
back
and
just
kind
of
back
link
some
of
the
key
ones
and
call
out
you
know
things
that
people
would
be
interested
in,
like
the
helm
chart
or
whatever.
But
thank
you
so
much
jordan.
A
I
really
appreciate
you
just
consistently
knocking
that
out.
It's
amazing
and
brooks.
I
think
you
want
to
talk
a
little
bit
about
a
proposal
that
you
have
today.
C
I
did-
and
I
just
wanted
to
get
my
screen
sharing
right
before
so
something
that
we've
had
on
our
issues
in
in
the
wasn't
cloud
org.
You
know
this
is
wash
for
example,
but
we
have
it
on
the
otp
repo
and
our
other
repositories,
as
well
as
the
label
called
good
first
issue,
and
you
know
this.
This
should
be
something
that
somebody
coming
into
wasn't
cloud
for
the
first
time
as
long
as
they
have.
C
Maybe
a
basic
knowledge
of
rust,
for
example,
should
be
able
to
pick
up
this
issue
and
contribute
for
the
first
time,
and
what
I
really
wanted
to
do
is
drive
this
forward
so
that
community
members
can
know
which
ones
are
are
important
to
us.
I
mean,
obviously,
every
issue
that
we
have
is
important,
but
ones
that
really
are
are
fully
scoped
out
and
something
that
they
could
pick
up.
So
the
one
one
that
I
created
this
morning,
which
is
when
I
kind
of
had
this
idea,
is
wash
has
environment
variables
like
wash
rpc
host.
C
I
didn't
yeah
like
wash
rpc
host
and
that's
how
you
specify
the
nas
server
that
you're
using
for
rpc.
When
you
launch
a
wasm
cloud
host,
you
can
use
an
environment.
Variable
called
wasn't
cloud
rpc
host
to
designate
the
nas
server
that
you
want
to
use
for,
like
remote
procedure
calls.
So
the
idea
is
that
if
we
merge
these
together-
or
you
know,
and
in
wash
instead
of
having
its
own
wash
rpc
host,
have
it
use
wasn't
quite
rpc
host.
C
Instead,
then
you
could
feasibly
start
up,
gnats
start
up
an
otp
host
and
then
immediately
start
using
wash
to
issue
or
like
to
call
actors
or
to
issue
control
interface
commands
things
like
that.
But
anyways.
This
issue
is
something
that
could
be
a
great
starting
point
for
somebody
to
get
into
the
wash
code.
Take
a
look
at
some
of
the
otp
host
code
to
look
at
environment
variables.
C
So
I
wanted
to
kind
of
pioneer
something
or
like
an
experimental
section
of
the
community
call
which
you
know
I
want
to
do
it
here
for
everybody,
who's
attending
and
then
post
in
slack.
Afterwards,
with
you
know
every
week,
one
or
two
issues
that
we
have
on
our
backlog,
that
would
be
great
for
somebody
who
wanted
to
contribute
for
the
first
time
or
repeat,
contribute
and
just
kind
of
draw
attention
to
that.
C
And
then,
if
anybody
is
interested
in
taking
it
on,
you
can
just
leave
a
comment
in
the
issue
and
say:
hey,
I'm
going
to
take
a
look
into
this
and
try
to
issue
a
pr
as
a
general
rule.
We
would
want
these
issues
to
be
very
well
scoped
out.
I
think
this
one
needs
a
little
bit
more
description
very
well
scoped
out,
so
that
all
the
information
you
need
is
in
the
issue
and
also
something
that
doesn't
take.
C
You
know
20
hours
of
time
when
somebody
as
a
community
member,
you
know
everybody's
got
things
going
on
so
not
taking
more
than
an
hour
or
two
of
free
time
or
something
like
that
to
to
get
in
and
contribute
so
interested
to
hear
any
feedback
on
that
gonna
try
and
go
through
some
of
our
issues
in
the
in
the
github
backlog
for
the
otp
host
and
wash
and
capability
providers.
Things
like
that
and
and
see
what
we
can
do
for
that
segment.
C
B
A
You
know
I
have
been
thinking
about.
You
know,
like
I
don't
know.
If
there's
a
service
that
will
mail
stickers
on
demand
right,
you
know,
we've
done
that.
We've
opened
up
t-shirts
before
and
I'm
happy
to
just
like
you
know,
have
nicole
throw
out
a
permalink.
You
know
out
there
every
month,
but
maybe
I
could
you
know
like
bribe
my
kids
to
mail
stickers.
You
know
I
mean
we
don't
know,
we
would
even
have
that
many
print
off
labels
once
a
month.
A
C
A
There
are
places
that
do
t-shirts
on
demand,
so
jordan,
if
or
brooks,
if
you
want
to
have
this
written
up
so
that
you
know,
like
closing
issue,
get
a
t-shirt
or
something
I'm
I'm
happy
to
to
just
fund
that
so
I'll
pull
up
with
you
on
slack
and
we
can
kind
of
maybe
figure
out
a
path
forward
there,
because
there
are
a
lot
of
places
that
you
know
what
we're
doing
with
the
previous
t-shirt
place
was
fine,
but
they're
dedicated
like
drop
shippers,
that
you
can
ship
a
couple
cases
of
t-shirts
to,
and
then
they
just
kind
of
ship
them
out
on
demand.
D
Okay,
great
well,
thank
you
kevin.
Did
you
sorry
kevin?
Were
you
saying
something
yeah
I
just
wanted.
E
To
point
out
specifically
about
this
one
issue
that
it's
one
where
the
amount
of
effort
is
pretty
low,
but
the
impact
is
pretty
high
because
you
know
anything
that
smooths
out
or
makes
easier
the
developer
experience
is,
is
absolutely
worth
it.
So
this
is
one
of
the
good
ones
where
it
won't
take
much
work
to
make
things
better.
A
Thank
you
kevin.
That's
an
awesome
call
out
as
well
so
well,
I
think
a
stewart
you
were
next.
You
had
something
you
want
to
talk
about
this
week,
which
was
a
new
capability
provider.
You've
just
been
experimenting
with
in
your
free
time,
go
ahead.
F
Okay?
Yes,
I
I
was
just
playing
around
trying
to
build
a
provider
that
wraps
a
postgres
database
with
a
graphql
api,
not
necessarily
for
exposing
directly
to
the
browser,
but
I
suppose.
F
A
few
providers
and
the
sql
provider
now
as
which
is
great,
but
I
thought
it
might
be
so
I've
used
this
thing
before
called
postgraph.
I
don't
know
if
anyone's
seen
it,
which
is
quite
cool,
postgraphial
introspects,
the
the
the
sql
database
schema
and
generates
a
graphql
schema
and
then
provides
resolvers
to
do
queries
and
mutations
against
it
and
it's
like
relay
compatible,
etc,
but
it's
written
in
typescript
and
you
know
short
of
asking
the
maintainers
to
rewrite
it
in
rust.
F
F
Normally,
you
use
neon
to
go
the
other
way,
so
it's
really
for
writing
native
node
modules
in
rust,
for
instance,
that
you
can
call
from
from
node
from
typescript
to
javascript
whatever.
Oh
sorry,
I
shouldn't
have
clicked
on
that
and
the
other
way
around.
So
we
effectively
just
link
the
road,
then
runtime
and
start
it
up
and
then
use
neon
to
to
talk
the
other
way.
F
So
it
does
work.
I've
got
this
a
pass-through
actor
which
is
just
there
for
testing
purposes
so
in
in
the
I
don't
know.
If
you
can
see
all
of
this
get
rid
of
that,
so
we've
got
an
http
server
provider
and
this
graphql
provider
and
this
password
a
link
to
the
hp
server
and
a
link
to
the
graphql
provider.
That
just
has
a
database
url
on
it
and
that's
the
only
configuration
it
has.
F
Basically,
so
it
knows
how
to
get
to
the
postgres
database
and
then
everything
else
is
is
just
done
by
with
reflection
effectively.
So
I've
got
a
query
here,
which
literally
just
runs
a
graphql
query
could
have
formatted
this
a
bit
better.
So
then,
if
I
do
a
post
to
the
endpoint,
then
it
returns
that
it's
usually
quite
fast.
F
There
we
go,
there's
no
data
in
the
database
to
speak
of
really,
but
but
it
was
just
really
cool
and
then
I'm
not
sure
if
this
is
actually
going
to
work,
because
this
is
this
is
a
graphql
playground,
one
of
the
problems
with
there's,
so
I've
got
two
end
points
effectively,
one
for
the
graphql
queries
and
one
for
actually
for
the
graphql
playground,
but
I'm
not
using
that
I'm
using
the
built-in
one
so
ignore
that,
but
it
sends
an
introspection
query
which
could
be
bigger
than
a
megabyte.
F
So
I'm
not
sure
if
it's
actually
going
to
use.
No,
it's
not
going
to
work.
This
is
a
problem.
We
need
to
sort
out
which
I've
been
talking
with
steve
about
recently,
which
is
about
what
happens
when
you,
when
you
kind
of
send
a
big
payload.
I
think
that
payload
is
just
under
a
megabyte.
F
Let's
try:
we
should
no
okay,
all
right.
Okay,
demo
didn't
work,
but
curl
works.
It
is
poor
ath.
Isn't
it
yeah
it's
working
right
anyway.
It's
there.
If
anybody
wants
to
have
a
look
at
it,
I
just
did
it
from
from
for
playing
around
with
really,
but
the
the
interesting
code
is
in
the
provider
that
has
so.
This
is
the
provider,
that's
the
configuration
that
just
takes
the
url
database
url
and
then
it
calls
into
into
this
graphql.
F
Upstream,
which
uses
neon-
and
it
literally
just
sets
up
javascript
functions
and
calls
into
into
node
into
javascript,
and
then
there's
there's
a
callback
which
is
called
when
the
request
is
finished,
and
then
it
ties
to
the
two
things
up
up
together
on
a
channel
and
then
sends
a
response
back.
But
it's
quite
an
interesting
concept
because,
like
it's
well
I
mean
it
possibly
means
that
you
can
have
providers
written
in
modern
time.
You
know
it's
messy
because
it
has
to
compile
the
type
script
together.
F
So
the
files
are
included
as
part
of
the
binary
for
the
provider,
which
is
a
bit
messy
and
it
unpacks
them
into
a
template
directory
so
that
it
runs,
but
but
yeah
it
does
work
anyway.
I.
A
Thought
it's
worth
just
playing
around
with
it.
Stuart!
That's
awesome
man.
I've
got
some
questions,
but
anyone
want
to
start.
E
Sure
I'll
start
first,
this
is
it's
fantastic.
I
love
seeing
people
play
with
with
stuff
like
this.
E
F
Generic
yeah,
so
I
didn't
show
you
the
smithy
contract,
but
it's
literally
just
got.
Let's
show
my
screening
and
show
it
to
you.
It's
literally
just
got
to
where
is
it
interface,
graphql
provider.smithy?
F
So
it's
literally
just
got
a
query:
there's
an
operation
to
get
the
graph,
the
graphical
playground,
but
I'm
I
wasn't
using
that
before
yeah.
This
query
just
has
a
query
request
and
a
query
response,
and
so
it's
completely
generated
the
contract.
F
So
the
client
makes
the
graphql
query,
so
the
actor
would
would
effectively
generate
a
graphql
query
and
submit
it
to
the
provider.
F
Very
cool,
so
it
could
be
so
it
could
be
used
against
any
database.
So
what
I
was
going
to
do
is
include
in
in
this,
including
this
repo
like
a
to
do
database
with
some
some
stuff
in
it
or
whatever,
just
as
an
example
and
then
in
the
docker
composer
and
we'll
just
bring
that
up
as
part
of
the
demo.
Basically.
F
F
So
I
know
that
that
could
be
up
to
eight
megan
and
even
further
to
64,
but
but
I
think
there
is
still
an
outstanding
problem
with
sending
large
payloads
through
which
haven't
got
to
the
bottom
of
yet,
but
but
I'm
using
this
to
try
and
sort
of
flush
those
out
if
there
is
a
problem
flushing
out
yeah.
E
One
of
the
one
of
the
potential
future
features
that
we've
talked
about
is
the
the
notion
of
chunked
invocations.
E
So
one
potential
solution
to
this
would
be
so
that
when,
when
the
host
runtime
sends
invocations
and
receives
invocation
responses,
it
does
so
in
chunkable
or
chunked
fashion,
so
that
it
can,
you
know,
inspect
the
original
payload
and
then,
if
necessary,
split
it
into
you
know
a
number
of
micro
requests
and
then
have
the
other
side,
collect
them
and
then
reassemble
the
payload
into
the
larger
one.
F
Yeah,
I
think
that
that
would
be
amazing
because
quite
often
like
a
web
page,
for
instance,
if
it's
got
like
assets
embedded
in
or
whatever
could
or
you
know,
can
easily
go
over
one
meg
thing
and
it'd
be
nice
not
to
have
to
run
on
a
nats
infrastructure
with
large
message.
The
size
configured.
G
F
E
Be
amazing,
I
think,
even
if
you
even
if
you
set
the
message
size
higher
than
one
meg,
I
think
nats
actually
has
a
hard
upper
limit
of
two
legs
right.
You
can't
go
higher
than
that.
F
F
Which
would
be
great
this?
You
know
I'm
more
than
happy
to
try
transfer
this
reaper
over.
If,
if,
if
at
any
point
it
seemed
you
know
it's
deemed
useful
or
to
move
the
contents
of
it
into
providers
reaper
or
whatever,
wherever
you
know
a
little
bit.
A
A
Think
I
think
when
we
work
through
some
of
these
issues
and
we
get
it
stable
and
we
feel
like
this
is
something
that
is
generally
usable.
Then
absolutely
you
know
this
should
yeah.
I
mean
all
the
things
you
know
stuart
are
welcome
and
then
we
can
get
it
published.
I
think
we
would.
We
haven't
really
ever
talked
through
the
sort
of
like
how
do
we?
A
What
do
we
publish
as
official,
but
I
am
inclined
to
think
that
you
know
you
know
we
would
want
to
pull
in
as
many
things
as
possible,
but
we
can
have
a
sidebar
on
that
on
slack
or
whatever.
The
case
would
be,
especially
when
it
adds
a
lot
of
interesting
utility
like
this.
Does.
F
This
is
just
I
mean
the
good
thing
about
it
is
that
it
is
completely
generic.
So
there's!
No
apart
from
your
database
connections,
you
know
the
database
url,
there's
no,
there's
no
additional
configuration,
I
mean
there
could
be
because
you
can
tune
postgraph
file
to
do
different
things,
but
you
know
it
does
work
out
of
the
box
without
it.
A
Cool
any
other
questions,
any
other
questions
for
stuart.
A
Okay,
well
just
turning
over
to
a
couple
things,
we
obviously
had
a
very
successful
kubecon
the
week
before
last,
a
lot
of
good
feedback
and
discussions
and
discovery
of
large
users
out
in
the
world
that
are
doing
amazing
things
with
blossom
cloud.
We're
going
to
continue
to
encourage
those
people
to
share
their
demos
and
use
cases
as
we
work
through
them.
A
planning
is
already
underway
for
kukan
eu,
which
will
be
in
valencia,
spain
next
year,
as
well
as
another
wasm
day
event
that's
put
on
by
the
cncf.
A
So
no
details
on
that
yet,
but
we're
just
planning
an
early.
You
know
workshop
or
something
along
those
lines
sometime
before
valencia,
because
there's
a
large
demand
for
it
and
it
looks
like
there
will
be
a
webassembly
event
for
us
government
people
in
washington
dc
in
the
near
future,
early
stages
on
planning
on
that
as
well.
For
a
web
assembly
wasn't
cloud
focused
event,
but
otherwise
I
will
keep
anyone
abreast
any
other
community
issues
to
share
or
notes.