►
From YouTube: Insomnia Stream 2022 02 22
Description
A weekly stream focused on Insomnia's open source community.
Join us on the #stream channel over on the Insomnia community slack: https://chat.insomnia.rest/
A
A
B
Hi
everyone,
so
this
is
the
insomnia
stream
thanks
for
joining
in,
we
have
on
the
stream.
Today,
I'm
dimitri
a
developer
here
on
the
insomnia
team.
C
James,
likewise,
working
on
this
on
your
team.
B
B
Yeah
all
right
gaz
has
a
pr
up,
so
we
can
just
so
we'll
filter
by
the
end,
you
guys
can
see
my
screen
right
so
we'll
filter
by
the
insomnia
stream
label
and
here's
the
one
okay.
So
oh,
it
closes
two
things.
That's
always
a
it's
always
a
bonus
open
api
importer,
okay.
B
So
this
is
about
open
api,
specs
and
oauth
2.,
okay
from
october
and
february
2021,
okay,
great
so
existing
importer
for
the
open
api
spec
requires
specifying
scheme
bear
so
any
any
overall
before
I
like
read
it
word
for
word
any
overall.
Do
you
remember
it's
been
a
while.
I
guess
what
this
is
all
about.
D
Yeah,
it
has
been
a
little
while
so
yeah.
I
was
trying
to
get
an
open
api
spec
using
open
off
and
using
go
off
into
insomnia.
The
the
syntax
highlighting
was
complaining
that
it
was
an
invalid
specification
because
if
you
didn't
specify
schema
borrower,
it
would
just
say
sorry
not
happy
with
it.
But
if
you
look
at
the
the
rscs
for
all
for
for
open
api,
not
the
app,
not
the
rscs,
there's
a
nice
little
stack
for
it.
D
Yeah
you're,
not
supposed
to
specify
that
scheme
attribute.
If
I
remember
correctly,
okay,.
B
B
So,
let's,
let's
try
to
create-
I
always
like
to
here
I'll
start
up.
The
app
oh
cool
james
is
in
here
yeah.
So
anyone
watching
you
can
come
on
the
insomnia
stream
channel
on
the
insomnia
or
the
stream
channel.
It's
just
called
stream
on
the
insomnia
community
slack.
Could
let's
see,
let's
put
a
link
out
to
that
I'll,
grab
a
link
in
a
second
and
put
that
in
the
in
the
chat
here
but
yeah
you
can
go
on
there
and
there's
a
link
to
a
vs
code,
liveshare,
so
yeah.
B
Actually
james,
would
you
mind
throwing
that
link
up
on
the
youtube
it's
in
the
here?
Okay,
so
so
we
have
insomnia.
B
Yeah,
so
just
for
anyone
to
know
everything
we
do
about
this
stream
is
done
in
the
open.
We
have
a
like
a
planning
document
here
that
we
have,
and
at
the
bottom
of
this
planning
document
is
some
links.
So
sorry,
I
guess
I'll
grab
it
and
I
can
throw
it
here
in
the
chat.
B
So
there's
the
insomnia
community
slack
and
there
is
just
chat
it
out
and
you
can
see
it
on
the
screen
so
I'll
put
on
the
screen
for
everyone
to
see
and
chat,
dot,
insomnia,
dot,
rest
and
then,
if
you
go
there,
you'll
see
the
channel
that
you
need.
Okay,
great!
So,
let's,
let's
get
going
over
there
you'll
find
the
vs
code
live
share
if
you're,
if
anyone's
interested
okay.
B
We'll
see
if
we
can
get
it
moving
here,
so
the
the
basic
point
is
we're
we're
opening
or
we're
importing
an
open
api
spec,
and
it
needs
to
contain
information
that
it
doesn't
about.
The
oauth
flow
is
that
right.
D
It
needs
to
yeah
yeah.
It
was
basically
for
me
fix
the
validation,
so
it
was
no
longer
complete.
It
was
no
longer
unhappy
about
that.
Lack
of
that
schema
attribute.
D
I
think,
at
the
same
time
I
I
kind
of
did
a
little
bit
of
tidying
just
so
that
you
could
specify
things
like
a
redirect.
Url.
Oh
this
is.
It
was
quite
some
time
ago
for
this
one
yeah.
D
Yeah,
I
think
there
was
a
bit
more.
There
was
kind
of
effectively
a
bit
of
being
helpful
going
on
going
on
here.
So
I
know
you
can
specify
your
kind
of
environment
very
you
kind
of
like
your
variables
in
in
insomnia,
so
that
you
can
kind
of
reuse
them
across
across
requests.
D
Yeah,
that's
basically
what
this
is.
That's
also
part
of
what
this
pr
is
doing
as
well.
So
if
you've
got,
I
know
if
you
want
to
specify
like
a
client
id
or
redirect
you
url
for
a
wolf
you
can
just
kind
of
rather
than
having
to
copy
it
into
every
single
request.
D
B
D
Yeah,
so
you
don't
have
to
kind
of
rather
than
having
to
manually
kind
of
like
fiddle
about
with
the
authentication
for
every
request.
That's
been
imported.
It'll
set
the
client
id
the
redirect
url
on
all
the
requests,
that's
important
that
it's
supported
and
it
will
set
it
to
an
environment
variable.
C
Yeah,
I'm
saying,
like
that's
pretty
nice,
to
have
on
the
environment
because
you
can
reuse
it.
It's
a
powerful
feature
actually.
B
D
Yeah
so
yeah,
you
can
see
that
it's
yeah,
so
property
scheme
is
not
expected
to
be
here,
but
if
you
didn't
do
that,
I
think
the
previous
set
up
insomnia.
I
think
the
previous.
The
way
insomnia
previously
worked
is,
if
you
didn't
specify
that
scheme
borrower
it
refused
to
yeah.
It
wasn't
happy
importing
the
oauth
settings
at
all.
B
B
Let's
see
6
0
auth
two,
there
we
go
or
if
this
is
a
little
small
here,
let
me
make
it
legible.
So
I'm
gonna
add
your
remote
and
then
check
out
this
branch.
B
B
Sweet!
Oh
yeah,
because
it's
mostly
yeah
that
that
open
api
file
doesn't
get
touched
a
lot
and
then
the
rest
are
fixtures.
Okay,
so
cool!
Let's
it's
recompiling.
B
B
Yeah
I
mean
this
is
the
the
there's
not
really
too
much
that
changed,
which
is
really
cool,
so
it's
like
I
love
pr's
by
the
way
that
are
like
80
tests
10
to
20
code,
so
I
think
I'm
probably
gonna
have
to
yeah
I'll
delete
this
one
to
regenerate
it
and
then
we'll
come
back
over
here
and
see.
So
it's
not
it's,
not
the
working.
B
Oh,
when
you
like
actually
import
the
spec,
oh
yeah,
okay,
so
then
that's
like
a
like
a
bit
of
a
bug
there,
but
maybe
it's
unrelated
and
we
can
handle
it
separately.
So
I'm
going
to
come
over
here
and
say
import
from
file.
Okay!
Actually,
let's
import
from
clipboard
new.
A
B
It
may
have
been
something
changed
since
you,
since
you
did
this
work,
so
I
didn't
push
on
purpose.
Let
me
grab
your
let's
solo
here,
I'm
going
to
solo
your
what's,
it
called
oauth
2
there
solo
that.
C
B
B
Okay,
let's
we
can
test
this
out
real
easily.
Let's
just
let's
go
grab
your
pr,
it's
here
somewhere.
Where
is
it?
Oh,
that's
my
local!
I
want
to
see
yours
here,
let's
check
it
out
again
yep
and
I
also
want
to
see
develop
on
origin.
B
B
Oh
because
something
has
changed,
some
pretty
significant
things.
Okay
run
hard,
something
has
changed.
Probably
let's
we
can,
while
that's
running
in
the
background,
we
can
check
out,
take
a
peek
at
this
file,
so.
B
Oh
and
12.18
yeah
we've
upgraded
node
since
this
time.
Okay,
again,
oh
learning
that
found
okay.
B
B
D
B
D
B
So
we're
good,
so
I
don't
think
anything
changed,
I'm
not
sure
why
it
didn't
work.
But
if
we
find
out
that
the
the
spec
that
we
were
testing
with
also
doesn't
generate
the
right
thing,
then
it's
likely.
We
just
have
something
wrong
in
our
specs.
So
let's,
let's
prepare
for
that
scenario,
and
so
you
added
this.
D
B
B
Right
these
are,
these
are
more
oh,
okay,
but
this
is
not
really
the
right
way
of
looking
at
it
at
the
problem.
I
think
because
this
is
showing
it
already
added,
but
with
the
environment
variable,
that's
testing,
like
environment
variable
stuff,
but
we
want
to
see
the
importing
if
it
comes.
D
B
C
So,
just
to
do
a
small
like
reiteration
on
on
the
theme
in
open
api,
you
can
define
an
authorization
component
right,
so
you
can
reuse
it
in
your
in
your
requests.
Let's
say
yep
and
here
what
we're
doing
is
basically
we're
adding
this
barrier
scheme,
so
we
can
order,
then
the
the
auth
window.
C
When
someone
like
goes
to
insomnia,
they
can
see
the
authorization
method
for
that
specific
request
and
currently
that
mapping
doesn't
work.
So
we
wanna
make
sure
it
also
supports
this.
D
C
D
B
Them
got
it
do
we
know
it'd,
be
cool
to
know
which
one
it's
working
for
if
we
can
figure
that
out.
But
let's
try.
Let's
try
with
this,
because
this
okay,
so
this
is
the
one
that
I
grabbed
before,
but
let's
migrate
from
designer.
No,
so
let's
just
like
delete
this
and
and
then
go
oh
well,
we
have
to
delete
it
from
something
where
it'll
generate,
so
the
flow
for
generation
should
be
the
same
as
importing
is
one
of
my
presumptions
there.
It
is
so
something
has
changed.
B
This
is
exactly
the
one
we
thought
wasn't
working,
so
we
are
not
going
to
bother
rebasing
this
right
now.
Let's
just
try
to
get
to
the
heart
of
this
job
and
I'll
go
back
later
and
try
to
find
out
what
happened
and
why
it
doesn't
work
after
being
rebased.
There
could
be
a
genuine
bug
in
insomnia
that
has
nothing
to
do
with
your
work
and
we
don't
want
to
get
distracted.
So
we
are
in
good
shape,
then.
B
So,
let's
take
a
look,
I
wish
we
could
see
them
side
by
side,
but
anyway,
let
me
make
this
a
little
bigger.
B
Oh
oops!
Is
it
is
that
as
big
as
we
can
get
okay,
this
is
as
big
as
I
can
go.
Really:
okay,
anywho!
Here
we
have
security
scheme,
oauth2
authorization
code
flows
great
and
then
it
imports
correctly
grant
type
authorization
code,
and
then
this
links
to
an
environment
variable
right
right.
D
Yeah,
so
I
think
it
what
if
I'm
ever
looking
at
the
pool,
if
you
look
at
the
pr,
I
think
it's
stated
that
I've
added
it
for
redirect
url
as
well.
B
That's
the
redirect
url
sorry
redirect
url
for
author
requests
to
a
variable.
This
is
a
required
parameter
of
amazon.
Cognito!
Oh,
are
you
using
cognito,
that's
cool.
I
like
cognitive
octa,
is
also
cool
so
being
able
to
set
it
easily
for
all
requests.
So
so
we
sort
of
you
made
a
you
made
a
like
a
design
choice
about
what
that
variable
should
be,
or
did
it
come
from
the
spec.
D
I
made
the
design
choice.
I
I
named
it
based
on
what
the
existing
variables
were
called
so
there's
already
off
to
client
id
and
client
secret,
so
I
just
went
with
off
to
redirect
url.
B
D
D
B
B
That's
cool
so
so,
let's
grab!
Let's
then
I
don't
want
to
solo
on
develop.
Let's
just
look
at
your
pr
straight
up
so
yeah.
This
is
the
only
code
file.
So,
let's
see
if
that
happened
here.
B
And
if
these
two
are
here,
then
what
is
this?
What
is
this
reducer
security
schemes?
Oh,
this
is
yeah
this
is
it
yeah,
okay
right?
This
is
what
you're
saying
is
that
oauth2,
username
and
password
were
already
added
you
just
added
this
redirect
url1,
but
the
auth
client
id1
was
already
added
to
okay
cool.
B
I
understand
what
that
says
now
and
I
didn't
before,
and
then
here
we're
linking
it
boy
it'd
be
nice
if
we
didn't
write
a
string,
it's
kind
of
error-prone
to
put
a
string
in
there.
I
wish
we
had
pulled
these
things
out
before,
but
I
don't
want
to
that's
okay.
I
don't
wanna
get
too
fancy
with
it,
but
okay.
So
if
the
grant
type
is
authorization
code,
then
fill
okay,
so
this
is
saying
fill
out
the
form
basically.
D
B
C
Well
kind
of,
but
you
know
I'm
still,
I'm
still
getting
to
the
sense
of
how
the
whole
feature
works,
but
I
understand
the
big
picture.
At
least
I.
B
D
Oh,
let
me
pull
up
the
specs
for
for
open
api,
okay,
which,
funnily
enough,
I
was
writing
an
opinville
study
of
the
night.
So.
D
I
know
I
was
just
on
my.
I
have
a
laptop
if
you
find.
B
It
but
okay,
so
this
is
let's:
let's
look
so
oauth
2
is
a
security
scheme
type,
so
this
is
saying
this
code
is
saying:
if
the
I
make
this
any
bigger,
I
can't
make
this
any
bigger.
So
let's
go
to
open
api
3
and
then
what
is
it
line?
330.
B
D
Yeah
so
oh,
I
really
wish
specs
for
this
actually
had
a
proper,
deep
links.
C
James,
so
currently
we're
checking
if
the
security
scheme
is
like
if
the
scheme
type
is
security,
type
or
oauth,
and
then
we're
running
this
logic,
but
before
we
were
doing
it
for
every
every
scheme,
type
right.
C
A
D
Yeah
so
I
removed
the
borough
check
if
you
look
at
the
link,
I've
just
put
in
the
private
chat
that
will
take
you
to
the
open
api
specs
for
the
security
objects
and
you'll
see.
That
is
it
scheme
that
should
only
reply
to
http
off
it
shouldn't
apply
to
oauth
so
yeah.
If
you
look
at
the
third
column.
B
D
D
I
think
I
think
there's
some
quite.
I
think
this
is
quite
clever.
Oh
a
few
people
also
kind
of
like
in
the
book
in
a
in
the
book
associated
with
the
chat,
who
were
kind
of
like
saying
yeah
that
that's
kind
of
wrong.
D
Yeah,
let's
go
right
at
the
top.
B
Showing
us
oh
so
this
is
okay.
So
then
maybe
I
do
understand
so
this
check
is
looking
for
something
that
should
be
an
unreachable
case
and
we're
forcing
users
to
add
bearer
when
otherwise,
it
really
shouldn't
be
there.
Yeah
and
the
linting
is
correct.
The
linting
understands
that
bearer
shouldn't
be
there,
and
so,
in
order
to
actually
get
this
to
work,
you
have
to
have
a
lint
error,
yeah.
B
So
so,
okay,
so
then
the
fix
is
to
take
off
bearer
because
that's
an
invalid
state
anyway.
Okay,
so
I
just
want
to
do
one
quick
thing
before
we
go
forward
which
is
to
which
is
this
open
api
three
was,
I
actually
was
I
well,
let's
go
find
it
again
open
api
three,
it's
always
ooh
nope.
This
is
wrong.
Open
api
3.,
it's
always
like
good
to.
B
B
B
This
is,
this
is
typescript
phase
one
and
two,
so
I
can
go
to
this.
I
can
go
to
this
commit
actually
here
and
then
I
can
go
to
parent
and
browse
files,
so
we're
gonna
do
a
little
get
splunk
just
for
a
second.
If
you'll
entertain
it
oops
import
open.
Three.
B
B
B
B
So
I'm
always
we're
always
on
the
lookout
for
something
because,
like
there's
always
there's,
usually
too
many
things
for
like
somebody
to
know,
we're
always
trying
to
reduce
that.
Of
course
we
want,
we
really
want.
I
mean
that's,
why
we're
doing
this
stream?
We
really
want
this
project
to
be
something
that
people
feel
like
they
can
contribute
to
for
our
own
sanity
as
well.
It
helps
keep
things,
keep
the
code
base
clean,
but
okay.
Well,
I
think
we
did
the
diligence
on
making
sure
that
there
wasn't
some
reason.
B
Why
bearer
was
there
in
the
first
place
and
the
only
thing
I
kind
of
want
to
do,
but
I
maybe
was
not
appropriate
to
do-
is
figure
out
this
I'd
like
to
make
like
an
is
supported,
scheme
function
that
can,
what
do
you
guys
think
is
that
is
that
out
of
scope
I'll?
Do
it
right.
C
Before
we
go
there,
I
would
like
to
continue
continue
like
on
the
flow
on
what
we
changed
and
what
is
there
just
to
get
a
sense
of
what
what
we
are
doing?
Okay,
then,
perhaps
we
can,
we
can
see
if
we
can
improve
it.
So
if
it's
needed
right,
like
perhaps.
C
Sort
of
good,
so
after
after
the
scheme,
we're
checking
if
there's
like
a
flow
authorization
code
and
flow
implicit
and
then
we're
putting
in
the
accumulator
and
os
to
redirect
url
to
localhost,
so
I
would
need
like
if
I,
if
I
entered
that
code
path,
I
would
need
some
explanation
on
why
this
is
like
happening.
C
D
Yeah,
I'm
just
kind
of
flipping
back
and
forth
between
this,
so
I
can
kind
of
see
your
screen.
I've
just
got
your
I've
got
the
stream
kind
of
like
full
screen.
So
I
can
okay.
B
Is
it?
Is
it
super
blurry?
I
don't
know
what
yeah
I
wish
it
wasn't.
I
don't
see
anything
on
my
side
that
could
make
it
less
less
bad,
but.
B
Well,
anywho
cool,
so
so
james.
What
what
do
you
think
we
should?
How
do
you
think
we
should
test
that.
C
It's
more
of
a
question
like
why,
if
you
go
to
the
code
line,
397
we're
doing
a
check
yeah,
this
is
this
is
something
that
was
added
in
this
pr
right.
So,
what's
like,
why
is
this
a
localhost
redirect
url?
If
one
of
those
two
things
is
true?
C
B
I
have
a
question
before
gats
tries
to
answer.
If
he
can,
would
this
make
it
better.
D
D
D
This
is
the
redirect
url
when
you
are
sent
to
your
render
when,
when
you're
sent
to
your
identity
provider,
this
is
the
redirect
url
that
you're
going
to
come
back
to,
and
it's
also
the
one
that
you'll
see
specified.
If
you
click
on
like
the
of2
tab,
the
authentication
tablet
you've
got
where
it
matches
up
with
the
redirect
url
there
as
well.
D
C
Okay,
so
there's
some
logic
here,
so
these
are
like
always
two
variables
as
far
as
I
can
see
on
the
on
the
code
and
we're
gathering
them
from
the
from
the
schema
that
we
have
and
based
on
what
is
available
from
the
schema
and
when
what
the
authorization
type
is
going
to
be.
We
insert
different
variables
to
the
environment.
A
D
D
You
you're,
not
you
don't
you're,
just
passing,
if
I
remember
rightly
you're,
effectively
just
kind
of
passing
actual
credentials
through
there's,
no
there's
no
redirect
involved,
and
this
is
more
of
kind
of
like
a
an
information,
in
fact
yeah.
If
you
click
on,
if
you
choose
the
grand
type
of
client
credentials,
you
can't
enter
a
redirect
url
in
in
insomnia
or
either.
B
B
Yeah-
and
they
only
get-
I
just
verified-
they
only
get
set
if
they
weren't
previously
there
so
like.
If
I
have
I,
I
change
it
to
asdf
and
if
I
go
back
and
forth
here,
it's
going
to
it's
going
to
leave
it
there,
but
if
I
remove
it,
what
I
should
see
what
I'm
expecting
to
see
anyway.
A
D
B
So
end
point:
let's
delete
these
other
ones,
so
we
don't
get
confused,
I'm
very
easily
confused
guys.
James
knows
this,
but
design
document
wait
a
second
no,
which
one
is
the
one
we
just
did.
B
D
Yeah,
this
is
the.
I
think
this
is
good.
This
is
a
this
was
like
a
test
case
where
we
had
lots
of
different
authentication
schemes,
so
lots
of
these
other
kind
of
paths
in
the
open
api
importer
will
have
kind
of
get
pulled
in
for
that
yeah.
This
is
yeah.
It's
the
one
with
lots
of
stuff.
C
I'm
curious-
and
this
has
nothing
to
do
with
the
pr,
but
I'm
curious
what
happens,
because
I
can
see
that
the
variables
get
accumulated
and
overridden
by
the
next
scheme.
So
what
would
happen
if
you
had
like
two
schemes
that
that
intervene
with
each
other
like
if
you,
if
you
have
something
that
has
authorization
code,
something
and
then
something
with
a
password?
C
Yeah,
that's
a
good
good
question.
C
C
B
B
Does
that
mean
we're?
Okay,
because
I
guess
it
could
mean
like?
Is
there?
So
maybe
you
could
rephrase
your
question
like
this?
Is
there
a
situation
where
having
authorization
code
and
implicit
at
the
same
time
means
that
you
would
not
want
to
set
this
variable.
B
C
Yeah,
exactly
I
think
we
can.
We
can
run
a
quick
experiment
if
you,
if
you
duplicate
the
scheme
and
if
it
makes
sense
in
an
open
api
spec,
do
you
have
like
two
yeah.
C
C
D
B
B
Well,
it
doesn't
mean
that
it
is
valid
because
it
doesn't
error,
you
know,
maybe
there's
just
not
a
rule
for
it.
What
is
this
one?
This
is
authorization
code,
but
it
doesn't
matter
in
the
end
anyway,
because
still
it's
going
to
be
set
to
the
same
thing,
one
way
or
the
other,
so
the
code
path
that
we're
looking
at
is
irrelevant,
but
I'm
now
I'm
curious.
B
B
B
B
C
B
C
C
C
You
go
to
the
design
spec
like
the
api
spec
and
see
that
specific
request.
If
you
find
it
whatever,
we
were
looking
before.
C
C
B
All
right,
so
I
can
remove
these
two
but
okay.
Well,
that
was
cool.
I
learned
something
I'm
curious
why
the
all
doesn't
work
like.
I
think
this
is
an
insomnia
problem,
not
a
like
open
api
type
problem,
but
if
I
try
to
switch
here
to
like
whatever
something
else
it
says,
current
authentication
settings
will
be
lost.
B
I've
never
hit
that
code
path,
so
this
is
something
in
the
drop
the
drop
down.
If
the
type
changes.
B
B
B
I
bet
you
that
the
data
structure
we
have
for
this
can
store
all
the
information
that
we
need
and
we
could
get
it
well
anyway.
We
have
15
minutes
left,
let's
try
to
see
because
it
clears
it.
It's
all
that
stuff
is
empty
now
when
it
shouldn't
be,
whereas
if
we,
if
we
remove
this
just
for
fun,
for
example-
and
we
do
something
that
doesn't
require
us
to
change
the
type
like-
let's
let's
say
we
do,
this
then
go
back
all
will
appear
and
it
will
have
the
right
stuff.
B
But
when
we
change
type,
so
if
I
go
to
basic
auth
yeah,
it's
going
to
tell
me
the
same
thing
anyway:
whatever
that's
fine,
let's,
let's
not,
let's
not
get
too
stuck
on
that.
So
what
needs
to
happen
for
this
pr
to
merge?
I
mean
I'm
okay,
we
don't
need
to
do
my
helper
function.
I
I
guess
it's
just
this
gets
a
little.
Let's
can
we
try
for
two
seconds?
Is
everyone?
Okay,
with
that
james.
C
Is
there
is
there
something
else
I
would
ask
guys?
Is
there
like
something
you
would
feel
comfortable
with,
and
what
would
you
feel
uncomfortable
like
what's
that?
What's
the
highest
risk
of
this
vr
in
your
opinion,
because
you
you
had
to
delve
into
this
and
figure
out
all
these
weird
cases
for
the
open
api
space?
C
So
what
what
makes
you
free?
What
like
is
that
is
the
biggest
concern
here
that
something
could
break
or,
or
you
know
something
could
go
wrong.
Yeah.
D
D
So
yeah,
that's
not
wonderful,
but
I
think
if
you
want
to
talk,
if
you're
going
to
tidy
that
up
and
check
some,
maybe
chuck
some
unit
tests
around
it,
it
probably
starts
to
feel
in
a
much
cleaner
state,
the
accumulator
stuff
setting
the
defaults.
I
don't
think
that's
too
much
of
an
issue
or
I
certainly
hope
it's
not,
and
I
think
that
the
the
removing
that
bearer
check
that
that's
it's.
D
C
I
agree,
I
would
say
like
this
one,
so
so
the
worst
thing
that
can
happen
there,
at
least
in
the
in
the
checking
for
the
security
type
oauth,
is
that
we
don't
generate
the
variables
right.
So
that's
not
a
big
deal,
I
would
say
in
the
things
that
could
be
broken.
That's
not
like
a
worst
case
scenario
right,
so
it's
okay,
someone
will
not
have
their
their
predefined
variables
there
yeah.
D
I
mean
I've
updated
the
text
for
the
the
test,
fixtures
and
the
tests.
I
I
seem
to
remember.
I
ran
them,
although
the
ci
will
run
them
anyway,.
D
C
A
bit
so
I
think
the
biggest
the
biggest
thing
that
changes
here
is
you
might
like
if
things
go
south
like
if
you,
if
you
were
using
the
better
scheme.scheme
thing,
what's
gonna
happen
to
you,
you're
not
gonna,
get
the
variables
generation
thingy.
Oh.
D
C
Exactly
so
so
it's
not
that
we're
breaking
something
it's,
I
would
say
we're
moving
towards
spec.
You
know
what
the
spec
does.
The
biggest
concern,
then,
would
be
this
out
scheme
fine
thing
which,
like
changes,
how
how
things
are
going
on,
but
I
think
there's
like
test
cases
to
cover
this
too
so
yeah.
B
B
B
It
means
we
can
unit
test
it
now
or
later
if
we
want,
but
let's
review.
So
if
the
scheme
is
let's,
let's
see
if
we
can
just
do
this
a
little.
B
C
C
B
B
If
it's
not
one
of
these,
then
return
false
okay.
If
it's,
this
then
return
true
and
if
scheme,
oh
okay,
so
scheme
is
only
in
oauth
types.
So
really
we
would
anyway,
the
way
to
do
this
nicer
would
be
to
to
let's.
Let's
do.
This
is
oauth
security
scheme,
because
oauth
security
scheme
is
the
only
one
that
has
scheme
in
it,
or
at
least
the
others,
don't
necessarily
have
scheme
in
it.
But
oauth
does
right,
there's
all
these
ones.
C
B
D
B
Would
say
it's
the
same
except
it's
an
or
here
I
I,
I
stashed
my
changes.
So,
let's,
let's,
let's
don't
save
that.
D
B
D
You're,
if
the
author
types
http,
then
you
need
to
choose
an
iona
authentication
scheme
which,
what's
in
so
yeah
one
of
basic
borrower
digest
or
that
kind
of
stuff.
C
I
think
that
is
supported
name
in
the
function,
kind
of
of
lies
about
it,
because
this
is
basically
if
this
is,
if
the
scheme
you're
passing
me
exists
in
the
defined
schemes,
which
is,
I
think,
the
the
supported
schemes
we
find
them
from
the
spec
yeah.
If
this
thing
is
there
and
follows
these
rules
or
some
of
these
rules,
then
we
get
an
out
scheme.
Otherwise
you
don't
get
an
ounce
scheme.
I
think
that's
like
the
the.
C
What
this
constant
authentication
thing
is
trying
to
to
wrap
as
logic
and
the
the
find
function
is
a
bit
complicated.
Perhaps
we
can
simplify
it
because,
like.
B
A
C
Six
months
later
or
three
days
later,
I
would
be
confused
to
add
an
extra
case.
B
Yeah,
I
think
I
think
that's
yeah.
Well
anyway,
we're
approaching.
C
A
B
Not
particularly
maintainable,
I
agree
so
then
I'll
I'll
go
after
this
and
rebase
this
and
push
it
up,
and
I
think
we
should
be.
I
don't
think,
there's
anything
that
needs
to
be
done
to
merge
this.
Do
you
do
you
have
anything
james.
C
No,
I
think
that
the
improvements
we
can
make
afterwards.
We
need
to
think
about
a
little
bit.
B
Okay-
well,
that's
just
about
our
time,
so
thanks
gaz
for
coming
on,
really
appreciate
it.
This
has
been
super
fun.
It's
we.
We
love
to
interact
with
people
from
the
community.
This
is
like
probably
my
favorite
part
of
this
job
and
it's
it's
really
nice
to
have
this
opportunity
to
to
hear
what
you
think
and
what
you
were,
what
you
were
doing,
while
you
were
trying
to
figure
this
out.
B
So
thanks
again
for
coming
on
and
thanks
everybody
who's
watching
on
youtube
and
elsewhere,
and
we'll
see
you
guys
next
week,
cool.