►
From YouTube: Code Review Weekly Workshop - Sep 02, 2022
Description
In this session we discuss some topics pertaining to code review!
0:00 - Code Review Show & Tell & Questions
27:22 - Code Review Pairing
A
A
So
let
me
share
my
screen
so
that
we're
looking
at
the
agenda
all
together,
yeah,
we
have
a
meager
agenda,
but
that
doesn't
mean
that
we
might
not
tap
into
a
gold
mine
of
conversation,
but
I
was
I
ran
into
this
this
last
week.
I
always
find
mrs,
like
this
interesting,
would
love
to
hear
just
your
all
perspective
on
how
you
would
approach
reviewing
mrs
way.
So
let
me
open
this
up.
A
Doug
sol
did
a
fantastic
job,
putting
together
an
mr.
That
concludes.
An
experiment
and
experiments
are
kind
of
like
feature
flags.
So
there's
like
a
candidate
and
a
whatever
the
opposite
of
candidate
is
a
control
group
and
there's
a
whole
code
branch
that
we're
going
to
remove.
Now
that
we've
concluded
the
experiment,
so
57
files
changed,
you
know
removing
all
sorts
of
large
amount
of
lines
which
is
lovely
omars
to
see,
but
I
don't
know.
But
how
would
you
approach
reviewing
an
mri
like
this?
A
A
C
A
B
Yeah,
I
mean
it's
kind
of
what
I
do
when
I'm
adding
stuff
in
different.
Like
I'm
working
on
something
at
the
moment,
that's
got
a
lot
of
changes
and
even
though
all
the
specs
get
updated-
and
you
know
you
had
you
had
some
automated
tech,
it's
kind
of
like
just
as
a
sanity
check,
just
kind
of
click
around
a
bit
to
make
sure
everything
looks
like
it's
working,
normally
yeah
but
yeah
I
mean
I
don't
I
don't.
B
C
B
B
A
A
We
need
to
kind
of
just
remove
everything
on
on
the
other
side
of
it,
so
he
did
a
great
job
of,
like
I
think,
directing
back
into
how
to
review
it.
Some
other
things
stick
out
to
me
is
is
sometimes
we'll
have
an
mr
of
like
okay,
we're
going
to
remove
the
feature
flag
now
but
like
it
hasn't,
even
been
enabled
on.com
yet
and
like
so
just
from
a
process
perspective.
E
E
A
F
A
One
one
way
of
looking
at
this
too,
because
I
do
think
it's
helpful
going
for
you
know
a
line
for
line
read-through,
even
though
I
don't
envy
any
back.
Ender,
that's
going
to
read
all
2000
removed
lines.
You
know.
If
we're
removing
the
whole
file,
you
don't
have
to
read
the
whole
file.
That's
maybe
that's
worth
clarifying
the
well.
A
A
It's
really
easy
to
test
like
did
this
thing
like
does
this
thing
not
exist
anymore,
and
you
can
also
like
just
follow
through
the
the
same
trail
that
doug
probably
followed
when
you
put
this
together
of
like
okay,
if
we
remove
this
entry
point,
what
all
was
that
entry
point
using?
Are
those
modules
need
anymore?
Can
I
remove
those
modules
and
like
there's
a
so
from
the
front-end
perspective?
There's
a
there's,
a
lot
that's
removed
here
like
even
this.
You
know
here's
some
components,
so
these
are
modules
used
to
you
know
power.
A
Certain
views
could
be
reused
elsewhere,
but
it's
not
so
now
that
this
isn't
used
at
all.
We
need
to
remove
it
and,
unfortunately
I
don't
think
we
have
any
linting
that
actually
detects
whether
modules
are
used
or
not
man.
That
would
be
that
can
actually
be
nice.
It's
worth
writing
an
issue
about.
I.
B
G
C
A
A
A
But
whenever
we're
removing
code
there's
this
references
that
you
can
just
start
searching
on
and
you
can
find
you
know
usually
find
you
know
a
couple
of
things
that
just
slipped
through
the
cracks.
The
other
thing
was
I'll.
Do
a
project
wide
search
for,
like
the
keywords
related
to
what
we're
removing
and
so
like
this
whole
experiment
and
everything
was
called
combined
registration.
A
So
I
did
a
project
wide
search
for
combined
dot,
star
registration
and
that
came
across
a
reference
to
a
file
that
was
going
to
be
removed,
but
isn't
isn't
around
anymore.
So
we
could
hey,
there's
still
more
references
that
we
can
find.
Probably
the
most
interesting
thing
that
I
was
I
managed
to
to
find
here
was
doug
accidentally
and
removing
something
that
wasn't
used
anymore
removed
the
spec
for
it,
but
didn't
get
to
commit
actually
removing
the
underlying
function
that
the
specs
were
for.
A
So
we
had
just
these
changes
on
the
spec
file
without
it.
Actually
you
know
the
reflecting
change
being
on
the
unit
file,
so
that's
a
that
symmetry
of
like
we
removed
something
from
the
unit.
I
expect
to
be
removed
from
the
spec
like
some
whenever
that's
off,
that's
something
to
look
for.
We
touched
the
spec
file,
but
we
never
touched
the
unit
as
a
sign
that
maybe
something
was
overlooked.
Yeah
cool
thanks.
Let
me
share
that.
I
think
these
mris
are
interesting.
E
Absolutely
I'm
adding
on
to
that,
because
you
just
mentioned
the
dead
code.
If
you
are
opening
up
the
agenda
I
this
week,
I
was
an
author
of
mr
and
anna
did
an
excellent
maintainer
review
on
this,
and
if
we
actually
encountered
a
so
called
that
test
or
I'm
referring
to
it
as
that
test
and
which
I
think
is
quite
a
mean
mean
thing
to
encounter
in
very
short,
what
happened
here,
we
are
expecting
some
fit.
E
This
is
this
was
a
medium
medium-sized
refactoring,
so
in
this
test,
we're
expecting
something
which
is
no
longer
there
and
we
are
expecting
it
to
be
something
which
is
also
no
longer
there.
So,
at
the
end
of
the
day,
we're
expecting
undefined
to
be
undefined,
which
will
be
true.
This
test
was
absolutely
passing,
but
it
was
still
quite
rubbish.
E
A
It
is
a
really
tricky
generic
assertion
because
it
accepts
a
whole
lot
and
it
could
very
easily
go
out
of
style
when
the
underlying
code
changes
yeah
thanks
for
bringing
it
up.
That's
a
really
great
example.
So
in
this
case
we
were
wanting
to
make
sure
that
the
length
we
wanted
to
make
sure.
I
guess
that
these
links
weren't
empty,
but
it
looks
like
it's
just
not
that
kind
of
it's
look
I'm
guessing.
This
is
an
object,
not
even
a
ray
right.
E
A
E
A
E
I'm
happy
to
drive
you
if
that's
okay
for
you
yeah,
that's
mine,
shouldn't
be
too
much
clicking
around.
So
one
thing
I
run
into
continuously
and
as
a
disclaimer,
I
am
not
the
most
sophisticated
deaf
when
it
comes
to
to
graphql,
like
I
know
what's
going
on,
but
I've
never
really
never
really
digged
into
it
very,
very
deep.
E
So
reviewing
these
things
with
with
graphql
changes
is
kind
of
challenging
to
me,
and
I
be
curious
about
how
you
consider
these
kind
of
things,
but,
like
I
can
debug
a
query
if
you
go
for
the
changes,
yeah
the
changes
tab.
There
is
exactly
like
a
query
is
now
introduced,
which
obviously
there
has
been
interface
defined
for
it
already
because
it
works,
and
that
is
something
I
refuse,
and
I
could
check
that
we
are
kind
of
computing
all
the
values
getting
from
the
graphical
query,
correct.
G
E
A
A
C
F
A
A
A
A
A
A
E
G
A
E
C
A
E
E
It
is
basically
the
decline
would
subscribe
to
any
kind
of
value
and
then
the
server
without
any
other
interaction
would
tell
let
the
client
know
and
that
this
thing
has
changed
and
then
just
update
has
somebody
ever
encountered
this
and
developed
on
our
code
base
because
I
feel
like
that
is
that's
a
good
opportunity
for
good,
good,
x,
behavior
or
all
the
polling.
We
are
still
seeing
it's
probably
the
most
spotting
code
base.
I've
ever
touched.
A
D
B
E
A
Historically,
we've
had
and
the
I
know
that
the
back
end
has
needed
to
figure
out
how
they're
going
to
solve
the
websocket
problem.
I
think
that
they've
approached
ways
to
solve
it,
but
it
probably
adds
a
little
bit
of
complexity
to
the
interface,
but
I'm
not
entirely
sure.
I
really
don't
know
so.
That'd
be
okay,
yeah,
that's
a
great!
That's
a
great
question
and
yeah:
we
don't
do
it
a
whole
lot.
We
probably
should
consider
using
it
instead
of
pulling
if
we
need
to
do
it
and
yeah.
That's
a
great
question.
G
A
A
A
Let
me
let
me
sign
in-
and
I
have
a
handful
of
mrs,
that
are
somewhat
non-trivial,
so
I
think
that
should
be
interesting
yeah.
I
was
really
wanting
to
look
at
this
one.
I
will
share
my
screen
and
you'll
help
me
pick
all
right.
This
is
an
mr
I'm
reviewing
the
front
end
of
it
and
the
front
end
change
is
really
small,
so
we
would
very
likely
be
able
to
finish
reviewing
this
ammar
during
our
time.
A
This,
mr
the
front,
end
change,
is
much
larger
and
it's
doing
something
fairly
significant
and
I
believe
that
there's
other
ways
we
can
achieve
some
of
that
this
one
this
mmr
will
take.
This
mmr
is
fairly
non-trivial,
but
and
then
there's
this
one,
which
is
quite
large,
I
feel
like
it's
large,
so
marcel
the
his
super
awesome
ux
manager,
but
he
also
contributes
code
often,
and
so
I
know
this
is
well
tested.
A
I
know
he's
like
and
he's
included,
so
many
screenshots
he's
included,
so
many
screenshots
I
had
to
like
scroll
down
a
whole
lot
to
get
to
the
okay,
so
I
have
a
feeling
this
is
well
tested,
but
the
one
thing
that
stands
out
to
me
with
this,
mr
is
it
is
touching
a
lot
and
I'm
a
little
concerned.
If
maybe
we
should
split
this
up,
that's
kind
of
what's
my
thoughts,
but
is
there
one
that
you
want
to
dive
into
together?
We
got
this
one
that.
B
E
A
E
A
C
A
A
E
Before
we
start
with
this
quick
question,
let's
what
do
you
think
about
highlighting
the
front-end
changes,
because
sometimes
on
hours
like
that
that
are
so
heavy
on
the
back-end
side,
they
were
like
changing
a
test
id
on
the
front
end
or
something
like
with
those
I
don't
even
bother
reproducing
I'll,
be
like
whatever
you've
been
doing.
This
looks
good
to
me.
Yes,.
A
G
E
A
A
A
A
G
G
G
A
A
D
G
E
It
is,
it
is
a
question
that
came
up.
That's
not
coming
back
to
me,
since
I
asked
this
earlier
this
week
to
myself
is
because
I'm
seeing
it
here-
let's,
for
example,
go
with
the
glw
full
css
class
in
line.
Seven
like
these
are
classes
from
our
design,
design
system,
that's
probably
or
utility
class.
That
is
50.,
I'm
looking
for.
How
would
you
approach
searching
for
those,
because
it
happens
to
me
too
often
that
I
kind
of
more
or
less
have
the
css
in
my
head
like
I'd?
E
A
A
A
D
A
A
E
A
A
A
A
D
E
A
See
I
use
pages
for
conventional
comments,
but
I
use
I've
just
registered
this
domain
with
gitlab
gitlab
set
up
the
gitlab
setup,
the
ssl
stuff,
all
sorts
of
stuff
with
it.
I
was
just
able
to
say
yes,
this
is
definitely
my
domain
and
they
verified
it
and
then
they
were
able
to
go
forward
with
it.
So
I'm
pretty
sure
it's
with
jesus,
okay,
it.
A
A
A
A
A
A
A
A
D
A
Do
you
remember
the
first
spider-man
with
toby
mcguire?
Remember
it's
a
big
word
but
yeah!
Okay,
do
you
remember
when
you
know
he
runs
into
the
he's
doing
his
first
fight
and
here's
that
focus
again
he's
doing
his
first
fight
and
the
the
guy
that
booked
them
was
like
hey
you're,
the
guy's,
the
peter
parker
was
like
you're
supposed
to
give
me
all
the
money
it's
like
yeah,
but
you
knocked
him
out
in
two
minutes.
A
So
you
only
get
this
much
and
peter
parker's
like
well,
that's
not
fair,
and
that
guy
says
I
miss
the
part
where
that's
my
problem
and
then
later
the
guy
gets
robbed
and
the
guy
and
the
guy
saying
hey
help
and
peter
parker
says
I
miss
the
part.
But
that's
my
problem.
So
that's
my
yeah,
that's!
What's
going
on
in
my
head
right
now,
when
I'm
like,
I'm
not
gonna,
I
don't
have
to
solve
this.
A
There's
a
pretty
good
someone
made
a
pretty
good
edit
of
that
spider-man
movie,
where
it's
all
these
clips
of
him
and
throughout
the
whole
movie,
but
they've
they've
just
played
that
audio
part
in
front
of
it.
It
was
like
they're
all
you
know,
they're
all
at
the
end,
sorry,
I'm
spoiling
it
forever
hasn't
seen
it,
but
if
they're
all
at
the
end,
green
goblins
died
and
harry
osborne
is
upset
or
whatever
and
peter
parker
says
I
missed
the
part.
But
that's
my
problem
like
this
is
really
funny
yeah
all
right.
E
C
A
F
A
A
D
D
D
F
A
A
G
E
A
E
A
A
A
That's
pretty
fine
yeah,
that's
pretty
fine!
I
yeah,
I
think
the
only
other
things
then
I
I
now
realize
why
these
rows
are
showing
up
in
the
body.
Maybe-
and
so
this
is
a
non-blocking
comment,
but
it's
because
this
is
actually
a
sibling
to
the
head.
For
some
reason,
I
thought
it
was
under
the
head,
the
t-head,
but
that's
because
I'm
just
going
to
leave
that
as
like
a
a
thought
non-blocking.