►
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
Okay,
so
it's
september
13,
2021
distribution,
team
demo,
and
today
I
wanted
to
show
you
guys
some
of
the
things
that
I
worked
as
a
kind
of
auxiliary
tools.
More
than
anything
that
I
utilized
while
I
was
performing
some
tasks
and
especially
the
three
hours
week,
so
I'll
jump
right
into
the
triage
related
stuff.
So
I'll
share
my
screen.
A
Okay,
so
as
I
outlined
in
that
in
the
demo
notes,
I
created
this
geo
workflow
kind
of
tool,
kit
project
things
that
we
do
manually
but
could
be
done
automatically
so
and
the
main
tool
there
is
this
gl
flow,
that's
the
main
engine
that
actually
it's
a
rather
longish
python
client
for
the
lab
that
can
go
and
do
certain
things,
and
there
is
the
readme
file
that
outlines
most
of
them.
A
Hopefully,
if
not
I'll,
update
this,
but
there
is
the
watch
list
issues
get
issued,
get
them
our
clone
labels.
Change
labels
apply
labels.
Oh
there
are
actually
there
is
no
close
issue
documented,
but
that's
okay.
It's
it's
implemented,
but
not
documented,
oh
well,
but
first
things.
First,
as
it
relates
to
the
triage.
A
So
for
the
triage,
it
assumes
certain
things
about
the
standard
and
how
you
maintain
your
triage
issue,
and
it
also
assumes
a
few
other
things.
So,
for
example,
like
I
was
watching
dj
doing
his
triage
week
like
I
was
looking
out
over
his
shoulder
and,
for
example,
what
he
does
first
thing
he
does.
A
He
actually
sets
up
the
watch
for
three
projects
for
the
gitlab
chart,
gitlab
operator
and
gitlab
omnibus
so,
and
he
has
to
click
manually
through
those
things
every
single
time-
and
I
was
like
that's
not
me-
I'm
not
doing
this,
so
I
put
this
stuff
in
place,
so
the
triagi
ammo
is
kind
of
like
quick
configuration
for
things
related
to
the
triage.
So
if
I
want
to
watch
something
all
I
have
to
do
is,
let's
see,
let's
do
the
watch
here,
so
I
can
do
watch.
A
And
it
automatically
will
it
informs
me
that
for
those
three
it
changed
setting
from
global
to
watch
on
all
those
things,
and
if
I
want,
if
it's
my
the
end
of
my
triage
week,
I
do
the
reset
flips
it
back.
It's
nothing.
Fancy
at
all
like
this
is
like
the
really
bare
bones
function,
but
then
I
figured
out
a
few
other
things
that
I
was
kind
of
tired
of
doing
because
I'm
lazy.
A
So
I
automated
the
labeling
of
the
issues,
so
you
can
have
sets
of
labels
because
we
do
have
certain
things
where
we
need
to
apply
multiple
labels
at
the
same
time
to
either
mr
or
the
issue.
So
I
just
define
those
sets
of
labels
and
the
tool
the
geoflow
can
actually
do
it
for
you.
Let
me
see
if
I
have
it
in
read
me:
get
an
issue
list
issue:
where's
labels
yeah,
so
change
labels.
You
can
do
because
the
command
is
a
little
bit
more.
A
And
so
yeah,
of
course,
so
that
you
know
it's
a
work
in
progress,
so
just
don't
throw
too
big
of
a
you
know
stones
at
me
for
that
not
being
that
much
ready,
but
let's
go
in
the
code
and
just
take
a
look
at
what
the
parameters
are:
change
labels
right
there
so
most
of
them
they
take.
Oh
that's
another
thing,
because
sometimes
I
had
to
deal
with
direct
reference
to
the
issue
or
mr,
so
you
have
that
really
short.
A
A
So
you
can
add
multiple
labels
to
the
existing
issue
or
you
can
delete
multiple
labels.
You
can
apply
that's
the
thing,
that's
there's
apply
labels
and
there's
change
labels
and
that
the
difference
between
the
two
is
apply.
It
applies
it
on
top
of
the
pre-existing
labels
and
change
labels
is
whatever
labels
you
have.
It
will
change
them
to
whatever
you
provide,
so
it
kind
of
resets
the
whole
thing.
A
A
Oh,
there
was
one
thing
that,
because
we
were
talking
about
this
in
context
of
issue
bot
when
issue
bot
can
actually
mark
multiple
issues
with
a
pipeline,
and
they
can
only
be
linked
to
a
specific
one
issue,
the
core
issue
that
we
are
trying
to
resolve
and
then,
when
we
close
one,
preferably
we
might
want
to
close
all
the
rest
of
them,
so
they're
not
just
dangling
around,
and
for
that
I
added
the
added
the
cascading
closure.
A
A
A
So
you
would
have
commented
on
the
workflow
on
issue
2,
that
is
closing
it.
Then
it
will
close
it.
Then
the
same
thing
would
happen
to
one,
so
it
kind
of
found
all
the
children
so
to
speak,
because
it
assumes
that
the
first
issue
that
you
provided
with
is
a
root,
and
then
it
will
close
everything
cascading
down
and
it
tries
to
detect
the
loops.
A
So
it
will
bail
out
a
loop
and
I
haven't
tested
on
really
complicated
setups,
but
it
seems
to
work
so
for
the
cascading
delete.
It
will
do
that
and
the
same
thing
with
the
cascading
comments.
So
if
you
want
to
do
multiple
comments
so
again,
if
we
have
a
lot
of
issues
related
to
a
particular
one-
and
we
just
want
to
comment
on
all
of
them
saying
this
is,
for
example,
this
is
inapplicable
to
us
we're
going
to
close
this
in
so
many
days
or
whatever
you
can
do
that
too
with
that
tool.
A
So
those
are
two
mass
operations
that
I
kind
of
put
in,
and
the
mass
comment
was
just
the
side
effect
of
cascading
closure
because
I
already
had
the
functionality
so
might
as
well
just
put
that
in
quicker
labeling.
We
went
through
that
and
issue
mr
disambiguation
url
versus
reference.
I
kind
of
spoke
about
that
at
the
beginning.
A
Oh
then
enhanced
api
processing
of
issues
nmrs-
and
this
is
where
it
was
kind
of
interesting
to
me,
because
I
use
the
api
and
what
I'm
doing
I'm
actually
fetching
the
entire.
So
here's-
oh,
that
was
that
was
the
wrong
one.
Sorry,
my
json
is
probably
going
to
be
that
yeah.
So
when
you
fetch
the
issue
I'll
try
to
maximize
this,
if
I
can.
A
Nope,
okay,
it
doesn't
want
to
cooperate
right
now,
but
that's
okay,
so
what's
happening
here
is:
if
we
go
there,
it
puts
in
the
discount.
So
it
generates
the
json
object,
but
it
attaches
to
that
json
object.
The
discussions
there
are
actually
a
separate
api
call.
A
A
If
that's
all,
you
want
to
do
no
recursion,
just
direct
links
and
we'll
plug
them
in.
So
you
have
a
little
bit
more
background
on
the
issue
or
mr
that
you
fetched
and
you
want
to
operate
on.
So
I
saw
this
entire
thing
as
more
of
a
way
to
plug
into
and
yeah
the
pipeline,
for
this
project
actually
generates
the
image
that
you
could
use,
and
that
was
the
main
intent
is
something
that
we
could
potentially
use
within
our
workflows
and
pipelines.
A
But
that's
kind
of
far
out
immediately
it's
like
if
you
want
to
do
some
manual
operational
things,
that's
what
it
does
so
yeah,
that's
about
it
for
the
workflow
stuff
comments,
whatever
like
whatever
what
you
guys.
Feelings
are
on
this
subject.
B
I
really
like
the
idea.
I
think
that
it
helps
us
the
same
way
that
the
issue
bot
did
with
kind
of
bringing
to
light
some
inconsistencies
in
how
we
do
triage
or
opportunities
for
us
to
do
things
better.
So
I
think
especially
the
fact
that
your
program
has
dry
run
is
awesome,
because
we
can
kind
of
test
it
and
confirm
that
it's
doing
what
we
want.
B
I
think
when
I
do
triage
there,
especially
even
just
that
roll-up
feature
where
I
can
just
kind
of
summarize
all
the
issues
that
I
touched,
whether
they're
maintained
or
discussion
or
closed
or
whatever
is
immediately
helpful,
and
it
would
provide
some
consistency
for
these
triage
issues
and
then
the
more
we
can
automate,
even
if
it's
just
kind
of
finding
issues
with
without
the
labels
and
applying
them.
I
know
sometimes
the
gitlab
bot
comes
around
and
does
that,
but
I
don't
think
that's
always
consistent.
B
So
that
would
be
another
helpful
point
as
well.
I
don't
know
how
many
projects
we
have
on
the
distribution
team
that
are
in
python,
but
I
don't
know
if
that's
a
huge
it's.
A
B
A
The
other
thing,
and
thanks
for
mentioning
it
like
the
triage
report,
this
is
a
separate
script
that
actually
takes
that
json
that
that
the
other
tool
produces
and
it
extracts
all
that
requested
data,
and
I
made
it
smart,
so
it
consumes
both
types
of
our
triage
issues
when
you
actually
use
label
as
an
real
label
with
tilde
or
you
just
mark
it
with
a
pseudo
label.
So
it's
phrase
column
something
so
you
will
extract
either
way.
So
there
are
two
examples
like
one
I
extracted
dj.
A
I
used
you
as
a
guinea
pig
here,
so
I
extracted
from
dj's
issue
and
he
was
using
just
a
plain
text
stuff,
and
this
is
what
I
was
using.
I
was
using
the
tildas
and
it
doesn't
really
matter
because
when
we
pop
it
back
into
the
issue,
it
presents
itself
nicely
either
way,
so
it
tries
to
be
smart
to
a
degree
and
those
two
triage
summary
and
three
hours
week.
A
B
A
B
C
D
Any
other
comments
on
this
before
just
to
echo
on
you
know,
mitch
mentioned
the
word
consistent
or
consistency
a
few
different
times
and
that's,
I
think,
from
my
perspective,
the
like
the
more
consistent
we
are
across
the
team,
the
easier
it
is
to
report
and
show
people
what
we're
working
on
and
then
the
other
example
of,
like
hey.
Occasionally,
things
get
missed
because
they're
not
following
kind
of
the
known
the
known
pattern,
and
so
I
think,
having
tools
that
generate
these
patterns
goes
a
long
way
to
toward
keeping
those
from
falling
through
the
cracks.
A
Like
I
provide
enough
infrastructure
for
people
to
be
able
to
tinker
with
this,
especially
with
the
docker
file
and
the
image
being
immediately
available
for
people
to
kind
of
pick
up
and
see
what
they
can
do
with
it,
it
could
be
broken.
A
It's
a
proof
of
concept,
so
let
me
know,
put
the
issue
like
either
way
so
yeah
that
that's
how
much
I
have
to
say
on
this
particular
one
on
the
geo
workflow-
and
I
was
gonna
like
this
gl
flow-
was
the
base
kind
of
workhorse
of
this
entire
thing,
and
I
was
gonna,
expand
on
top
of
that,
and
you
would
notice
that
it's
fairly
generic,
it's
not
specifically
driven
for
one
task,
it's
just
all
sorts
of
operations
and
then
I'm
just
plugging
in
smaller
tools
to
extract
the
results.
A
I
want
and
my
ideal
scenario.
I
would
like
to
make
it
more
of
a
kind
of
pipeline
processor
when
you
can
specify
multiple
commands.
Get
issue
close
like
comment
on
the
issue.
Close
the
issue
kind
of
it's
a
longer
story
like
the
functions
are
not
there
yet,
but
it
would
be
interesting
because
I
like,
and
I
will
show
you
with
the
other
tool
that
approach
that
I
found
quite
useful
in
the
past
and
I'm
kind
of
recycling
that
idea
of
over
and
over
again.
A
C
So
it's
cool
so
apart
from
putting
the
same
format
of
comment
into
issues,
what's
the
what's
the
gain
here,
because
it
looks
like
it's
a
whole
lot
of
extra
commands
to
do
what
we
already
do
and
it's
already
done
with
macros
and
our
current
triage
setup.
So
I'm
trying
to
figure
out
what
this
adds
to
the
game,
because
it
feels
like
a
lot
of
boilerplate.
A
Okay,
how
do
you
two
things
at
the
start
and
if
we
just
talking
about
the
triage
just
triage
alone,
because
that's
probably
the
scope
of
your
question
and
within
triage
that
initial
step,
when,
like
I
said,
I
was
following
dj's
advice
and
you
know
his
flow
the
way
he
does
this,
so
it
may
not
be
applicable
to
everybody
but,
for
example,
setting
up
a
watch
on
multiple
projects
that
you
want
to
watch
and
every
single
time
you
have
to
remember
to
go
click
it
do
it
like
it's
all
through
the
gui.
A
This
way
you
can
actually
script
it.
You
can
have
your
own
script.
Saying
like
this
is
those
are
the
things
I
want
to
do
at
the
start
of
the
week
at
the
end
of
the
week
and
the
end
of
the
week,
it
actually
summarizes
everything.
Oh
one.
Other
thing
that
I
forgot
to
mention
is
that
I
added
let's
go
back
to
this.
A
The
triage
report
actually
will
look
at
the
labels
like,
for
example,
there
are
label
expressions,
so
you
can
put
the
minus
minus
in
front
of
a
label.
A
For
example,
I
put
accidentally,
I
put
the
for
scheduling
label
on
on
an
issue,
and
then
I
took
it
off
later
on
after
discussion
with
a
team
member
and
the
idea
was
okay,
so
we
leave
it
there
because
I
did
those
two
actions,
but
I
don't
want
this
counted
as
there's
one
extra
issue
for
scheduling
because
I
put
another
mention
of
it.
So
I
put
that
prefix
when
the
first
symbol
is
minus.
That
means
that
that
label
is
to
be
removed
from
the
reporting,
so
it
just
takes
it
out.
A
And
it's
entirely
up
to
you,
like
the
json,
the
triage
report
json
actually
generates
the
json
that
you
can
do
whatever
you
please
with
like
I'm
extracting
just
a
subsection
of
it,
the
tags,
but
it
has
the
reverse.
It
has
the
history
of
an
issue.
So
if
you
want
to
take
a
look
at
what
you've
done
with
the
issue
throughout
your
triage,
so
you
could
have
done
like
apply
tag.
A
C
I
guess
I
mean
that's,
I
don't
do
any
of
that
manually
at
all.
I
have
an
auto
calculated
tabular,
it
just
does
it.
So
that's
just
that's
why
I'm
kind
of
confused
because,
like
we're
supposed
to
watch
all
projects
all
the
time,
so
we
don't
turn
watches
on
and
off
so
like
turning
the
watch
on
the
watch
should
be
on
all
the
time
on
all
of
our
projects
and
then
we
use
our
automated
filters.
That's
part
of
the
handbook
and
then
for
this
you
know
when
we
do
our
triage
and
we
track
it's.
C
You
know
you,
you
keep
a
listing
of
what
you're
tracking
and
that
you
know
you
can
automatically
tabulate
that
so
I'm
just
struggling
because
to
me
this
is
an
awful
lot
of
things
for
stuff.
We
already
do
that's
already
in
our
handbook,
so
I'm
trying
to
find
the
what
we're
doing
here
and
you
know
again,
I
think
it
comes
down
to
what
we
need
is
a
docs
update
in
our
handbook,
because
it
sounds
like
there's
just
things
that
aren't
clear
and
documented
well.
Well,
because
we
already
have.
We
already
have
a
lot
of
this
covered.
A
As
far
as
triage
goes
yeah,
I
can
take
that
like
like
I
said,
I
only
took
a
sample
of
one
and
I
didn't
have
a
chance
to
kind
of
watch
more
people
how
they
do.
The
triage,
like
I
went
through
the
triage
page
and
tried
to
kind
of
follow
that
to
a
t.
Well
as
much
as
I
can
so
if
things
are
missing
or
if
things
are
not
necessary,
it's
fine
too.
It's
like
the
main
goal
here
was
not
specifically
the
triage
was
kind
of
the
inspiration,
but
the
tool
itself
was
made.
A
D
Again,
I'm
gonna
use
the
word
consistency.
I
think
we're
I
think,
on
the
tree
outside
right.
We've
identified
through
this
is
that
people
are
doing
it
different
ways,
and
so,
whether
or
not
everyone
uses
the
same
tool,
at
least
we
have
a
tool
that
represents
the
consistent
way
to
do
it.
If
folks
want
to
use
it,
that's
that's
where
my
head's
at
on
it.
Like
I
get
that
hey.
D
A
A
So
this
is
where
I
said
that
it
kind
of
follows
that
pipeline
executor.
So
I
within
the
script,
I
provide
a
multiple
different
functions
and
it's
up
to
the
end
user,
basically
to
combine
them
into
something
that
they
want.
I
kind
of
pre
pre-compile
some
list
of
invocations,
but
in
general
you
can
do
auto
deploy
so
like,
for
example,
there
is
deploy
operator
wait
for
operator
to
play
git
lab,
but
sometimes
you
just
want
to
do
the
deploy
the
operator.
So
all
you
do
is
just
auto
deploy
oh,
come
on
and.
A
And
that's
it
like
everything
else
you
know
is
done
so
it
kind
of
to
me
it's
easier
to
manipulate
the
operations
on
the
fly
more
kind
of
a
debugging
tool
as
well,
but
I'm
not
gonna
go
into
details
of
this
particular
script.
We
can
talk
about
it
next
time,
but
I
just
wanted
to
show
that
concept.
The
stuff
that
I
was
talking
about,
the
previous
one,
where
I'd
like
to
have
something
like
that
implemented
there.
A
So
you
can
actually
do
multiple
operations
in
the
same
indication
because-
and
I
will
explain
why-
because
I
implemented
caching
within
that
tool,
so
it
actually
can
do
everything
in
memory
without
committing
anything
back.
So
you
can
do
the
dry
run
and
all
your
say
modify
issue
is
going
to
be
reflected
in
memory
but
never
committed.
But
then,
when
you
do
the
get,
you
will
see
that
committed
result.
So
you
can
do
multiple
operations
within
the
same
and
it'll
be
all
discarded
at
the
end.
C
A
Though
the
auto
deploy
yes,
the
issue,
the
drill
flow
stuff,
it
actually
like
there's
an
entire
project.
So
if
anybody
wants
to
pop
any
issues
there
go
for
it
issue
boards
open
on
that
project,
yeah
like
I,
I
did
not
want
to
assume
that
the
group
is
willing
or
ready
to
take
this
too
long.
So
it's
in
my
private
place
space.
So
if
we
are
in
the
condition
to
move
it
over
to
the
more
common
space,
it's
fine
too,
but
for
the
moment
yeah,
it's
in
the
private
space.