►
From YouTube: Magento MSI Open Demo. January 19, 2018
Description
Agenda:
- @iminiailo - module InventoryIndexer introduced, indexation mechanism improvements
- @roma-glushko / @iminiailo - Make stock source link part of API (this PR unblocks by priority algorithm and return processing)
- @Yaroslav Rogoza / @vnayda - Status of `Low Stock Notification`
A
On
previous
demo,
I
will
show
you
several
prototypes
and
the
proof
of
concept
regarding
some
of
the
functionality,
for
example,
catalog
search
and
the
work
with
a
shopping
cart.
During
this
week,
we
spent
quite
a
lot
of
time
to
make
this
prototypes
said
to
be
unsatisfied
for
our
guidelines
and
the
code
code
requirements
we
have
internally
in
magenta.
That's
why
we
have
quite
a
lot
of
technical
stories
during
period
technical
stories
and
cost
during
the
current
week
during
the
current
sprint.
Actually,
so,
according
to
today's
agenda,
we
will
start
from
from
the
first
subject.
A
A
We
already
got
an
approval
from
our
counter
owner
from
our
intern,
that
we
will
update
source
item
always
with
an
update
on
safe
mode,
and
we
would
not
support
update
by
schedule
because
actually,
our
reservation
placement
and
the
work
with
the
inventory
is
a
synchronous
by
its
nature.
So
even
update,
update
by
update
on
save,
is
also
a
synchronous
operation.
So
we
no
need,
additionally,
to
support
update
by
schedules
that
would
be
just
confusing
and
actually
making
a
refactoring
of
that
indexation
code
and
the
older
code
devoted
for
the
indexation.
A
We
raise
that
question
several
times
regarding
similar
cases
with
Anton
krill
and
actually
that's
really
cool
when
it
has
the
index
in
there
in
a
separate
module,
because
the
indexation
as
well
as
it's
pretty
similar
like
with
the
cache
is,
should
not
be
in
the
same
entity
and
actually
entity
should
know
nothing
that
it's
being
cached
or
in
the
indexed,
because
this
is
just
a
supplementary
work
just
to
speed
up
speed
up
the
whole
process.
That's
why
actually
we
introduced
this
in
inventory,
indexer
model
and
potentially
inventory
indexer
is
also
switchable
module.
A
So
potentially
we
can
run
the
system
without
inventory
index
here.
So
this
is
pretty
good
from
the
modularity
point
of
view.
As
well
as
it's
pretty
good
from
from
that,
we
don't
have
a
lot
of
dependencies
and
our
system
is
like
could
be
built
on
top
of
basic
layer.
It
could
be
extensible
and
it
could
be
some
additional
complexity
could
be
introduced
along
the
way
that
actually
was
the
main
decision
and
let's
improve
the
code
quality.
A
So
this
is
the
first
update.
The
second
one
would
be
presented
by
Roman
Google
score
from
ethics,
so
he
will
show
the
introduction
of
the
source
link
as
the
part
of
the
API.
So
probably
the
name
of
the
story
sounds
little
bit
technical,
but
I.
Believe
Rahman
will
describe
you
in
more
detail.
The
business
idea
behind
the
technical
story,
so
Ramon
feel
free
to
share
your
screen.
B
They
are
stored
on
their
level
and
we
decided
to
store
in
the
stock
source
links
information
regarding
the
validity
of
exactly
the
links
that
link
foreign
source
stockin
source.
So
yeah.
That's
why
we
need
we
need
to
introduce
a
new
entity
so
yeah,
that's
interface,
related
to
this
entity
data
model.
We
introduced
it
in
in
the
inventory,
API
module
and
also
introduced
a
couple
more
interface
related
to
managing
this.
B
This
entity,
which
which
are
remove
interface,
say
interface
and
get
these
interfaces
as
so
yeah,
and
we
also
reflected
all
code
that
related
to
using
the
legacy
legacy,
implementation
of
links
and
also
we
already
factored
legacy
code
in
newly
introduced
inventory,
indexer
module
so
yeah
and
the
next
steps
here
is
to
introduce
actually
the
priority
and
purity
field
to
this
entity
and
remove
it
from
source
a
source
entity.
And
then
we
will
be
able
to
unlock
a
couple
of
other
tasks
and
most
of
them
related
to
priority
source
selection,
algorithm.
A
We
need
to
have
that
priority
and
the
level
of
the
source
to
stock
link.
That's
actually
why
we
decided
to
introduce
this
API,
which
Roman
implemented,
and
we
will
use
a
priority.
So
the
next
task
would
be
introducing
the
priority
attribute
to
this
link,
so
that
will
help
us
to
implement
the
priority
based
algorithm.
Another
business
story
which
unlocked
you
go.
C
C
C
A
Its
priority,
as
I
mentioned,
we
have
the
next
task
introducing
priority.
So
the
priority
is
not
ok,
we
will
add
the
priority
so
and
we
will
add
this
like
a
typical
attribute,
so
we
would
not
add
it.
As
extension
attribute,
we
decided
to
use
this
like
a
normal
attribute
here,
so
we
belong
that
like
it
should
attribute
or
again
it
won't
be
here,
but
priority
would
matter
well.
A
A
We
will
use
priority
for
another
business
story,
so
we
have
business
case
that,
for
example,
for
the
RMA,
a
very
strong.
We
also
need
to
make
a
decision
to
which
source
we
will
return,
so
we
will
have
will
have
the
actually
will
make
a
return
to
that
the
most
to
the
stores
with
the
biggest
priority
as
well.
So
why.
A
Part
will
have
interface.
We
will
introduce
the
one
to
that
interface.
We
share
the
name
of
the
interface
from
free
get
get
like
gets
source
for
the
return
based
on
the
stock
ID
provided
and
based
on
that
we
will
return
the
source
ID
in
our
default
implementation,
because
it's,
the
MVP
we
will
return
I,
will
make
a
decision
based
on
the
biggest
highest
priority.
Okay,.
C
C
D
C
A
Lee
we
had
the
design,
it
was
documented.
Like
quite
quite
a
long
ago.
I
mean
it's
still,
actually,
no
actually
so
the
main
idea.
So
initially
we
want
to
just
want
to
say
they
regard
ins
a
priority,
so
the
initial
idea
was
to
implement
it
at
the
extension
attribute,
but
taking
into
account
the
performance
aspect
and
also
because
we
will
use
the
priority,
at
least
in
the
MVP
part.
Quite
quite
a
lot.
We
decided
to
edit
as
a
as
a
default
default
attribute.
C
C
D
D
A
B
A
D
E
D
I
presented
them,
Carlos,
don't
know
the
source
library
and
we
have
further
web
studies
from
different
sources.
So
we
have
our
stock
configuration
that
is
higher
for
some
Model
T
that
then
their
product
that
will
show
this
product.
This
will
sell
the
previous
week.
It
was
initial,
like
implementation
of
this
fictional
I'll
teach,
and
this
week
we
already
have
working
filters,
so
we
came
through
and
also
we
have
to
the
meeting
to
filter
by
source
code
using
the
drop-down
represented
all
other
sources
in
the
system.
At
the
moment
also,
we
have
finished
the
export
functionality.
D
E
A
A
A
That
sort
of
delivery
awareness
to
order
page,
so
it
was
shown
previously,
and
we
agreed
that
we
can
close
it.
So
I
will
close
this
story
right
after
this
demo
and
run
full
selection
algorithm
by
request
on
order
page.
This
story
also
has
been
presented
and
we
get
an
approval
that
we
can
close
it.
So
I
will
close
both
of
the
story
right
after
this
demo
and
we'll
update
the
amount
of
studies
and
make.
A
A
C
A
Actually,
it's
pretty
typical
stuff
like
that.
We
can
update
in
there
several
threats
some
entity
and,
for
example,
like
to
Edmonds
work
in
parallel
in
parallel,
can
update
the
same
entity
and
while
admin
one
is
still
working
on
like
an
entity,
form
admin
to
can
already
make
it
persistence
and
when
the
admin
one
makes
a
decision
to
to
save
the
form
as
well.
Actually
he
will
just
overwrite
already
existent
data,
and
we
don't
have
like,
in
a
general
case
of
general
solution
for
this
case,
but
actually,
in
our
case
this.
A
This
could
be
a
real
big
issue,
because
currently
we
are
talking
about
the
product
update
form
and
we
because
we
are
supporting
coarse-grain
api's
and
we
save
all
the
product
dates.
And
currently,
this
inventory
data
is
also
part
of
the
product
data,
but
it
shouldn't
be
probably
in
the
ideal
world
but
like,
for
example,
if
the
admin
making
an
update
of
the
product
form
and
the
order
is
placed
like
simultaneously
visit
and
even
if
the
admin
that
updates
a
stock
item
quantity
because
he
wanted
just
to
make
update
of
the
media
data
for
the
product.
A
He
will
override
existing
data
for
for
source
items
and
actually
the
main
idea
and
I
hope
that
would
be
solved
on
the
level
of
the
UI
RM,
which
will
provide
some
kind
of
the
kind
of
MVCC
exertion
of
the
entity
and
making
making
the
persistent.
We
will
see
that
we
are
in
a
state
are
but
entity
already
in
a
state
B.
So
we
can
just
proceed
with
persistence
because
we
are
in
our
own
state.
But
actually,
ideally,
we
need
to
make.
A
We
need
to
differentiate,
inventory,
update
and
the
product
data
update,
because
currently
we
we
can
get
to
this
to
this
problematic
behavior
for
some
merchants
and
we
already
have
the
github
issue
and
actually
this
pub
issue
was
opened
and
today
we
had
a
discussion
with
the
maintainer.
But
so
what
do
you
see
about
this?.
C
A
We
need
to
align
here
on
this
or
on
track
that
or
amiss
actually
will
support
this
case
like
entity
versioning
and
another
stuff.
Would
who
do
you
think
regarding
the
separation
inventory
out
of
product
so
editing
product?
We,
so
we
illusion
here
the
coarse
grain
API
so
because
the
inventory
would
be
like
another
API
call
and
we
can't
like
now
save
the
inventory,
along
with
the
product
data.
Let's
think
it
makes
sense.
C
A
A
Actually,
the
main
point,
but
actually
in
this
case
we
will
need
to
reconsider
our
look
and
feel
because
currently
it's
on
the
same
page
on
the
inventory.
Currently,
so
probably,
we
need
to
introduce
a
dedicated
start
with
the
inventory
data
and
it
should
not
be
on
the
product
product
edit.
That's
important
topic
and
I
think
it's
being
raised.