►
From YouTube: Backdrop Weekly - May 20th, 2021
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
Hello,
it's
may
20th
2021,
and
this
is
the
weekly
backdrop
developer
meeting
where
we
talk
about
the
latest
tickets
and
priorities
for
backdrop,
development
we
like
to
do
little
introductions
before
the
start
of
these
meetings.
I'll
go
first.
My
name
is
nate
lampton,
I'm
a
quick
sketch
on
the
internet
from
oakland,
california,
and
today
yeah.
It
looks
like
we
have
a
pretty
short
little
agenda
things
to
cover.
Let's
go
to
robert
lang.
D
I
guess
hey
I'm
greg,
usually
in
australia,
but
now
in
greece,
just
joining
for
the
regular
updates
with
the
updates
and
help
wherever
I
can
I'll
pass
it
on
to
joseph.
D
F
A
All
right,
we
don't
have
any
updates
today.
As
far
as
for
referral
websites
like
the
forum
or
localization
site,
or
even
even
the
main
site
itself,
it
looks
like
we
mostly
are
just
gonna
go
straight
into
product
update.
A
So
the
big
news,
of
course,
is
that
on
saturday
this
past
week,
we
release
119,
so
that
new
version
of
backdrop
is
now
out
and
in
the
wild
very
exciting,
and
is
there
a
blog
post
out
about
the
119
release?
No,
not.
A
Looks
like
marketing
we're
working
on
catching
up
with
that,
but
that
was
very
exciting.
We
had
a
kind
of
a
mad
rush
there
at
the
very
end,
not
getting
in
new
features,
but
compiling
a
list
of
last
bug
fixes
and
patches.
In
particular,
we
had
a
batch
of
triple
seven
cross
ports,
one
issue
that
we
worked
through
about
like
16
or
so
crossboards
that
pulled
into
backdrop.
A
Nothing
really
substantial
there
in
that
in
that
cross-porting
list,
though
just
like
small
things
here
and
there
gregory
is
there
anything
of
note
that
you
can
recall
that.
D
Not
much
small
changes,
as
you
said,
just
to
bring
myself
some
priority
in
court
priority
with
b7
as
much
as
possible.
A
Yeah,
so
yeah
119
is
now
out
the
gate,
so
that
makes
118.
We
also
made
a
new
release
of
118
118.4.
That's
the
final
release
of
118.
Unless
there's
security
releases
that
might
be
coming
out
down
the
road,
we
continue
to
main
118
maintain
118
until
two
releases.
So
when
120
comes
out,
then
118
will
get
support
like
fully
stopped
getting
updates.
A
So
119
is
now
the
next
bug
fix
release.
119.1
will
be
the
first
set
of
bug,
fixes
that
we
will
we'll
be
working
on.
A
One
big
issue
that
has
cropped
up
is
an
issue
with
backup
and
migrate
module
and
the
php
8
support
that
we
added
into
119..
This
is
a
problem
both
with
drupal
7
and
with
backdrop.
It
appears
that,
because
we
took,
for
the
most
part,
the
same
approach
as
dribble
7
did
regarding
php
8
support,
and
there
was
one
really
big
problem
with
the
database
abstraction
layer
that
the
database
abstraction
layer
previously
had
been
extending
the
pdo
classes
from
php
itself,
and
that
is
not
allowed
anymore.
So
the
workaround
of
kind
of
internalizing,
the.
F
A
Object
inside
of
the
select
queries
has
caused
a
has
caused
a
compatibility
problem
with
backup
and
migrate.
Joseph
did
I
make
a
technical
error
there.
E
A
Perfect
yeah
thanks
for
clarifying
that,
so
we
have
an
issue
5094
that
is
tracking
this
sort
of
problem.
The
backup
and
migrate
maintainer
for
backdrop
has
been
working
with
the
backup
and
migrate
maintainer
for
drupal
7,
because
they
both
have
the
same
problem.
It's
unclear
quite
what
the
solution
is.
At
this
point,
it's
likely
that
there'll
be
some
combination
of
like
fix
and
backup
and
migrate,
plus
a
fix
in
drupal
core.
A
It
seems
like
initially
like
there
was
some
larger
problem
with
backup
and
migrate
that
they've
already
made
the
maintainers
already
made
a
new
version
of
backup
and
migrate
for
backdrop,
but
my
understanding
is
that
there's
still
a
bigger
issue
that
may
require
some
core
changes.
A
I
think
that
that's
the
state
of
things,
because
we
posted
an
announcement
saying
that
a
new
version
of
backup
and
migrate
was
out,
and
that
is
recommended
to
upgrade
it
at
the
same
time,
but
there
still
seems
to
be
an
issue
that
requires
some
collaboration
between
the
core
team
and
and
the
maintainer
for
backup
and
migrate.
A
So
it's
still
kind
of
an
evolving
situation.
We're
not
quite
sure
what
the
solution
is
there.
A
Yet,
let's
see,
does
anyone
have
anything
to
add
on
that
issue,
any
additional
information
that
you
guys
have
no,
okay,.
E
Sorry
in
one
of
the
issues,
someone
recommended
adding
a
implementing
the
the
underscore
underscore
call
magic
method
and
redirecting
that
to
the
pd
redirecting
anything
that
comes
into
there
to
the
pdo
object.
E
A
Generally,
no,
I
think-
and
I
I
yeah
I
don't
know
about
official
official
policy
honestly,
I
only.
F
A
Personal
preference,
which
is
magic
methods,
are
extremely
difficult
to
track
how
they're
working
and
what
is
getting
called.
So
I
prefer
not
to
use
them
and
ides
can't
track
stuff
that
goes
through
magic
methods.
So
I
I
have
a
strong
preference
for
avoiding
them,
because
they're
hard
to
document
discover
and
to
debug
like
there's,
almost
nothing
good
about
magic
methods.
In
my
opinion,.
D
D
A
Yeah
and
and
yeah
I
I
would
avoid
that.
I
I
think
that
joseph,
if
it's
a
matter
of
making
it
so
that
the
child
or
like
the
pdo
methods,
are
accessible
at
another
layer,
then
maybe
the
pdo
object
can
just
be
made
available
as
a
property
or
through
a
method
or
something
like
that,
and
then
you
could
call
the
call
the
function
that
you
need
directly
on
the
the
pdo
object
from
outside
of
the
query
object.
You
know.
A
E
Right
well,
the
issue
is
that
the
pdo's
prepare
method
is
no
longer
available
at
the
top
level,
so
we
can't.
E
So
I
get
I
guess,
yeah
making
it
available.
Making
it
a
public
property
instead
of
a
protected
property,
would
probably
fix
that.
A
Or
or
maybe
we
can
leave
it
protected
or
private
and
put
like
a
get
pdo
or
something
like
that
method
on
there
to
like
return
the
object,
okay,
but
I
think
yeah,
let's
communicate
with
the
backup
and
migrate
maintainer.
You
know
we'll
keep
tabs
on
what
drupal
7
is
doing,
because
I'm
sure
that
they're
facing
the
same
problem
and
try
to
come
to
a
solution
that
everybody
agrees
on.
A
It
would
be
nice.
You
know,
because
we
are
constantly
trying
to
track
the
changes
that
are
happening
in
drupal
7.,
like
eventually
we'll
probably
get
an
issue.
That's
like
you
know,
port
the
fix
from
drupal
7
core.
So
so
really
would
be
nice
if
we
landed
on
the
same
solution
there
rather
than
needing
to
reconcile
those
two
things
later.
A
Okay,
let's
see
speaking
of
php
eight
support,
our
our
php
8
support
isn't
100
complete.
A
We
have
issue
5076
to
track
further
changes
that
are
needed
for
full
php
aid
support
this
one's
only
split
out
because
the
the
initial
issue
that
we
had
for
php
support
was
getting
rather
long
and
unwieldly,
and
so
this
new
issue
is
intended
to
be
a
meta,
and
then
we
make
new
issues
for
the
individual
fixes
that
we
need
there.
A
So
the
the
obvious
thing
for
us
to
start
from
is
that
there's
still
still
a
number
of
triple
seven
patches
that
we
can
pull
from,
but
even
with
the
full
set
of
drupal
7
patches,
the
full
test
suite
doesn't
run
against
php
8.
At
this
point,
so
the
most
comprehensive
way
to
get
these
problems
fixed
is
to
run
the
test.
Suite
with
php
8
environment
using
lando
or
ddev
is
a
really
good
and
pretty
easy
way
to
accomplish
that.
A
A
Let's
see
and
one
last
thing
on,
crossports,
gregory
and
indigozella
have
been
working
back
and
forth
on
getting
up
to
date
with
drupal
core
crossports
gregory
opened
a
new
issue
for
crossports
from
triple
7.70
to
7.75
at
issue,
50,
91
and
cataloged,
all
of
the
patches
that
were
in
those
six
releases
into
one
big
meta
issue
and
is
working
through
pulling
all
of
those
those
items
in.
So
thank
you
for
the
work
there.
A
No
pull
requests
yet
on
those
issues,
but
you're
sorting
out
which
issues
actually
still
apply
to
backdrop
and
which
ones
don't.
D
A
Okay,
let's
see
that's
kind
of
the
state
of
thanks
for
119.1.
A
We
have
those
that
really
big
high
profile
issue,
about
trying
to
fix
compatibility
with
backup
and
migrate,
peripheral
to
the
php
8
support
and
then
continuing
to
expand
to
php
eight
support
so
that
every
part
of
backdrop
works
with
the
latest
version
of
php.
A
Let's
see
an
issue
that
cropped
up
during
the
code,
sprints
that
we
were
having
on
119
right
at
the
end
of
a
release
like
we
were
have
like.
We
have
been
these
past
couple
of
weeks.
A
There's
always
this
like
surge
in
activity
where
we
get
a
lot
more
collaborators
trying
to
get
in
the
things
at
the
the
very
last
moment,
and
whenever
that
happens,
it
makes
apparent
some
of
the
deficiencies
in
our
tooling,
because
a
lot
of
people
are
getting
frustrated
with
running
the
tests,
automated
tests
currently
being
provided
by
zen
ci
that
the
test
results,
like
the
pages,
wouldn't
be
accessible
or
they
would
go
missing
or
we
would
get
random
failures.
A
Some
of
those
are
related
to
our
tooling,
like
on
zen
ci,
and
some
of
them
are
issues
with
core,
but
it
caused
a
bit
of
frustration.
So
a
new
issue
is
opened
in
the
let's
see
where
it
was
there's
a
a
community
group
in,
in
the
backdrop
hang
on.
Let
me
try
to
find
out
where
this
issue
is.
A
Find
an
alternative
to
frequently
problematic
zen
ci.
It
is
under
yeah
backdrop,
slash
backdrop
community
issue
number
one.
A
So
I
think
this
is
a
new
new
group
for
discussion
about
around.
You
know
community
issues.
A
This
is
kind
of
interesting
because
this
is
almost
like
a
core
issue,
but
I
don't
know
anyways
the
the
discussion
started
there.
Alan
melz
opened
it
herbduel,
bw,
panda,
greg
joseph
and
some
others
have
been
communicating
back
and
forth
about
various
options
there.
A
The
most
obvious
candidate
is
that
we
would
use
github
actions
purely
because
it's
integrated
into
github.
It's
free
there's
no
limit
on
it
for
an
open
source
repository
such
as
hours
and
yeah.
It's
it's
already
there
so
and
we
are
planning
on
using
github
actions
for
code
linting
for
checking
like
php,
syntax
and
code
style
and
things
of
that
nature.
So
we
had
been
trying
to
see
if
it
would
also
work
as
a
tool
for
running
our
automated
tests
and
so
to
that
end,
there's
issue
4753.
A
That
is
just
like
if
we
were
to
use
github
actions.
What
does
that
look
like?
And
I
I
did
some
benchmarking
on
how
fast
github
actions
can
run
and
I
didn't
put
in
a
whole
lot
of
effort
into
making
this
comprehensive,
like
running
different
versions
of
php
in
different
versions
of
my
sequel
and
stuff
like
that.
But
the
proof
of
concept
gave
us
some
idea
of
the
performance
that
we
can
get
run,
github
actions
and
so
far
it's
a
bit
disappointing.
G
A
The
issue
number
it's
issue
4753
and
yeah
joseph
you.
Having
done
a
fair
amount
of
optimization
on
the
test
runs,
it
would
be
really
interesting
to
get
your
insight
there.
A
There
were
some
suggestions
that
I
recall
that
you've
said
that,
like
using
my
isam
is
something
that
can
really
speed
up
the
test
runs,
and
it
was
suggested
that
there's
a
new
version
of
my
ism
called
area-
that's
supposedly
supposed
to
be
like
the
the
new
version
that
is
supported
in
marietv
and
newer
versions
of
my
sequel
and
it
using
arya
did
help
a
little
bit.
A
But
then
there
was
also
some
suggestions
that
we
put
the
mysql
data
directory
directly
on
a
memory
partition
and
when
we
do
that,
that
speeds
up
test
runs
a
bit
as
well,
but
it
also
negates
the
benefit
of
using
the
aria
database.
Engine
and
nodb
ends
up
being
a
little
bit
faster.
Actually,
when
using
a
memory,
partition.
E
B
E
So
I
mentioned
this
a
couple
weeks
ago
in
a
meeting,
but
I
was
saying
that
on
certain
versions
it
doesn't
actually
convert
the
tables
to
my
eyesim
and
I
finally
figured
out
what
was
going
on
and
it's
essentially,
if
you're,
using.
A
Yeah,
I
think
I
ran
into
that
problem,
that
it
wasn't
converting
them
because,
and
it
would,
it
would
like
fatal
out
when
it.
F
A
Successful
in
converting
them,
so
I'm
pretty
sure
that
the
new
approach
actually
did
end
up
working,
but
but
I
was
but
I
ran
into
the
problem
when
trying
to
convert
to
my
I
sam
because
of
that
key
length
problem,
but
when
converting
to
aria.
Instead,
it
didn't
have
that
problem,
so
it
was
converting
them
into
aria
successfully,
but
I'm
not
actually
sure
if
arya
is
as
fast
as
what
my
isam
was
previously.
It
definitely
did
improve
the
test
run
times
overall,
using
aria
versus
nodb
straight
up.
A
I
like
them
like
25,
or
something
like
that
or
maybe
20,
so
it
was.
It
was
significant
enough
to
see
that
there
was
actually
something
different
happening,
which
was
good
okay,
but
then,
once
I
used
a
memory
partition
for
nodb,
the
benefit
was
was
gone,
and
I
saw
that
we
actually
have
some
code
comments
in.
F
A
Test
running
that
switching
to
nodb
really
only
gives
you
the
benefit
if
you're
not
running
on
a
memory
partition.
So
I
thought
it
was
funny.
We
actually
had
a
code
comment
saying
if
you
run
on
a
memory
partition,
this
probably
won't
help
and
sure
enough
that
that
kind
of
turned
out
to
be
the
case,
but
the
the
takeaway
with
all
of
this,
though,
is
that
github's
runners
are
a
lot
slower.
A
Unfortunately,
so
far
with
the
optimizations,
the
best
I
could
get
was
about
half
an
hour
to
run
the
tests
on
php
7
and
that's
versus.
Like
five
minutes
on
zen
ci,
I
mean
a
huge
difference,
six
times
slower
on
github
actions
than
it
is
on
zen
ci,
and
I
don't
know
quite
why
that
is
the
zen
like
the
the
number
of
cores
for
a
github
actions.
A
So
yeah
I'm
not
quite
sure
where
to
go
from
here
with
github
actions.
Honestly,
I
I
think
that
six
times
slower
runners
might
be
a
tough
pill
for
people
to
swallow.
So
I'm
I
don't
think
that
I
can
really
push
forward
saying
we
should
use
it
anyway,
when
it's
that
much
slower
another
option
would
be
using
external
runners,
like
github
actions,
that
you
have
hosted
runners.
A
A
A
Okay,
all
right.
Let's
see
that's
all
we
have
for
119.1.
As
far
as
priority
bug
fixes,
there
are
still
a
huge
number
of
issues
and
the
next
milestone
looks
like
76,
open
issues
for
119.1.
So
there's
still
lots
of
known
issues
to
prioritize.
A
If
you're
looking
for
something
actionable
to
to
move
on
right
now,
1
1920
one
1920
will
be
the
next
minor
version
of
backdrop.
A
We
have
the
summer
to
work
on
it
yay
kind
of
exciting,
so
it
will
be
coming
out
september,
15th,
four
months
from
now,
more
or
less,
and
we
kind
of
have
yet
to
revisit
like
what
are
the
features
that
we're
going
to
target
for
1.20.
A
The
most
obvious
things
are
the
ones
that
carried
over
from
119.
That
includes
an
exposed
file,
name
filter
for
the
media
library,
dialog
issue.
3293.,
let's
see
the
field
has
visibility.
Field
has
value
visibility,
conditions
for
layouts
issue,
4728
the
saved
draft
module
in
core
issue.
3157
actually
looks
like
I
thought.
Bw
panda
was
the
advocate
for
this
one.
But
gregory
did
you
just
add
this
to
the
agenda.
D
Yeah
at
some
point
so
peter
joins
the
company
that
I
work
for
so
he's
overwhelmed,
so
he
removed
himself
from
being
an
advocate
and
I
sort
of
like
decided
to
take
over,
because
I
want
to
see
that
getting.
But
if
peter
changes
his
mind
he's
more
than
welcome,
I
have
a
long
list
of
things
that
I
want
to
work
on,
so
there's
always
exciting
things:
okay,.
A
Let's
see
things
that
also
carried
the
advanced
caching
hashtags
and
cash
max
age
issue.
4127.
A
joseph
is
the
advocate
for
that
one
that
one
we
didn't
have
enough
reviews
and
feedback
to
provide
like
feedback
in
time
for
it
to
be
iterated
upon,
I'm
not
sure
joseph.
If
there's
anything
you'd
like
to
say
about
that
issue
today,.
G
I
don't
know
okay,
I
don't
like
that
issue.
That's
all
I
have
to
say
about
that.
A
Okay,
okay
and
then.
Lastly,
this
issue
is
not
one
that
carried
the
only
issue
that
we
have
prioritized
for.
1.20
is
a
new
issue
that
bug
folder,
that
robert
filed
issue,
50
67,
an
issue
with
node,
previews
and
layout
module
have
have
their
code
kind
of
all,
matched
together
right
now,
an
issue
50
67
is
an
issue
to
kind
of
create
a
new
api
to
decouple
those
two
things.
Do
you
have
any
updates,
robert
on
the
state
of
that
issue?.
B
It
has
a
pr
I've
tested
it,
it
works,
it
just
needs
more
eyes
on
it
to
make
sure
I
didn't
miss
anything
dumb.
Okay,.
A
Great,
that's
that
then
it's
yet
to
be
decided
about
kind
of
when
we'll
do
our
next
planning
meeting.
A
If
history
is
any
precedent,
it
would
actually
be
june
3rd,
but
we
probably
need
to
put
some
effort
around
promoting
and
discussing
you
know
like
hey
what
items
should
we
prioritize
for
1.20,
get
some
people's
thoughts
on
it
and
then
get
that
meeting
like
scheduled,
so
that
people
are
aware
that
it's
happening?
So
we
can
have
some
participation
and
discussion
in
that
that
get
together.
A
A
So
that's
exciting.
That
kind
of
brings
ready
to
wear
the
ready
to
wear
initiative.
To
a
close,
we
also
have
the
telemetry
initiative
that
saint
paul
tim
is
is
heading
up.
I
don't
think
that
we
have
any
updates
for
this
week,
so
I
won't
say
more
about
it
at
this
time.
A
That's
kind
of
it
for
the
things
that
we
have
on
the
agenda.
Do
you
guys
have
anything
you
guys
would
like
to
bring
up
and
discuss
today.
D
Well,
since
I
was
working
on
it
recently,
another
initiative,
the
automated
automatic
update.
D
Was
long
being
forgotten,
nothing,
I
think
it's
waiting
on
unite,
if
I
recall
correctly
just.
A
A
Yeah,
that
would
be
a
great
item
to
to
pick
back
up
again
the
code
signing
last.
We
left
that.
I
think
that
I
did
some
work
to
convert
the
code
signing
code
that
is
in.
A
A
I
think
there
were
other
things
with
automatic
updates
that
are
also
actionable.
There's
a
really
nice
drupal
7
module
for
automatic
updates.
Now
that
I
I
think
that
we
could
go
and
look
at
the
automatic
updates
module
that
the
drupal
contrib
team
has
put
together
and
see
the
way
in
which
it
works.
I
know
that
it
has
two
parts
that
we
could
benefit
from.
A
One
of
them
is
a
pre-flight
check
where
it
does
things
like
it,
checks
if
any
core
files
have
been
modified,
does
like
a
diff
type
functionality
as
well
as
other
things
to
like
make
sure
that
running
an
automatic
update
is
safe
as
best
as
you
can
know,
if
it's
safe
to
run
an
automatic
update,
so
those
preflight
checks
would
be
very
valuable
for
us
and
secondly,
I
haven't
confirmed
this,
but
assuming
that
the
drupal
7
automatic
updates
work
fully
really
curious
how
update.php
fires
you
know
how
database
updates
are
fired
because
right
now,
that's
the
problematic
thing
that
we
have.
A
A
All
right:
well,
let's
call
it
a
meeting,
then
just
one
question
before.
A
Two
weeks
after
the
release
yeah,
so
it
would
be
yeah
next
weekend-
would
be
the
the
likely
time
for
it
to
happen
some
somewhere
around
there.
A
Right,
that's
what
what
I
was
going
to
say
as
well
is
that
if
we
can
get
that
backup
and
migrate
issue
to
a
happy
resolution
as
of
right
now,
that's
the
only
thing
that
we
know
of
that
is
a
serious
regression.
A
And
so,
if
we
get
that
problem
fixed
to
everyone's
satisfaction,
we
may
do
a
release
immediately.
You
know
sooner
than
that
I
mean,
but
I
I
kind
of
doubt
that
it'll
be
faster
than
than
that
normal
two-week
timeline.
So
and
if
we
don't
fix
that
backup
and
migrate
issue,
there
might
not
be
any
reason
for
a
119.1
release
if
there's
not
other
substantial
fixes
in
there.
So
then,
what
the
two
weeks
after
the
119
release
is
usually
like,
if
necessary,
so
yeah,
it's
it's.
It's
all
a
little
bit
fuzzy.
A
Yeah,
but
I
definitely
would
say
that
should
be
our
number
one
thing
is:
is
trying
to
fix
that
regression
with
backup
and
migrate
and
take
a
look
at
what
triple
seven
is
doing
the
core
and
backup
and
migrate
sides,
as
well
as
working
with
our
maintainer
quackers
ate
me,
which
is
a
hilarious
username.
Who
is
actively
looking
for
solutions
right
now
in
the
contrib
version
of
backup
and
migrate
for
backdrop,.