Drupal 8 Migrate: Taming the wild west and learning new tricks

everybody this is talk about Drupal 8

migrate and how amazing it is originally

I was going to do this in kind of a Wild

West theme but doing that across like 40

slides or so it just seemed kind of dumb

so I'm leaving with it the Wild West

part out of it so it was me I've been a

software developer really since I was a

little kid I've been in the Drupal

community for about five years I just

kind of weaseled my way into a project I

had no business being on order to get

off of it another project and then kind

of fell in love with the community and

my form and all of you guys this is what

is what's kept me passionate about this

community for the past five years I'm

primarily back-end as you can see from

the design of this slide I do not have

an eye for design I mean I'll try to CSS

if I have to but if I could do

everything from the command line that is

ideal contributed to some important

modules I guess as many as I'd like but

you know some Grand Master and I just

love to eat my little fake Oh avatar

everywhere all right so what we're going

to talk about is how migrate works some

of the kind of must-have modules you

need to really be productive in it and

then how to write each of the different

types of plugins to handle the different

operations that migrate is going to do

we're going to talk a little bit about

destinations because there there is a

little bit special and then we're gonna

have live demos with maybe real coding

you get to watch me tight as we go

so before I start who here has done any

sort of migration in Drupal 8 at all

okay it looks like about half who has

tried and failed to do a migration okay

hopefully the the beginning of this see

how everything works we'll we'll bring

you up to speed and then the the real

hardcore people maybe at least hopefully

there's something you'll get out of this

but we said the real the real hardcore

Cody meat is all at the end what we're

not going to talk about is migrate

upgrade or migrating any legacy sites

with with migrate there's a whole lot of

good talks on that stuff already this is

outside the scope of what I'll be

talking about we're not really going to

be cataloging differences between all

migrated and new migrate although you

know I might just sort of rant about how

bad stuff used to be and we're not to be

talking about multi-language stuff

there's I don't know if I've seen the

talk for it but there are definitely

good posts about how to import

translations and things like that

we're not going to be dealing with that

so what is migrate how does it work

basically there's three parts to a

migration there's a source process and a

destination source this is where the

data comes from this is any data source

pretty much any data source you can

imagine there's already plug-ins for

Jason and CSV other Drupal all kinds of

web server anything you can imagine

there's probably already a source for it

process is what Maps that source data

into data that the destination is going

to import and the destination is where

that data is going to go most of the

time that's a Drupal entity not all the

time I'm going to cover something a

little bit later where we have a

destination that's not Drupal and

sometimes you want to export data from

your Drupal site out to something else

but for the most part most of what I see

most of people's you know issues that

I'm trying to correct or how quick it's

all transforming external data into

Drupal entities so let's start with a

source plugin this thing love to read

this is where your data comes from this

is how we're defining it every source

plug-in has a different set of

properties we're using my grade source

directory for this one that this is this

is one of my favorite ones it's it's

very illustrative of how the source

plug-in worse

and I'll be diving into this a little

bit later so you can see just how easy

it is to make a custom source plug-in we

can see that we're gonna be looking at

this particular folder so a lot of

questions that that I've seen evolve

like how do I check data you know my

source data into the repository how do I

point my module at it regularly you know

that's how so you just start at the your

web root and you can specify the path

all the way into your modules data file

we're only looking at PDFs and we have

an unlimited recursion level migrate

directory migrate source directory

actually allows you to limit how many

steps down you're gonna go we don't

we're only dealing with one directory

but you can limit that now we're gonna

look at the process this is what takes

that source data here we have a file

name and the URL

and we're going to map this into

destination data and then this is where

we plug in any of our tools that can

manipulate that data this is a simple

migration that scans a particular

directory that one for PDFs and then

using the parsed PDF process plug-in

that I wrote just based on a generic

kind of PHP PHP DF parser brings in that

data into our body field and then the

destination determines where this is

going to go and in this case it's going

to go into a node of the type document

we can see that we have no dependencies

in here but the enforced module

dependencies this is this is a pretty

key thing this is a good thing that that

migrated plus gives you that regular

migrate doesn't migrate plus will

enforce that enforced module dependency

so when you've disabled your module it

knows to remove all that config regular

role and migrate doesn't do that so when

you you know when you have a module with

your config done under migrating center

migrate plus when you uninstall you have

to delete all the config for each of

your migrations this handles it all for

you I know that is all really exciting

so how do you start well you already

have it it's in core but your real

must-haves to be super productive with

it you need migrate plus you need

migrate tools with the migrate that

comes in core there's really no facility

to just use it out-of-the-box it's code

only there's no command-line tools

there's no UI if you're gonna write a

migrate and then use it using only core

files you know only the core module

you're gonna have to write the code to

do the import

and then there's some add-on tools here

migrate devel is pretty wonderful that

it does a some some great things it

integrates nicely with drush and will

actually display you a snapshot of what

your source data looks like and then you

can make it put what the what the

destinations gonna look like - and then

config update config update really lets

you it's just so you can make quick

changes to your migrate camel and then

just paste them in and run it and it

will update the config for you otherwise

you're going to be doing a lot of you

know a lot of uninstalling and

reinstalling your module to update the

config migrate develop if you use that

migrate debug tag that it comes with the

command line option it will update your

import config but only after it runs

which is not super helpful because you

have to run it limit it roll it back run

it again and then you'll get your real

results but it's still faster than

uninstalling and reinstalling your

module so how do we do it really we just

create a module we think it's a

directory and an info yeah no file or

just Drupal GM if you were using the

Drupal console

we're gonna start with some config nasaw

goes into config install we need a

config for our migration group the

groups are really optional but you

should really use them and then a

migration a config file for the

migration itself and then when we enable

the module when we do address migrate

status

we'll just see it in the list that looks

something like this there we go so we

can see here's the different groups that

I have here I've got a group a list of

cars that comes from a CSV

here's that PDF one I was looking at and

then the special treat is pulling Mars

rover data straight from NASA because

we're trying to prove that migrations

are rocket science but well we'll make

it that way

and then rush em IM to run it now if

you're like me and you've been doing

this for a really long time you'll keep

typing and rush mi and it will yell at

you every time because the new dress 9

is all mi m so for dress 8 1 1 5 which i

think is what i'm on it'll just put a

warning up so remember him I am migrate

the import once you've got all that down

you're gonna want to test your migration

so I've written a little helper module

that does it optionally does a process

row but mostly is here to display what

your source data and your destination

data will look like you don't need to

worry about trying to copy any of this

down that is a shortcut to the github

repo of where all of this is so let's

run one

here's our our site with our favorite

theme and then can we read this I have

this my great helpers processor oh it

takes the ID of a migration how many

rows were going to process and then

whether we're going to process it

or not so with false that means all

we're going to see is source data but I

changed this to true will also see how

it maps during the process base let's

execute this so this is a CSV file

oops

oh boy

[Music]

so it looks like that

if we come down it's got this crazy

doubleheader here so the first header

line is what we'd expect it's column

headers and then they have the types on

the next line so we want to do something

special to handle this we don't want to

import string double in double double as

a car

[Music]

so we can see that we have a skip plug

in here during in our process space it

uses a static map so that if it

encounters the string string it will

turn it into a zero which skip on empty

registers as empty and it will skip this

row so that's why I have two in here for

how many rows were processing because

the first one won't actually process a

row you'll see when I when I run this

that it doesn't have any destination

because it skips in and then this is

what the file looks like so car would

give a string mpg cylinders displacement

horsepower etc etc then if we come down

to the next row which is the first one

with actual data in it we can see that

it's giving us a Chevelle mpg of 18 yeah

right because I have a modern v8 muscley

car and I don't get 18

that's 1970 their standards are probably

different bobbly

the ringing is not me it was coming to

go on last obsession - all right well

that's good all right maybe it'll do

that when we're waiting what I'm talking

know what's going again you have to be

plugging in on the adapter for the video

I don't see a USB that doesn't only have

it doesn't have a USB adapter as well

don't see it on this one son all right

well sorry for this body I'll just try

to you know if it flashes four leaves

with more than a second or two just

giving a sign not only

I'm used to working ahead of a WebEx

anyway so this isn't all that different

from what I normally do so all we're

seeing now is source data this is just a

representation of the source file that

we pulled in and the fields that are in

it if we look at what the mapping looks

like

this is excellent there we go we're

setting our destination fields to the

source fields and this is a pretty

simple just one-to-one migration so

we're gonna put cylinders and cylinders

horsepower and horsepower and so on go

and change this to true now and see what

it is that we're doing those were the

column headings in the CSB

yes

these

that composure

good good

there it is

all right

no that's going this is excellent

all right so we have this set to true so

it will run the process row on this and

we'll be able to see how it maps we can

see that this we don't even have our

first one in here because that was the

one that we told it to skip with that

skip on empty so we get our second one

it's giving us our source data and then

we can look at our destination we can

see now this goes to title our skip has

a value in it so it doesn't it's

assigning it just to admin for the

creator and then here all of the the

field values that it's doing so with

just this little bitty animal we have

all of this in here and working so let's

go to content as we can see is empty

and then we'll run the import you can

see that we start out we have 101 rows

in this spreadsheet we have imported

none of them I remember this time than I

am now CSV import should go pretty fast

it just spins through a file that it's

already got creates all of these notes

when we refresh here there they are

I know it's very impressive personal

collection

no this isn't really my era we're gonna

dodge Coronets and things in the 70s

this isn't really my my thing it was

just the easiest free car data source

that I could find we can do address em

are then to roll them back

and then once that's done we refreshed

content and they're all gone so it

handles the mapping of the source ID and

the destination ID for you so it knows

what to roll back and then if you have

any updates and you're running with an

update flag it will automatically update

the appropriate destination node based

on changes in your source node

all right so let's talk about sources

this is where your data it comes from

now for the most part there are sources

that can interpret everything that

you're going to want whether it comes

from you know some sort of relational

database a web service and some data

file files in a directory structure or

other systems whatever I haven't really

found a place where there's anything

lacking yet so yeah I really just I

haven't had to write a lot of sources I

don't think I've had to write any to

actually use inner projects there's

something that covers what we have but

let's look at my great source directory

this is a really good pattern to to

follow as as far as a source goes

but I said not scrolling over yeah this

is great

there it is I apologize for that can we

read this good

we make a little larger

apparently no

it's suddenly decided to go into

presentation mode I hit my shortcut

before all right the source plugin

really only needs a couple of things it

needs an initialized iterator and it

needs IDs and IDs this is what is going

to identify your source data as a unique

record iterator is how it actually gets

the data from the source this is a good

pattern to follow as opposed to some of

the core documentation the core doc is

not entirely correct

for instance source plugin tells you

that for initialized iterator that you

can give it an array it is not an array

this must be an iterator if you're

trying to return an array and this

initialize iterator like it says you can

it won't even register it as a source

and it will run blank so that's a gotcha

and then all it's doing is building up

its own couple of iterators to take the

source directory and loop down through

them and then since we have a file

extension set it checks the file

extension of everything that it found

that it finds and if it matches it'll

push it to the array that it wraps in

the array iterator again I cannot stress

how key this is because when I do write

a source I have forgotten to do this a

couple of times and it's baffling as to

why it comes up blank so let's go ahead

and look at that I'm going to turn

presentation mode off so you can see

what the directory looks like

it keeps flipping back to my other

phpstorm rather than coming out of where

it's supposed to come out too

[Music]

we're still in target come on open

jeez Louise all right so here's our

module and we have a data directory in

here there's three PDFs in here so

that's what it's going to use that's

what it finds as we process it so I have

one up here for our PDF import we

execute we can see what this source

looks like and all this is is the path

and the files that exist inside here and

here's just these three of them well

there's one okay

so here's our various rows

okay if you're writing a writing a

source I really recommend that you

follow migrate source directory to see

what is looking like

so demo time that's that's what I was

doing is I was skipping ahead of myself

let's talk about processes now this is

where it's really exciting this is where

you take your source data and manipulate

it and then map it out to where it's

going to go in the destination for the

most part most of your custom work is

going to be done here most of what

you're going to write is going to be a

process plugin

so these are just things that do stuff

today though to see what you already

have using the Drupal console you can do

a debug plug-in and look at my great doc

process it'll give you a list of

everything that's in there and where it

lives so yeah my great plus again I

cannot stress how important this is it

has half of the things that you'll need

to do just in it like you know building

arrays or a pop the rate pop is one of

my favorite examples for the process

plug-in because it's so simple although

I'll get to that in a minute all the

process plug-in really needs there's a

transform method that returns out

whatever data it's going to return you

know the day the output data

coming back to the other target what I'm

just going to close that

so for for a simple process plug-in this

is a bad as simple as it gets

we need to identify it so you identify

as a process plug-in give it an ID if

you're nice you'll explain how to use it

and then we extend our process plug-in

base which gives us this transform

method and as we can see all we're doing

is running a ray pop and returning that

value this is a fully functional really

viable process plug-in it doesn't have

to be more complicated than this but I

wrote one that's more complicated than

this just to show you guys

so I wrote this parse PDF just based on

a quick and easy PHP PDF PDF parser

library that I found so let's let's take

this apart let's see what we're doing

here it extracts text from a PDF source

here's our plugin identifier so when we

use this in a migration file we're going

to set our process plug-in as parse

underscore PDF this is how we're going

to use it here's our destination target

plug-in that we're using and then the

source which really just expects a file

a fully-qualified file name for where it

can find the file in our important

transform function we just do a little

bit of validation make sure that we have

a value make sure that it's not an array

then finally we make sure that the file

exists and if it does we open up the

parser get the text and return it so

let's hop back over here and turn this

to true so that we'll see what our

process row output looks like when we

execute this

we can see not only do we have our

source up here but in our destination

we've mapped body value to the text of

that PDF and here it is

it's a long one here we go

[Music]

so we cover area notation and our

transform method oh and value I should

go over that to value is what you're

actually giving the ammo has the source

it comes into your plug in under value

right there so we got value is we set it

to URL which is that fully qualified

file name and now let's actually run

this guy

so we concede we should we should be

empty here of PDFs imported is zero will

run our PDF migrate

and then we'll go back to content and

look there are our notes so if we had it

one we can see that we've pulled the

text out of the PDF

[Music]

let's swishing back and forth is great

isn't it there we go

you can also change these together well

an array of processes

apparently I didn't have this in

presentation mode when I was pre running

it because I had all of the tabs nicely

in a row ah streamline that don't worry

so we have our body value goes to this

plugin so if we were to write another

little plugin let's just copy this and

make this a two

and then here let's just make sure that

it's not empty that it's not an array we

don't care about this and let's make our

text

so yeah writing a new process plug-in

really is just this simple all we're

doing is doing a string to upper on the

value that comes in and returning it

we're going to slip that into our

migration here

we don't actually need a source line by

default it will use the return from

whatever is above it in the chain now

let's go ahead and roll that back

what was that oh you're right

okay now we're going to do this migrate

debug you can see that it's going to

give us what we had and then the next

time we run it the new config will have

loaded and then she need to do a my M

so this to LA I'm and specify your

actual migration this is why you all

come to a talk that says live code good

good good

I need to clear the cache because I

added a plug-in

is it in my hand but I'm Vanessa

I do not know it's a fine question

though

well I am mr. import I don't know what

else mi is supposed to be

and sometimes your migrations will get

hung up in a loop or if you work them

where they're you know if they if they

break so here are a handful of important

commands Josh migrate stop

and then drush migrate reset status this

will set it back to idle so that you can

do your upgrade again or import again

and there it is so this is what my great

debug gives you it gives you the course

this is a long PDF there we go

so right here from your dress output

it'll give you that same sort of thing

that my little helper it gave you in M

they execute window and then we can see

that our body value is all capitals now

so we just added a new process plug-in

and I gotta go in with just that the

whole thing was what 20 lines of code

and all the three of them were copy it

paste really

destinations this shows where this

determines where your data is going to

go however your your import process is

going to save data out

that's your destination plug in again

it's just my experience most of the time

you're not going to be writing these

just something like 90% of all of the

migrations that I've seen or to get

things into Drupal but to get things out

of Drupal migrate plus comes with a SQL

table output that you know handles

imports and deletes and and then rolls

them back on an individual record level

for you that's that's great

so I wrote a custom one that is going to

export data as a CSV

more just everybody's favorite the

migration itself looks like this mostly

standard stuff we're using that

[Music]

now we can see that our destination

plug-in is CSV and we're giving it a

path of just our doc route export that's

ESB

we're using a destination base that has

an import import is what it is that

actually writes that data so we're going

to get the path off of our configuration

and then if it doesn't exist we're going

to create the file and then for every

row that we have defined in our process

it's going to add it up blow it up into

a CSV and write it out and then it has

custom roll back functionality it

doesn't go in like row by row remove

records from the file although it could

be sophisticated enough to do that this

version of it just deletes the file when

it rolls it back so let's look at what

that looks like we can see that we have

no export that's CSV

now we're going to take all of that data

from one CSV and put it into another CSV

there it is

so here's all the hundred rows that we

did and note that it skipped the row we

told it to skip

if we roll it back

I'll still they're good

well ideally when you roll it back it

should delete the file too

Benetton

and this not actually storing anything

it's just converting it from one source

to the destination exactly this that

this migrate is just a conduit Drupal is

just an orchestrating everything it

doesn't say anything is an entity Drupal

the only changes in anything Drupal and

related now are the the map tables that

the migrate keeps track on and and

that's it so we have some more practice

with NASA data and this was pretty fun

to do we're going to import data about

Mars rovers and then get the pictures

that they take these cute little guys so

we have these in a group there's one

Jesus there's one that gets the actual

rover data one that gets the cameras

associated with each Rover and then one

that grabs the photos that it's taken on

that particular day it's going to import

terms and notes they're very exciting

but you're select

yes so this is just to prove that

there's no trickery going on here these

are in fact empty no cameras no Rovers

this is riveting

excellent so you could just do the full

migration on this group and it gives you

some cool stuff in fact we're gonna say

our group is NASA and now when you do it

by group it does no dependency why is

what needs to come before what and it

will run things in the appropriate order

the rover photos for instance need the

camera which needs the rover so won't do

the matter of order we can see that on

the on the date that I have this

configured with it it only had four

photos and let's just have a look at

those

now sometimes you need to clear the

cache here because it doesn't get good a

good good so on June 3rd 2015 this is

what curiosity took with the front

hazard camera and these are all nicely

tagged so we can look everything

curiosity took and here's all of our

notes so but is we're going to do

something even cooler than that remember

I mentioned all the way at the beginning

that regular migrate relies on code to

actually do the import it doesn't have

any command-line tools or UI or anything

like that well I happen to have some

code here that would do it so we have

our three Rovers

we'll leave poor spirit alone because he

stopped working a while ago

give me a give me a date something that

I don't have to click back too far

perfect now we submit this is loading up

our migration changing the source URL

property and then running the the

migrate and doing the import so we can

see that it automatically redirected us

to my little view here that shows

opportunity any camera and on the date

that we just specified and these ones do

need the cash to be clear down there

images images get a little finicky with

time we all know how this works

so well that's good they're way faster

than I expected it to you know that's

gonna need to be cleared go ahead write

it into your process function ordinarily

I would but I was hoping to be faster so

looking at our submit we're grabbing our

rover ID grabbing the the date that

we're doing this on and then for each of

the migrations that were doing for

running this get migration so you got a

plug-in manager to migration create an

instance based on your migration ID and

here you can get your executable which

gives you an object with all of your

source configuration now in my main

configuration for this you can see that

I've got

let's Rovers photos this is actually

coded in here for the third June 2015

what we're actually doing is we're

replacing this and this

with our values that we pass in our over

name in our date and then we're

returning that and then we run the

import

so we can actually control this from a

code level and run this same import on

any of these parameters that we want now

when we load we should have pictures

great

let's just drill into one of these while

it's going

so that's a good picture opportunity but

let's see everything that opportunity

got so here's some more data we got from

NASA when it landed on Mars when it

launched its current status which is

active again poor spirit doesn't say

active the Mars Sol which because Mars

doesn't exactly have a 24-hour day so

it's been there for 5018 local Mars days

the newest picture that it has is from

Wednesday and has taken almost 200,000

pictures and here they are so we can see

that some of these are you know not

super helpful but look that is real

rocks and treads on Mars

see if we have our view back yes here we

go

now that I've got all of this data in a

Drupal 8 site let's use this d8 source

plugin migrate Drupal is what does your

version diversion kind of imports and I

think for core version 8.5 this migrate

Drupal d8 is part of the core Drupal

migrated update but I'm not sure I did a

patch for a Drupal migrated d8 and then

was told that it was all rolling up so I

don't know if my documentation patch

actually went in there but if you're

looking for documentation on how it

works the Drupal or migrate Drupal da

now has a nice readme on how to set all

of this up on your own that had zero

documentation before so let's migrate

these cars again

and then while that's going we're going

to head over to our other site so here

we have the rat site it's just really

the destination for all of this it's

still my favorite theme but it's rats we

can tell it apart you can see that it

has no content empty

this one is done now let's open our

other shell this is rat to match the red

so we know as I'm switching back and

forth because I tried to switch back and

forth quickly in the video doesn't

always keep up which site we're on

mmm we can see it worked a minute ago

probably mistyped it I did I had one -

that does not do now we can see that our

Drupal 8 source site has these hundred

cars it's got the PDFs that we imported

it's got all of our over data so let's

go ahead and bring these cars over I

have the same content type

you know same field definitions and

everything on one site versus the other

they don't have to go one-to-one

I just set them up so that they were and

now we refresh on our target site BAM

all these come from the other Drupal

site right in here and they're all

complete

you can see they're anonymous I didn't

have users but that you can do one more

will do that documents too

yeah it always pauses at the fast parts

and here's our our documents that we

imported

without a body but I might not have

included the body

let's go ahead and roll back the

documents here

you can see it was three now if we

switch over here

[Music]

you can see in our source we're back

down to zero so it is getting these

things live from one site to the next

and forth so imagine there are some sort

of questions yeah

no it adds on to it I'll show you how

that looks

it makes the you have imported more than

it has as its source but its source is

hard-coded by default so here we can see

that we have imported 19 so we have

minus 15 left unprocessed it will just

keep going and if I do a rollback here

we're all although Mac will you

contribute that GUI migration along with

the

the helper that you have a link to your

slides yeah I can also I do have the

public repo and you can get it right

there

sure

so I've done a couple of these usually I

get it done and I kind of forget about

it

now I've had a couple sites where

there's content that keeps getting

created like oh I'm taking one site to

another site and there's like a few

weeks between right and stuff gets

graded I always worried about rewriting

the migration because like it first of

migrates the configurations we want to

read through that because I've changed a

lot configurations I'm going to do it

back is it safe to pick up new content

off the old site do another run of the

migration well let me let me clarify on

what you mean by the configurations

they're like so I've done about my grade

from Drupal Drupal okay it pulls the

month the configuration from the old

site it doesn't it doesn't have to okay

but what I wanted it to it was good

first time and then I've you know

changed some stuff will it go is it

possible to just to run like particular

content right yeah well if you're the so

the question was in a say a Drupal to

Drupal migrated rerunning the migrated

to grab new content you can target it to

run a particular a particular migration

so if you've got one migration for blog

posts one migration for articles you can

say only run the articles at that point

it will run differences so it will only

do the sources that it doesn't already

have unless you run it with a - - update

in which case it will pull all of the

data and and let you it'll overwrite all

of the new stuff but let's say the

antenna nodes in your original article

import and then two weeks later you've

got two more you're running at em I am

articles it will only pull in the to

change the two new ones if they've

changed on the source site unless you do

a - - update they won't come over the

old the ones through a reason I got it

right

okay thanks it's a drupal 7 if we were

pulling in content from a third party

API that's regularly updating on it's

like doing feeds so yeah feeds migrate

is migrated best solution which

programmatically calling the migrate on

are on schedule that is my preferred

solution but I'm kind of I'm anti feeds

that that's mostly the way that I do it

like with that we have a knowledge

management system that pulls in

citations and literature references and

things from Zotero and the previous

incarnation of the site did it in feeds

and it was kind of hard to manage you

couldn't really rollback or track the

changes super well so I wrote a migrate

for it and it runs on a schedule on a

crop just like that and and pulls in all

of the deltas now another thing that I

have it doing is actually checking the

source data against our version of the

target data and then flagging the

differences for updates so that when it

pulls it it knows which ones it wants to

update as well as which ones it wants to

do new

each of them less taxing so it really is

I mean we're talking yeah I mean it's

not that many records it's eight twelve

thousand something like that but over

HTTP to get that it's slow so getting it

once

checking it and then importing it was by

far the fastest way to go google

migration or any migration no sources it

doesn't if it knows if there is

additional source data because it checks

the the migrated mapping and if that

source ID doesn't exist in your map it

knows that that's a new record if

something has changed it doesn't know

that automatically you have to tell it

and so like for that's a taro example I

have a job that goes through that

migrate map table and there's a needs

update field in there and if you change

that to true it will know to pull that

in and the next the next migration

he used in a tuple seven migrate having

a high water field yeah yes I had such

bad luck with that in Drupal seven it

didn't always work correctly with

complex data that going against my own

advice using built-in tools and I for

the most part just check make the check

myself

and he he's the d8 source bug even in

court

Ortiz probably have more time for the

drupal DA again in Drupal 8.5 I think it

is part of the core migrate Drupal now

if it's not though its migrate Drupal da

vennett depends on how you need to

process it if it's like if there's a

process that's a whole you know like

really something complex it really

doesn't lend itself to just a line item

by line item I might have something that

imports the data transform you know it

does whatever he needs to do on it and

then import it but for the most part

yeah I'll just do whatever did

operations I need to do kind of field by

field run by row in my process section

yeah that more specific example looks

like you're still thinking of something

yeah

there are well the dates of a UNIX

timestamp I don't know that there is one

that simply does it off the shelf now

but there's a callback one you can just

you know call a like a time ghouls three

more times

yeah and as for a concatenated there is

a concatenate plus if I come with

regular migraines I don't remember and

it will kids put those things yet are

your slides they can be okay I'll had it

alright anything else well thank you all

very much