Paragraphs migration example?

I think you must use both target_id and target_revision_id like here:

  field_paragraphs:
    plugin: sub_process
    source: paragraphs
    process:
      target_id:
        -
          plugin: migration_lookup
          migration:
            - paragraph_box
            - paragraph_grid
            - paragraph_text_embedded
          source: pid
        -
          plugin: extract
          index:
            - 0
      target_revision_id:
        -
          plugin: migration_lookup
          migration:
            - paragraph_box
            - paragraph_grid
            - paragraph_text_embedded
          source: pid
        -
          plugin: extract
          index:
            - 1

Extract is used because paragraph use compound ID (array of target_id and target_revision_id).

@safetypin : have you got the solution for this ? Am also facing same issue with paragraph and content mapping.

The solution that works perfectly for me (based on answer #4, thanks @havran) in latest Drupal 8.6.2 is below.

Please note that I didn't need to migrate past revisions of paragraphs. I just wanted to migrate the last (actual) paragraph revision from the source and attach it properly to the parent node. In reality, I think it's very common use case, because one might not need to fill in the target DB with mostly obsolete past revisions. But if you'd need to migrate *all* the revisions from the source paragraphs as well, you would probably need to extend the code below and migrate also proper revisions references.

  field_paragraphs:
    plugin: sub_process
    source: field_source_paragraphs
    process:
      target_id:
        -
          plugin: migration_lookup
          migration:
            - paragraph_box
            - paragraph_grid
            - paragraph_text_embedded
          source: value
        -
          plugin: extract
          index:
            - 0
      target_revision_id:
        -
          plugin: migration_lookup
          migration:
            - paragraph_box
            - paragraph_grid
            - paragraph_text_embedded
          source: value
        -
          plugin: extract
          index:
            - 0

These items you will likely need to customize:

• field_paragraphs: name of the target paragraph field (typically attached to the parent node etc.)
• field_source_paragraphs: name of the source paragraph field (typically attached to the source node)
• migration: array of *all* paragraph migrations the source paragraph field can use. It is important to list all migrations there and migrate paragraphs with all that migrations before. If you don't do thiat and the source ID will not be found between any of source migration, you will likely end with a not very explanatory error "input should be an array", resulting from failure on the following extract plugin that expects an array on input (and instead, it gets nothing)
• source: is always a "value" as that's how the source paragraph reference field is named in Drupal 7. If "value" doesn't work for you, check the "field_data_[field_source_paragraphs]" table in source D7 databse, where [field_source_paragraphs] is the value you entered above. There is one field that holds the value of reference, in my case it was [field_source_paragraphs]_value
• index: is always 0 as we are getting a first index from the returning [value, revision_id] pair both for target value and also for a target revision.

I'm setting the same "value" and "index" values for both target_id and target_revision_id fields. That's possible, because without migrating all the past revisions, we are okay to assume that the current revision is the first one, thus with same id as target_id. This would be probably a place to extend if you would need to migrate all the revisions.

Note: This pattern works perfectly for field collections as source as well. Just enter your field collection migration instead of paragraphs into the "migration:" field and follow the guide above for other fields.

Actually, the previous example can be even simplified like this:

  field_paragraphs:
    plugin: sub_process
    source: field_source_paragraphs
    process:
      target_id:
        -
          plugin: migration_lookup
          migration:
            - paragraph_box
            - paragraph_grid
            - paragraph_text_embedded
          source: value
        -
          plugin: extract
          index:
            - 0
      target_revision_id: '@target_id'

#10 works for me with no issues.

Hey Guy, I'a facing the same problem without solution:
here is the structure of my json file:

{
    "data": [
        {
            "uuid": {
                "value": "0142f404-0289-42ec-b26b-ab1ce7798fd5",
                "parent": false
            },
            "name": {
                "value": "P1",
                "parent": false
            },
            "supplierName": {
                "value": "U1",
                "parent": false
            },
            "description": {
                "value": "",
                "parent": true
            },
            "mandatoryInformation": {
                "value": "",
                "parent": true
            },
            "technicalInformation": {
                "value": "",
                "parent": false
            },
            "packageInformation": {
                "value": "",
                "parent": false
            },
            "productSpecifications": {
                "data": [
                    {
                        "name": {
                            "value": "p11",
                            "parent": false
                        },
                        "npp": {
                            "value": "1385002",
                            "parent": true
                        },
                        "amount": {
                            "value": "28",
                            "parent": false
                        },
                        "packagingUnit": {
                            "value": "lo",
                            "parent": false
                        },
                        "packagingSize": {
                            "value": "NN",
                            "parent": false
                        },
                        "atcCode": {
                            "value": "0o",
                            "parent": false
                        },
                        "labelling": {
                            "value": "",
                            "parent": false
                        },
                        "dosageForm": {
                            "value": "Tabletten",
                            "parent": false
                        },
                        "activeSubstances": {
                            "value": "Pioglitazon hydrochlorid",
                            "parent": false
                        },
                        "activeSubstancesWithAmounts": {
                            "value": "plo",
                            "parent": false
                        },
                        "indications": {
                            "data": [
                                {
                                    "code": {
                                        "value": "00.00",
                                        "parent": true
                                    },
                                    "codeDescription": {
                                        "value": "leo",
                                        "parent": true
                                    }
                                }
                            ]
                        },

the firts level i.e. uuid, name, descript... represents the field taht will be migrated to a entity:node with machine name 'produkt'

the content for my paragrap is the array 'productSpecifications'

Here is the yml file configuration for migrating the paragraphs:

id: produkte_paragraph
label: Migrate Paragraph content from json file into Drupal 8
migration_group: produkt

source:
  plugin: migrate_source_url
  track_changes: true
  data_fetcher_plugin: file
  data_parser_plugin: json
  urls: data/json/data.json

  item_selector: /data
  fields:

    -
      name: name
      label: 'Produkt name'
      selector: productSpecifications/data/name/value

    - name: npp
      label: 'Produkt Npp'
      selector: productSpecifications/data/npp/value

  ids:
    npp:
      type: string

process:

  field_produkt_name: name
  field_produkt_npp: npp



  uid:
    plugin: default_value
    default_value: 1

  type:
    plugin: default_value
    default_value: productspecifications(the paragrap_type)




destination:
  plugin: 'entity_reference_revisions:paragraph'
  default_bundle: productspecifications


migration_dependencies:
  required: {}
  optional: {}

Here is the yml configuration for my node (the node migration always works):

id: migrate_json_file
label: Migrate content from json file
migration_group: produkt

source:
  plugin: migrate_source_url
  track_changes: true
  data_fetcher_plugin: file
  data_parser_plugin: json
  urls: data/json/data.json


  item_selector: /data
  fields:
  - name: uuid
    label: 'uuid'
    selector: uuid/value

  - name: name
    label: 'Produkt Name'
    selector: name/value

  - name: supplierName
    label: 'Anbietername'
    selector: supplierName/value

  - name: description
    label: 'Beschreibung'
    selector: description/value

  - name: mandatoryInformation
    label: 'Pflichtangaben'
    selector: mandatoryInformation/value

  - name: technicalInformation
    label: 'Fachinformationen'
    selector: technicalInformation/value

  - name: packageInformation
    label: 'Gebrauchsinformationen'
    selector: packageInformation/value



  ids:
    uuid:
      type: string

destination:
  plugin: entity:node

process:

  langcode:
    plugin: default_value
    default_value: "und"

  type:
    plugin: default_value
    default_value: produkt

  title: name
  field_uuid: uuid
  field_suppliername: supplierName
  field_description:  description
  field_mandatory_info: mandatoryInformation
  field_technical_info: technicalInformation
  field_package_info: packageInformation
  field_productspecifications:(the paragraph field)
    plugin: iterator
    source: productSpecifications/data
    process:
      temporary_ids:
        plugin: migration_lookup
        migration: takeda_produkte_paragraph
        no_stub: true
        source: npp
      target_id:
        plugin: extract
        source: '@temporary_ids'
        index:
          - 0
      target_revision_id:
        plugin: extract
        source: '@temporary_ids'
        index:
          - 1



  uid:
    plugin: default_value
    default_value: 1

migration_dependencies:
  required: {}
  optional:
    - produkte_paragraph

dependencies:
  enforced:
    module:
      - migration_json_file

I'm really hopeless with this, I'm working on that since one week without sucess, The migrate on import one paragraph.

Don't know what to do.

Help needed!

I can't get #10 to work. It would be useful if the whole yml is posted so I can see the source and destination. It would also be useful to see an example of the yml for one of the previous migrations e.g paragraph_box.

The nearest I can get is #7 but I was hoping to not manually copy the mappings.

Migrate Slideshow Drupal 6 to Drupal 8.

Field by field mapping in migrate_plus.migration..yml file

field_slide_caption: field_slide_caption
field_slide_desc: field_slide_desc
field_slideshow_image: field_slideshow_image

Create custom_migrate.module file and Implement HOOK_migrate_prepare_row().

  use Drupal\paragraphs\Entity\Paragraph;

/**
  * Implements hook_migrate_prepare_row().
  */

 function custom_migrate_migrate_prepare_row(Row $row, MigrateSourceInterface $source, MigrationInterface $migration) {
 
//SlideShow
//GET data form drupal 6 database.
   $query_paragraphs_fid = $source->getDatabase()->select('content_field_slideshow_image', 'fsi');
   $query_paragraphs_fid->fields('fsi', array('nid','field_slideshow_image_fid','delta'));
   $query_paragraphs_fid->orderBy('fsi.field_slideshow_image_fid', 'ASC');
   $query_paragraphs_fid->condition('fsi.nid', $row->getSourceProperty('nid'));
   $mt_results_paragraphs_fid = $query_paragraphs_fid->execute()->fetchAll();
   foreach ($mt_results_paragraphs_fid as $key => $value) {
    if ($value->field_slideshow_image_fid) {
    //Caption
    $query_caption = $source->getDatabase()->select('content_field_slide_caption', 'c');
    $query_caption->fields('c', array('field_slide_caption_value'));
    $query_caption->condition('c.nid', $row->getSourceProperty('nid'));
    $query_caption->condition('c.delta', $value->delta);
    $mt_results_caption = $query_caption->execute()->fetchAssoc();
    //Description
    $query_description = $source->getDatabase()->select('content_field_slide_desc', 'd');
    $query_description->fields('d', array('field_slide_desc_value'));
    $query_description->condition('d.nid', $row->getSourceProperty('nid'));
    $query_description->condition('d.delta', $value->delta);
    $mt_results_description = $query_description->execute()->fetchAssoc();
    //Credit
    $query_credit = $source->getDatabase()->select('content_field_image_credit', 'ic');
    $query_credit->fields('ic', array('field_image_credit_value'));
    $query_credit->condition('ic.nid', $row->getSourceProperty('nid'));
    $query_credit->condition('ic.delta', $value->delta);
    $mt_results_credit = $query_credit->execute()->fetchAssoc();
    //end
   $file = File::load($value->field_slideshow_image_fid);
   $ppt_values = array(
       'id' => NULL,
       'type' => 'slideshow',
       'field_slideshow_description' => array('value' => $mt_results_description['field_slide_desc_value'], 'format' => 'full_html'),
       'field_image_credit' => $mt_results_credit['field_image_credit_value'],
       'field_slideshow_caption' => $mt_results_caption['field_slide_caption_value'],
       'field_image' => $file,
       );
    // Set Data in Drupal 8 database.   
    $ppt_paragraph = Paragraph::create($ppt_values);
    $ppt_paragraph->save();
    $target_id_dest = $ppt_paragraph->Id();
    $target_revision_id_dest = $ppt_paragraph->getRevisionId();
    $paragraphs[] = array('target_id' => $target_id_dest, 'target_revision_id' => $target_revision_id_dest);
   }
   }
  $row->setSourceProperty('field_image_slideshow', $paragraphs);
  //Slideshow End
 }

Here is my implementation for migrating from D7 field collections to D8 paragraphs.

Migrating data from D7 field collection to D8 paragraph.

migrate_plus.migration.he_fc_education.yml

migration_group: he
id: he_fc_education
label: 'Field Collection (Education)'
source:
  plugin: d7_field_collection_item
  field_name: field_education
process:
  #id: skip, newest only
  #revision_id: skip, newest only
  #uid: n/a
  #created: n/a
  status:
    -
      plugin: static_map
      source: archived
      map:
        1: 0
        0: 1
    -
      plugin: skip_on_empty
      method: row
  field_education_degree:
    plugin: he_clean_string
    source: 'field_education_degree/0/value'
  field_education_institution:
    plugin: he_clean_string
    source: 'field_education_institution/0/value'
destination:
  plugin: 'entity_reference_revisions:paragraph'
  default_bundle: education

Migrating field references from D7 field collection to D8 paragraph.

migrate_plus.migration.he_node_person.yml

migration_group: he
id: he_node_person
label: 'Nodes (Person)'
source:
  plugin: d7_node
  node_type: person
process:
  uid: node_uid
  status: status
  created: created
  changed: changed
  title: title
  field_education:
    plugin: sub_process
    source: field_education
    process:
      target_id:
        -
          plugin: migration_lookup
          migration: he_fc_education
          source: value
          no_stub: TRUE
        -
          plugin: extract
          index:
            - 0
      target_revision_id: '@target_id'
destination:
  plugin: 'entity:node'
  default_bundle: he_person
migration_dependencies:
  required:
    - he_user
    - he_fc_education

Note, migration uses revision data, so if you have issues, make sure the data exists in the revisions tables. I had a D7 module installed to delete old revision data to help keep the database small. That is what caused my field collections to paragraphs migration to not work properly. It would migrate without data. To fix it, I simply truncated the field collection revision tables and did an INSERT of the corresponding data tables into them. Make backups first so you don't screw anything up. Another option is to make a custom source plugin that doesn't use the revision data.

Although there are some hopeful comments is this thread I can't seem to find out how to migrate my content types with paragraphs from Drupal 7 to Drupal 8. Comment #9 and #10 look promising. But they are only a small part of the full solution.

If i'm right they show a part of the code for migrating the paragraph field and not the paragraph entities. Can someone help me out with a working example or good documentation on this paragraph migrating? Thanks in advance.

@zebda, yes the #9 and #10 (#10 is recommended as it's simpler) just show migration of the content nodes that have paragraphs attached. The items like paragraph_box, paragraph_grid, paragraph_text_embedded are paragraph entities, and you need to create a migration for each of these. That's an another topic, however a simpler one as these migrations are quite straightforward. An example of that migration (D7 paragraph to D8 paragrah) that worked for me was like following:

id: paragraph_box
label: Example paragraph box migration from D7 to D8
migration_tags:
  - 'Drupal 7'

migration_group: mygroup
source:
  plugin: d7_paragraphs_item
  bundle: pg_box # change to your D7 paragraph bundle here
process:
  # example fields here, adapt to your needs
  field_head: field_head
  field_introduction: field_intro
  field_text/format:
    plugin: default_value
    default_value: full_html
  field_text/value: field_text/0/value
destination:
  plugin: 'entity_reference_revisions:paragraph'
  default_bundle: pg_box_d8 # change to your D8 paragraph bundle here

Common case however is that in D7 you do not paragraphs as a source. For instance, a popular pattern in D7 was field collections instead of paragraphs. In that case, you need to change source part a bit. For field collections the source would look like following:

source:
  plugin: d7_field_collection_item
  field_name: field_myfield # enter field name of source field collection field in your D7 here

For the ones who were missing the full node migration in #10, I'm adding the full yml that worked for me:

id: my_node
label: Example node migration with paragraph field field_paragraphs inside
migration_tags:
  - Drupal 7
deriver: Drupal\node\Plugin\migrate\D7NodeDeriver

migration_group: mygroup
source:
  plugin: d7_node
  node_type: my_node # enter D7 source node type here
destination:
  plugin: entity:node
process:
  type:
    plugin: default_value
    default_value: my_node_d8 # enter D8 destination node type here
  title: title
  # other node fields come here
  # now copied the paragraph field from #10
  field_paragraphs:
    plugin: sub_process
    source: field_source_paragraphs
    process:
      target_id:
        -
          plugin: migration_lookup
          migration:
            - paragraph_box
            - paragraph_grid
            - paragraph_text_embedded
          source: value
        -
          plugin: extract
          index:
            - 0
      target_revision_id: '@target_id'
  # end of copy from #10
  status: status
  created: created
  changed: changed
  promote: promote
  sticky: sticky
  revision_uid: revision_uid
  revision_log: log
  revision_timestamp: timestamp
migration_dependencies:
  required:
      # these are "expected" migrations, as I wrote in example in #18. You need to adapt this to your own paragraph migrations
      - paragraph_box
      - paragraph_grid
      - paragraph_text_embedded

@mirsoft, thanks a lot that was just what I was looking for. You made my day. I have one follow up question. For some reason only half of the nodes is created the others have an error Array index missing, extraction failed.. A quick search on the internet points me in the direction a skip_on_empty plugin. To skip when a field is empty.

Is this the solution and how can I use this in the code you provided? Thanks again!

I am going to see if I can eliminate my need for this plugin (because I'd rather use out-of-the-box functionality), but this is what I use to process node and paragraph lookups the same way:

http://agileadam.com/2017/09/drupal-8-migrate-process-plugin-migration_l...

I can just use the migration_lookup_target_ids after any migration_lookup and it just works (on my sites, at least). Please test for yourself.

Note that I've had to patch migrate to handle multiple values: https://www.drupal.org/project/drupal/issues/2890844#comment-13158685.

Changing this to a task. If there is no docs. there should perhaps be one at https://www.drupal.org/docs/contributed-modules/paragraphs ?

And I think it should be in here.

@mirsoft, I followed the method described in the #19 and it was working fine in my local. Migrating the paragraphs and the nodes having paragraphs. The lookup table in my local was below:

| source_ids_hash                                                  | sourceid1 | destid1 | destid2 | source_row_status | rollback_action | last_imported | hash |
+------------------------------------------------------------------+-----------+---------+---------+-------------------+-----------------+---------------+------+
| 03c0d876fdff8735c14d4b47d74dd95c4fbe03973f7d96d3450b8ff5e55b56b5 |     73791 |      66 |      66 |           0 |               0 |             0 |      |
| 0649d449bb0c6287362f6304be143e043587b3a316a312804eda4bee94272687 |     69705 |      34 |      34 |     0 |               0 |             0 |      |
| 098d568d9eb9066bd9b696d73a150a67efa7c9be6ff2006748e1a3a8a470f1a4 |     73746 |      57 |      57 |        0 |               0 |             0 |      |

But if i do the same in my development the destid1 and destid2 are different. like below:

| source_ids_hash                                                  | sourceid1 | destid1 | destid2 | source_row_status | rollback_action | last_imported | hash |
+------------------------------------------------------------------+-----------+---------+---------+-------------------+-----------------+---------------+------+
| 03c0d876fdff8735c14d4b47d74dd95c4fbe03973f7d96d3450b8ff5e55b56b5 |     73791 |      4787 |      4566 |           0 |     0 |           0 |      |
| 0649d449bb0c6287362f6304be143e043587b3a316a312804eda4bee94272687 |     69705 |      4788 |      4567 |     0 |     0 |            0 |      |

Do you have any solution or the cause for the issue identified?

#24 error occurs when we had existing contents or manually added content in the site. This issue is solved after mapping the target_revision_id with index 1 like below

  field_paragraphs:
    plugin: sub_process
    source: field_source_paragraphs
    process:
      target_id:
        -
          plugin: migration_lookup
          migration:
            - paragraph_box
            - paragraph_grid
            - paragraph_text_embedded
          source: value
        -
          plugin: extract
          index:
            - 0
      target_revision_id:
        -
          plugin: migration_lookup
          migration:
            - paragraph_box
            - paragraph_grid
            - paragraph_text_embedded
          source: value
        -
          plugin: extract
          index:
            - 1

Hi there,

Thanks for all the great insight, everyone. The approach below worked for me.

My original approach came from this Softecu post. After replacing the plugins and changing the custom machine names, I was able to migrate the paragraphs but not wire up the paragraphs to the nodes.

https://softescu.com/blog/how-migrate-paragraphs-drupal-7-drupal-8

function sstone_migrate_migrate_d7_node_gta_slide_prepare_row(Row $row, MigrateSourceInterface $source, MigrationInterface $migration) { 
	$values = $row->getSourceProperty('field_branch_points'); 
	$value_new = []; 
	if ($values) {
		foreach ($values as $value) { 

        //$value_new[] = ['item_id' => $value['value']]; 

			$value_new[] = [
				'item_id' => $value['value'],
			];
			echo 'pushing ' . $value['value'] . '/' . $value['revision_id'];
		}
	} 
	echo 'count: ' . count($value_new);
	$row->setSourceProperty('field_branch_points', $value_new); 

	return $row;
}

And my YAML file looks like this:

langcode: en
status: true  
dependencies: {  }  
id: d7_node_gta_slide
class: null  
field_plugin_method: null  
cck_plugin_method: null  
migration_tags:  
 - 'Drupal 7'  
 - Content  
migration_group: sstone_general 
label: 'Nodes (GTA Slide)'  
source:  
 plugin: d7_node  
 node_type: gta_slide 
process:  
 # debug logic
 dump_sourcevar:
   plugin: callback
   callable: var_dump
   source: sourcevar
 nid: nid  
 vid: vid  
 langcode:  
   plugin: default_value  
   source: language  
   default_value: und  
 title: title  
 uid: node_uid  
 status: status  
 created: created  
 changed: changed  
 promote: promote  
 sticky: sticky  
 revision_uid: revision_uid  
 revision_log: log  
 revision_timestamp: timestamp  
 field_branch_points:  
   -  
     plugin: skip_on_empty  
     method: process  
     source: field_branch_points
   -  
     plugin: migration_lookup  
     # custom paragraph name
     migration: d7_paragraph_branch
     no_stub: true  
   -  
     plugin: sub_process
     process:  
       target_id: '0'  
       target_revision_id: '1'  
destination:  
 plugin: 'entity:node'  
 default_bundle: gta_slide
migration_dependencies:  
 required: {  }  
 optional: {  }

Best,
Nevin

This is I think the clearest article on the topic of migrating paragraphs to Drupal 8: https://understanddrupal.com/articles/introduction-paragraphs-migrations...

It has helped us enormously.

(Check the whole series at https://understanddrupal.com/31-days-of-migrations.)

Hi, I am having 3 paragraphs contents which I need to combine to one paragraph field.

  para_image:
    plugin: migration_lookup
    migration: d7_paragraph_gallery
    no_stub: true
    source: nid
  para_text:
    plugin: migration_lookup
    no_stub: true
    migration: d7_paragraph
    source: nid
  para_split:
    plugin: split_into_paragraphs
    source: body_value

  field_paragraphs:
    plugin: sub_process
    source:
      - '@para_image'
      - '@para_text'
      - '@para_split'
    process:
      target_id: '0'
      target_revision_id: '1'

when I add @para_split the migration does not work anymore, when only adding @para_image and @para_text it works. My question is on how to add the split_into_paragraphs to it so it gets imported… when running only the following it works:

field_paragraphs:
        plugin: split_into_paragraphs
        source: body_value 

I found how its working, I had to use merge
https://git.drupalcode.org/project/migrate_plus/-/blob/8.x-5.x/src/Plugi...

  para_image:
    plugin: migration_lookup
    migration: d7_paragraph_gallery
    no_stub: true
    source: nid
  para_text:
    plugin: migration_lookup
    no_stub: true
    migration: d7_paragraph
    source: nid

  temp_images:
    plugin: sub_process
    source:
      - '@para_image'
      - '@para_text'
    process:
      target_id: '0'
      target_revision_id: '1'

  temp_body:
    plugin: split_into_paragraphs
    source: body_value

  field_paragraphs:
    plugin: merge
    source:
      - '@temp_images'
      - '@temp_body'

I couldn't find an example that worked for my situation (multiple paragraph fields per content type, multiple values per field, and multiple paragraph types to choose from within each field). I was able to write a custom module that handled it. After I finished, I abstracted the module out to make it a bit easier to understand and created its own repository for it.

The repositories and an explanation video are available here: https://www.arsnova.cc/drupal-7-drupal-8-paragraph-migration

As someone suggested using the same value for target_id and target_revision_id I'd like to point out I had weird outcomes. Data was migrated but paragraphs would load incorrectly (a different paragraph showing up). This is hard to replicate because it depends on the data previously stored. In the end I used https://www.drupal.org/project/migrate_process_entity to pull the current revision_id after doing the lookup for the paragraph. This seems to have worked well, the field process section became:

  field_contacts:
    plugin: sub_process
    source: field_contacts
    process:
      target_id:
        plugin: migration_lookup
        migration: contact_section
        source: target_id
        no_stub: true
      target_revision_id:
      - plugin: entity_value
        source: '@target_id'
        entity_type: paragraph
        field: revision_id
      - plugin: extract
        index:
          - 0
          - value

Thank you for #25, which finally helped me with my field_collection migration -- all the older examples didn't help at all. I'm posting this in the hopes it will save someone else a bit of time.

  field_my_paragraph:
    -
      plugin: sub_process
      source: field_my_field_collection
      process:
        target_id:
          -
            plugin: migration_lookup
            method: process
            migration: my_field_collection_migration
            source: value
          -
            plugin: extract
            index:
              - 0
        target_revision_id:
          -
            plugin: migration_lookup
            method: process
            migration: my_field_collection_migration
            source: value
          -
            plugin: extract
            index:
              - 1

++ on #25 thank you Santhosh and cmeh.

As an alternative to Paragraphs, in case a more basic solution might work, migrating nested multivalue data into Custom Field is easy. A structure like this:

"hyperlink":
  [
    {
      "linkurl_source": "https://drupal.org",
      "linktext_source": "Drupal"
    },
    {
      "linkurl_source": "https://gitlab.com/",
      "linktext_source": "GitLab"
    }
  ]

... can be mapped with this:

fields:
    -
      name: hyperlink
      selector: hyperlink
[...]
process:
  field_hyperlink:
    plugin: sub_process
    source: hyperlink
    process:
      linkurl: linkurl_source
      linktext: linktext_source

From Migrate multivalue fields into unlimited Custom Fields field.

I am working on a Drupal 10 to Drupal 10 migration. All the examples here did not help. Looks like they are related to Drupal 7.

My solution was an own migrate process plugin. You can find it here: https://git.drupalcode.org/project/sites_migrator/-/blob/1.0.x/src/Plugi...

The plugin must be implemented into the process like this:

  field_paragraphs:
    plugin: paragraphs_sub_process
    source: field_paragraphs
    migration:
      - wg_paragraph_media_text
    process:
      target_id: '0'
      target_revision_id: '1'

I hope this example helps somebody.

#35 works fine, but made a slight adjustment for multiple migrate sources #3531281: ParagraphsSubProcess plugin does not merge paragraph items when using += operator