Import job definition

You can import user profiles from the ReachFive Console. This makes using existing user profiles that you may already have easier as well as enabling you to batch update users via a JSON or CSV file.

This page shows you all you need to know regarding formatting for importing profiles as well as the validation process, import and merge logic, and import limitations in the following sections:

Import format
Import and merge logic
Import limitations
Force data update from file
Import Lite profiles only
Job reports

For instructions on how to create and run an import job from the ReachFive Console, check out the Import profiles page.

Sequential imports

Imports are sequential. This means that an import file can create and update the same profile during an import.

For example: if user A with the name Marie is created early in the import and later on in the same import file, user A has the name updated to Maria. When the import is done, user A keeps Maria as her name.

Import format

Here, we look at how you need to format import files.

JSON
CSV
File validation

JSON

The schema for the imported user is based off of the user profile object. In this format, all fields are optional.

{
  "external_id": "1",
  "email": "foo@gmail.com"
} \n (1)
{
  "email": "bar@gmail.com",
  "name": "Joe",
  "gender": "M",
  "custom_fields": { (2)
      "has_loyalty_card": true
  },
  "consents": { (3)
    "exampleConsent": {
      "date": "2021-11-23T11:42:40.858Z",
      "consent_version": {
        "language": "fr",
        "version_id": 1
      },
      "granted": true,
      "consent_type": "opt-in",
      "reporter": "managed"
    }
  },
  "addresses": [ (4)
    {
      "id": 0,
      "default": true,
      "address_type": "billing",
      "street_address": "10 rue Chaptal",
      "address_complement": "4 étage",
      "locality": "Paris",
      "postal_code": "75009",
      "region": "Île-de-France",
      "country": "France",
      "recipient": "Matthieu Winoc",
      "phone_number": "0723538943",
      "custom_fields": {
        "custom_field_example": "custom_field_example_value",
        "custom_field_example_2": 42,
      }
    }
  ]
}

1	Each profile must separated by a line feed character (`\n`).
2	Import `custom_fields` as an object containing a series of fields.
3	Import `consents` as an object or as a flattened field with the format `consents.<consent>.<parameter>` shown below. `... "consents.cgu.date": "2021-09-03T19:08:01Z", "consents.cgu.granted": true, "consents.cgu.consent_version.version_id": 2, "consents.cgu.consent_version.language": "fr", "consents.cgu.consent_type": "opt-in", "consents.cgu.reporter": "managed" ...`
4	You can import addresses including custom address fields.

CSV

The CSV format uses a free format where columns are flattened JSON paths.

For example a path { "foo" : { "bar" : "test" } }` becomes foo.bar in the CSV header.

If a column has no value, then the corresponding field will not be used for the current line.

email,name,gender,custom_fields.has_loyalty_card,consents.newsletter.consent_type,consents.newsletter.granted,consents.newsletter.date,consents.newsletter.reporter,addresses.0.custom_fields.custom_field_example
bar@gmail.com,Joe,M,true,opt-in,true,2018-05-25T15:41:09.671Z,managed,custom_field_example_value

File validation

Once parsed, every imported profile is validated to ensure that only consistent data is imported.

Profile import files must comply with the following requirements:

At least one unique field should be provided (several can be provided):
- email (verified or unverified),
- phone_number (if the SMS feature is available, verified or unverified),
- a valid provider
- a valid external_id
At most a single matching profile should be returned for merging. The matching is made on:
- uid (corresponding to the profile id),
- email (verified or unverified),
- phone_number (if the SMS feature is available, verified or unverified),
- provider
Only one other profile can match the imported profile. If more than one user is matched, we can’t automatically define how to merge them, so an error will be returned.
Any provided consent should be defined for an existing key with a date anterior to the current date.
Any provided custom field should be defined for an existing key.
Any provided password hash should specify a valid hash method.

Import and merge logic

This section details various import and merge logic details for import jobs.

Subsections:

Multi-threaded imports
Delete existing fields
Delete an address
Create a user via import file
Encrypt import file
Social Login
Merge

Multi-threaded imports

Our import jobs are multi-threaded. This means that many profiles are imported simultaneously. Therefore, it’s important that you ensure all profiles are unique; otherwise, you may get duplicate profiles.

We strongly encourage you to check the following fields to ensure each profile is unique:

email
external_id
custom_identifier
phone_number (if SMS feature is enabled)

See the User Profile for more on these fields.

Delete existing fields

If you want to delete an existing field, you need to pass a null value as part of the import file.

The updated_at value of the import file must be more recent than the existing profile.

JSON
CSV

{
  "updated_at": "2021-06-04T14:16:34.658Z",
  "email": "bar@gmail.com",
  "name": "Joe",
  "family_name": null, (1)
  "gender": "M"
}

1	Pass `null` to delete the existing `family_name` field.

external_id,updated_at,email,name,family_name,gender
1,2021-06-04T14:16:34.658Z,foo@gmail.com,,,,,,bar@gmail.com,Joe,__null__,M (1)

1	Pass `__null__` to delete the existing `family_name` field.

Delete an address

If you want to delete an existing address, you need to pass the to_delete boolean as part of the address object in your import file.

JSON
CSV

{
  "email": "test@example.com",
  "addresses": [
    {
        "id": 1,
        "to_delete": true (1)
    }
  ]
}

1	Pass `to_delete` as `true` along with the desired address `id` to delete the address.

email;addresses.0.id;addresses.0.to_delete (1)
test@example.com;1;true (1)

1	Pass `to_delete` as `true` along with the desired address `id` to delete the address.

Create a user via import file

You can create a user via an import file. When you create a user this way, the created_at and updated_at fields are automatically set to the profile creation date.

If the created_at and/or updated_at fields are already present in the file for the particular user being created, the dates in the file are used.

Import A = You create a user and the file has no created_at or updated_at fields. The profile creation date is used for the fields.

Import B = You create a user and the file does contain the created_at or updated_at fields. The dates currently in the file are used for the fields.

Encrypt import file

When importing files with ReachFive, it is inherently secure and encrypted. However, you may wish to provide additional encryption for the import process.

To encrypt your desired import file, run the following command from the root of the file location:

openssl aes-256-cbc -salt -pbkdf2 -iter 10000 -in file_to_import -out file_to_import.enc

Command section Notes

Command section	Notes
`-salt`	Salts the password.
`-pbkdf2`	Declares use of PBKDF2 function.
`-iter`	Specifies the number of iterations of PBKDF2. Example: `-iter 10000`
`-in file_to_import`	The name of the original file to import. Example: `-in file_to_import`
`-out file_to_import.enc`	The name of the output file (encrypted). Example: `-out file_to_import.enc`

-salt

Salts the password.

-pbkdf2

Declares use of PBKDF2 function.

-iter

Specifies the number of iterations of PBKDF2.

Example: -iter 10000

-in file_to_import

The name of the original file to import.

Example: -in file_to_import

-out file_to_import.enc

The name of the output file (encrypted).

Example: -out file_to_import.enc

You can import multiple social login profiles using the identities field. The provider name must be a valid provider name, and the user_id field should be the ID of the user from the provider’s site.

The user’s profile will be populated with provider information after the first login of the user using the corresponding social provider.

Merge

If a corresponding profile is found during a merge, the profile will not be created but merged with the existing profile. The matching is based on the following fields:

id (corresponding to the profile id),
email (verified or unverified),
phone_number (if the SMS feature is available, verified or unverified),
provider

A profile should never match more than one other profile, or an error will be returned, since there is no way to know the merge priority in a 3+ way merge.

The merge is a safe merge, meaning that (following the priority) no field is overwritten. Fields can either be completed when they are lists or free objects (consents, custom_fields), created if empty, or completely ignored if already defined.

Merge priority

The priority of the merge is defined by the attribute updated_at. If the attribute is not present on both profiles, then the existing profile has priority over the imported profile.

Consents are the exception. If a consent is present in both merged profiles, the unified profile takes the consent with the most recent value based on the consent’s date attribute.

Import limitations

Notification limitations

The Import Module does not trigger or send any notifications via an import. It only brings in the necessary information as defined in the CSV or JSON file as it relates to user profiles and consents.

Any status that would trigger a notification typically within the ReachFive architecture and flow would need a follow-up action after the import itself.

Examples

Email confirmation of email address
SMS confirmation of phone number
Double opt-in consent confirmation (SMS or email)
Webhook and Pub/Sub hook triggers

Provider limitations

There is no way to define to which provider an email or phone_number belongs to. Keep in mind that after the first login using a social provider, the profile’s information will be updated using the provider’s own information, completing the missing fields of the profile.

Force data update from file

The Force data update from file option is available on the ReachFive Console when importing profiles. This option does not use the typical merge logic for user profiles but instead forces the data you import to update all existing user profile fields.

The Force data update from file option does not apply to the consent date. This will always reflect the date at which the consent decision occurred.

There may be unwanted behaviour if you use this option so be sure you understand your requirements before choosing this approach. See the import and merge logic section for more details on standard merge logic.

How does it work?

If there are fields with values in the current ReachFive profile that are not in the import, these fields retain their current value. However, all fields with values replace the existing fields in ReachFive.

Import profiles in Testing mode

When you import profiles in Testing mode, ReachFive tests:

access to the server
access to the import file
whether the the import file content is valid

Testing mode does not handle database modifications.

Import Lite profiles only

import lite profiles

When you import Lite profiles only, you are not importing the fully managed profiles, but rather those lite profiles that registered with your site. You have to import Lite profiles separately from managed profiles.

Separate import jobs

Whether you are importing new profiles or updating existing ones, managed profiles and lite profiles should be imported separately as different jobs. If you try to import a mixed group of managed and lite profiles, it could cause production issues.

Created and updated fields

If created_at is present in the import file, keep this value in the file.
If created_at is not present in the import file, value it with the execution date of the import.
If updated_at is present in the import file, keep this value in the file with maximum tolerance of + 10 minutes compared to execution date. If the date exceeds tolerance, bring it back to execution date + 10 minutes.
If updated_at is not present in the import file, value it with the execution date of the import.

Job reports

Job reports provide a snapshot of the different export and import jobs that you have run. Information such as the id, a definition, the job type, the progress of the job, the job’s status, and when the job was last run is provided for you so you can get a quick look at the latest jobs. To find more information about the job, you can view the job logs.

Only job reports run in the last 6 months are displayed on the Job reports page in the ReachFive Console. Older jobs are automatically deleted.

To access job reports:

Open your ReachFive Console.
Go to Settings Job reports.
Report statuses
The following statuses are possible for job reports:
- FAILURE which indicates the job has failed. Check out the logs for more information.
- WAITING which indicates the job has yet to finish.
- SUCCESS which indicates the job was successfully run.

Filter job reports

You can filter job reports from the ReachFive Console. Filtering the reports helps you more easily find the job report you’re looking for. The job report filters offers several ways to filter as shown in the table below. This includes setting a timeframe and how the results are displayed.

To filter job reports:

Open your ReachFive Console.
Go to Settings Job reports.
Click Filters in the top right corner.

Enter your desired filters in the dialog.

View from the console

Filter Description

Filter	Description
`Job ID`	Filter by a specific job ID that has been run. This filter is particularly useful when you need to look into a specific job. Perhaps, you obtained the ID through logs or need to investigate further.
`Job definition ID`	Filter by a job definition ID. This filter is useful when you want to look at a particular job definition. Maybe, you want to see how often it’s been run, or why a specific job definition is succeeding or failing.
`Job type`	Filter by the type of job such as export or import. The job type filter is good to filter out only the type of jobs you want to see.
`Job status`	Filter by the status of the job. Statuses `SUCCESS` `RUNNING` `WAITING` `WAITING_CANCELLATION` `CANCELED` `FAILURE`

Job ID

Filter by a specific job ID that has been run.

This filter is particularly useful when you need to look into a specific job. Perhaps, you obtained the ID through logs or need to investigate further.

Job definition ID

Filter by a job definition ID.

This filter is useful when you want to look at a particular job definition. Maybe, you want to see how often it’s been run, or why a specific job definition is succeeding or failing.

Job type

Filter by the type of job such as export or import.

The job type filter is good to filter out only the type of jobs you want to see.

Job status

Filter by the status of the job.

Statuses

SUCCESS
RUNNING
WAITING
WAITING_CANCELLATION
CANCELED
FAILURE

If desired, select a timeframe by clicking From for the starting date and To for the end date.
Choose the Order of how you want your results displayed.
Click Apply.

Job logs

To view more information on the job itself or download the logs, click Show logs found on the far right hand side of the table.

In the job logs, you can see the information Level, the Content of the information, and the Date on which it occurred.

Category Description Example

Category	Description	Example
Level	Denotes the level of the log information. possible values `ERROR` indicates an error has occurred while running the job, typically related to incorrect credentials or corrupted files (CSV/JSON). `WARNING` indicates there is something incorrect about the log, typically related to syntax or duplication. `LOG` indicates standard log content, typically denoting profiles, or the operation itself (export/import).	ERROR
Content	Gives a description of the log information.	"Export profiles from beginning"
Date	The timestamp for the log information.	`2023-04-01 19:00:06`

Level

Denotes the level of the log information.

possible values

ERROR indicates an error has occurred while running the job, typically related to incorrect credentials or corrupted files (CSV/JSON).
WARNING indicates there is something incorrect about the log, typically related to syntax or duplication.
LOG indicates standard log content, typically denoting profiles, or the operation itself (export/import).

ERROR

Content

Gives a description of the log information.

"Export profiles from beginning"

Date

The timestamp for the log information.

2023-04-01 19:00:06

Download logs

You can download full logs or just the errors by clicking the desired dropdown and choosing the format for the download.

Example JSON download

{"Level":"LOG","Content":"Export profiles from beginning","Date":"2023-04-02T00:00:00.899Z"}
{"Level":"LOG","Content":"Total profiles to export: 47","Date":"2023-04-02T00:00:01.296Z"}
{"Level":"LOG","Content":"Uploading to path: /uploads/weekly-export.csv","Date":"2023-04-02T00:00:01.298Z"}
{"Level":"ERROR","Content":"Error occurred: Invalid SFTP credentials.","Date":"2023-04-02T00:00:06.991Z"} (1)

1	Shows where the `ERROR` occurred and gives details on what the error was.

Export logs

You can export logs for a specific job using the Management API’s Export job logs endpoint.