This project is designed to import data into the FOLIO LSP. It provides a simple and efficient way to import data from various sources using FOLIO's REST APIs.
- Import MARC records using FOLIO's Data Import system
- Import User records using FOLIO's User APIs
- Batch post Instances, Holdings, and Items to FOLIO inventory storage
Using pip
pip install folio_data_importor uv pip
uv pip install folio_data_importTo install the project from the git repo using Poetry, follow these steps:
- Clone the repository.
- Navigate to the project directory:
$ cd /path/to/folio_data_import. - Install
uvif you haven't already. - Install the project and its dependencies:
$ uv sync. - Run the application using Poetry:
$ uv run folio-data-import --help.
Make sure to activate the virtual environment created by uv before running the application.
This package provides CLI commands for importing data into FOLIO:
# Main command with subcommands
folio-data-import <subcommand> [options]
# Or use standalone commands
folio-user-import [options]
folio-marc-import [options]
folio-batch-poster [options]Tab Completion: Install shell completions for better CLI experience:
folio-data-import --install-completionAll commands support environment variables for FOLIO connection credentials, allowing you to avoid repeating these parameters:
export FOLIO_GATEWAY_URL="https://folio-snapshot-okapi.dev.folio.org"
export FOLIO_TENANT_ID="diku"
export FOLIO_USERNAME="diku_admin"
export FOLIO_PASSWORD="admin"Once set, you can omit these parameters from your commands:
# Instead of:
folio-data-import users --gateway-url "..." --tenant-id "..." --username "..." --password "..." --user-file users.jsonl
# You can simply use:
folio-data-import users --user-file users.jsonlThis works for all subcommands: users, marc, and batch-poster.
Alias: folio-user-import
Import users to FOLIO with extended functionality beyond mod-user-import.
folio-data-import users \
--gateway-url "https://folio-snapshot-okapi.dev.folio.org" \
--tenant-id diku \
--username diku_admin \
--password admin \
--user-file users.jsonlService Point Management: Specify service points using codes instead of UUIDs:
{
"username": "checkin-all",
"barcode": "1728439497039848103",
"active": true,
"type": "patron",
"patronGroup": "staff",
"departments": [],
"personal": {
"lastName": "Admin",
"firstName": "checkin-all",
"addresses": [
{
"countryId": "HU",
"addressLine1": "Andrássy Street 1.",
"addressLine2": "",
"city": "Budapest",
"region": "Pest",
"postalCode": "1061",
"addressTypeId": "Home",
"primaryAddress": true
}
],
"preferredContactTypeId": "email"
},
"requestPreference": {
"holdShelf": true,
"delivery": false,
"fulfillment": "Hold Shelf"
}
"servicePointsUser": {
"defaultServicePointId": "cd1",
"servicePointsIds": [
"cd1",
"Online",
"000",
"cd2"
]
}
}
Flexible Matching: Match users by id, externalSystemId, username, or barcode:
folio-data-import users --user-file users.jsonl --user-match-key usernamePreferred Contact Type: Accepts FOLIO IDs or human-friendly strings (mail, email, text, phone, mobile). Set a default for users without a valid value:
folio-data-import users --user-file users.jsonl --default-preferred-contact-type emailUpdate Behavior: Control how existing records are updated:
⚠️ Breaking change in v0.6.0: Prior versions incorrectly merged incoming records into existing records, preserving fields absent from the incoming data. Starting in v0.6.0, the default behavior is full replacement — fields not present in the incoming record are removed. If you depend on the previous merge behavior, use--update-only-present-fields.
-
Full replacement (default): The incoming record completely replaces the existing record. Fields present in the existing record but absent from the incoming record are removed. The preferred contact type is preserved from the existing record when not specified in the incoming record.
folio-data-import users --user-file users.jsonl
-
Partial update: Only fields present in the incoming record are updated; missing fields are preserved from the existing record.
folio-data-import users --user-file users.jsonl --update-only-present-fields
Field Protection: Protect specific fields from being updated:
-
Job-level protection (applies to all records):
folio-data-import users --user-file users.jsonl \ --fields-to-protect "personal.preferredFirstName,barcode" -
Per-record protection (using custom field
protectedFields):{ "username": "jdoe", "customFields": { "protectedFields": "barcode,personal.telephone,personal.addresses" } }
User Deletion: Delete users listed in the input file(s) instead of creating or updating them:
folio-data-import users --user-file users.jsonl --delete-all- Only users that exist in FOLIO and match via the configured match key are deleted
- Staff and system users are automatically skipped to prevent accidental removal
- Associated request preferences, permission user records, and service point assignments are also removed
- An interactive confirmation prompt is displayed before deletions begin; use
--yes/-yto skip it (required in non-interactive environments)
JSON Lines format - one user object in the style* of mod-user-import (with extended support mentioned above) per line
*also supports dereferenced (UUIDs instead of reference strings) user objects (eg. directly extracted from /users)
Alias: folio-marc-import
Import binary MARC21 records via FOLIO's Data Import system using the change-manager APIs.
folio-data-import marc \
--gateway-url "https://folio-snapshot-okapi.dev.folio.org" \
--tenant-id diku \
--username diku_admin \
--password admin \
--marc-source-path records.mrcThe command will prompt you to select a Data Import Job Profile configured in your FOLIO tenant.
- Process single files or entire directories of MARC files
- Interactive job profile selection
- Real-time progress tracking
- Automatic retry on transient errors
Note: FOLIO's import logs can be unreliable. If you don't see a job summary when your job completes, check Data Import in FOLIO (Data Import > Actions > View all logs...).
Alias: folio-batch-poster
Efficiently batch post Instances, Holdings, and Items to FOLIO's inventory storage endpoints with support for creating new records and updating existing ones.
folio-data-import batch-poster \
--gateway-url "https://folio-snapshot-okapi.dev.folio.org" \
--tenant-id diku \
--username diku_admin \
--password admin \
--object-type Items \
--file-paths items.jsonl \
--batch-size 100 \
--upsert-
Multiple File Support: Process multiple files with glob patterns
folio-data-import batch-poster --object-type Items --file-paths "items_*.jsonl" -
Upsert Mode: Create new records or update existing ones
folio-data-import batch-poster --object-type Items --file-paths items.jsonl --upsert
-
Field Preservation: Control which fields are preserved during updates
folio-data-import batch-poster --object-type Items --file-paths items.jsonl --upsert \ --preserve-statistical-codes \ --preserve-administrative-notes \ --preserve-temporary-locations \ --overwrite-item-status
-
Selective Patching: Update only specific fields
folio-data-import batch-poster --object-type Items --file-paths items.jsonl --upsert \ --patch-existing-records \ --patch-paths "barcode,status,itemLevelCallNumber" -
Failed Records: Automatically save failed records to a file
folio-data-import batch-poster --object-type Items --file-paths items.jsonl \ --failed-records-file failed_items.jsonl
-
Progress Tracking: Real-time progress bar with statistics
- Disable with
--no-progressfor CI/CD environments
- Disable with
-
Config File Support: Use a JSON config file for complex configurations
folio-data-import batch-poster config.json
Example
config.json:{ "object_type": "Items", "file_paths": ["items1.jsonl", "items2.jsonl"], "batch_size": 100, "upsert": true, "preserve_statistical_codes": true, "preserve_item_status": true, "failed_records_file": "failed_items.jsonl" }
Input files should be JSONL (JSON Lines) format - one complete JSON object per line:
{"id": "item-001", "barcode": "12345", "status": {"name": "Available"}}
{"id": "item-002", "barcode": "12346", "status": {"name": "Available"}}
{"id": "item-003", "barcode": "12347", "status": {"name": "Checked out"}}Create new items:
folio-data-import batch-poster --object-type Items --file-paths new_items.jsonlUpdate existing items (by ID):
folio-data-import batch-poster --object-type Items --file-paths items.jsonl --upsertUpdate only barcodes and call numbers:
folio-data-import batch-poster --object-type Items --file-paths items.jsonl --upsert \
--patch-existing-records \
--patch-paths "barcode,itemLevelCallNumber"Process multiple files:
folio-data-import batch-poster --object-type Holdings \
--file-paths holdings_*.jsonl \
--batch-size 500 \
--upsertRun folio-data-import batch-poster --help to see all available options:
--object-type: Type of inventory object (Items, Holdings, or Instances) - Required--file-paths: Path(s) to JSONL file(s) - supports glob patterns - Required--batch-size: Number of records per batch (1-1000, default: 100)--upsert: Enable create-or-update mode--preserve-statistical-codes: Keep existing statistical codes during updates--preserve-administrative-notes: Keep existing administrative notes--preserve-temporary-locations: Keep temporary location (Items only)--preserve-temporary-loan-types: Keep temporary loan type (Items only)--preserve-item-status: Keep item status (Items only, default: true)--patch-existing-records: Enable selective field patching--patch-paths: Comma-separated list of fields to patch--failed-records-file: Path to save failed records--no-progress: Disable progress bar (useful for CI/CD)
All CLI commands can also be used programmatically in your Python applications.
import asyncio
from folioclient import FolioClient
from folio_data_import.BatchPoster import BatchPoster
async def post_items():
# Create FOLIO client
folio = FolioClient(
okapi_url="https://folio-snapshot-okapi.dev.folio.org",
tenant_id="diku",
username="diku_admin",
password="admin"
)
# Configure batch poster
config = BatchPoster.Config(
object_type="Items",
batch_size=100,
upsert=True,
preserve_statistical_codes=True
)
# Post records
async with BatchPoster(
folio,
config,
failed_records_file="failed_items.jsonl"
) as poster:
await poster.post_records("items.jsonl")
print(f"Posted: {poster.stats.records_posted}, Failed: {poster.stats.records_failed}")
asyncio.run(post_items())import asyncio
from folioclient import FolioClient
from folio_data_import.UserImport import UserImporter
async def import_users():
folio = FolioClient(
okapi_url="https://folio-snapshot-okapi.dev.folio.org",
tenant_id="diku",
username="diku_admin",
password="admin"
)
config = UserImporter.Config(
user_file="users.jsonl",
user_match_key="username",
default_preferred_contact_type="email"
)
importer = UserImporter(folio, config)
result = await importer.do_work()
print(f"Imported {result.records_created} users")
asyncio.run(import_users())import asyncio
from folioclient import FolioClient
from folio_data_import.MARCDataImport import MARCImportJob
async def import_marc():
folio = FolioClient(
okapi_url="https://folio-snapshot-okapi.dev.folio.org",
tenant_id="diku",
username="diku_admin",
password="admin"
)
config = MARCImportJob.Config(
marc_source_path="records.mrc",
job_profile_id="profile-uuid",
job_profile_name="Bibliographic records"
)
job = MARCImportJob(folio, config)
result = await job.start_import()
print(f"Imported {result.created_records} MARC records")
asyncio.run(import_marc())For complete API documentation and advanced usage, visit our full documentation.
Contributions are welcome! If you have any ideas, suggestions, or bug reports, please open an issue or submit a pull request.
This project is licensed under the MIT License.