Skip to main content

Python interface to the Bulk API.

Project description


Salesforce Bulk

Python client library for accessing the asynchronous Bulk API.


pip install bulk-support


To access the Bulk API you need to authenticate a user into Salesforce. The easiest way to do this is just to supply username, password and security_token. This library will use the simple-salesforce package to handle password based authentication.

from bulk_support import SalesforceBulk

bulk = SalesforceBulk(username=username, password=password, security_token=security_token)

Alternatively if you run have access to a session ID and instance_url you can use those directly:

from urlparse import urlparse
from bulk_support import SalesforceBulk

bulk = SalesforceBulk(sessionId=sessionId, host=urlparse(instance_url).hostname)


The basic sequence for driving the Bulk API is:

  1. Create a new job

  2. Add one or more batches to the job

  3. Close the job

  4. Wait for each batch to finish

Bulk Query

bulk.create_query_job(object_name, contentType='JSON')

Using API v45.0 or higher, you can also use the queryAll operation:

bulk.create_queryall_job(object_name, contentType='JSON')


import json
from bulk_support.util import IteratorBytesIO

job = bulk.create_query_job("Contact", contentType='JSON')
batch = bulk.query(job, "select Id,LastName from Contact")
while not bulk.is_batch_done(batch):

for result in bulk.get_all_results_for_query_batch(batch):
    result = json.load(IteratorBytesIO(result))
    for row in result:
        print row # dictionary rows

Same example but for CSV:

import unicodecsv

job = bulk.create_query_job("Contact", contentType='CSV')
batch = bulk.query(job, "select Id,LastName from Contact")
while not bulk.is_batch_done(batch):

for result in bulk.get_all_results_for_query_batch(batch):
    reader = unicodecsv.DictReader(result, encoding='utf-8')
    for row in reader:
        print(row) # dictionary rows

Note that while CSV is the default for historical reasons, JSON should be prefered since CSV has some drawbacks including its handling of NULL vs empty string.

PK Chunk Header

If you are querying a large number of records you probably want to turn on PK Chunking:

bulk.create_query_job(object_name, contentType='CSV', pk_chunking=True)

That will use the default setting for chunk size. You can use a different chunk size by providing a number of records per chunk:

bulk.create_query_job(object_name, contentType='CSV', pk_chunking=100000)

Additionally if you want to do something more sophisticated you can provide a header value:

bulk.create_query_job(object_name, contentType='CSV', pk_chunking='chunkSize=50000; startRow=00130000000xEftMGH')

Bulk Insert, Update, Delete

All Bulk upload operations work the same. You set the operation when you create the job. Then you submit one or more documents that specify records with columns to insert/update/delete. When deleting you should only submit the Id for each record.

For efficiency you should use the post_batch method to post each batch of data. (Note that a batch can have a maximum 10,000 records and be 1GB in size.) You pass a generator or iterator into this function and it will stream data via POST to Salesforce. For help sending CSV formatted data you can use the salesforce_bulk.CsvDictsAdapter class. It takes an iterator returning dictionaries and returns an iterator which produces CSV data.

Full example:

from bulk_support import CsvDictsAdapter

job = bulk.create_insert_job("Account", contentType='CSV')
accounts = [dict(Name="Account%d" % idx) for idx in xrange(5)]
csv_iter = CsvDictsAdapter(iter(accounts))
batch = bulk.post_batch(job, csv_iter)
bulk.wait_for_batch(job, batch)
print("Done. Accounts uploaded.")

Concurrency mode

When creating the job, pass concurrency='Serial' or concurrency='Parallel' to set the concurrency mode for the job.

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

bulk_support-0.3.tar.gz (11.9 kB view hashes)

Uploaded source

Built Distribution

bulk_support-0.3-py2.py3-none-any.whl (10.3 kB view hashes)

Uploaded py2 py3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page