By category - Collections

This group of tasks covers two data structures - Dictionary and List - along with the tasks that let a Business Process create, read, update, and inspect them while it runs.

• A Dictionary is a key/value store. Each entry has a unique String key mapped to a value of any type. Use it when you need to look something up by name or code - vendor IDs, currency mappings, UOM conversions, error messages keyed by field name.
• A List is an ordered, indexed collection. Use it when order matters or when you're accumulating items as you go - a batch of records to process, a running set of validation errors, a queue of output documents.

The tasks include:

General reasons to use Collections tasks

1. In-memory lookups instead of repeated queries. Rather than querying a database or re-scanning a source file every time you need a reference value, you load it once into a Dictionary at the start of the BP and do fast key-based lookups (Has Key, Get) for the rest of the run. This is the single most common use case in integration BPs — mapping trading partner codes, account numbers, or product IDs to internal equivalents.
2. Accumulating results across a loop. Lists are the natural place to collect things a BP encounters while iterating — validation errors, matched records, generated output filenames — so you can act on the full set after the loop finishes (e.g. checking List - Size to decide whether to send an exception alert).
3. Dynamic data whose shape isn't known up front. Unlike fixed BP variables, Dictionaries and Lists can grow or shrink at runtime, which is essential when a BP is processing a variable number of trading partner documents, line items, or API response records.
4. Passing structured data between BP steps or subprocesses. A Dictionary is a convenient way to bundle several related values (e.g. a parsed header record) into one object that can be passed around instead of tracking many individual variables.

Practical tips

• Put overwrites silently if the key already exists — useful for upserts, but worth guarding with Has Key if you want to detect duplicates rather than clobber them.
• Always initialize with Create before the first Put/Add — these tasks add to or read from an existing collection, they don't implicitly create one.
• Get/First/Last on an empty or missing entry will typically cause a runtime error, so pair them with Size or Has Key checks in branching logic rather than assuming the item exists.
• Dictionary keys are always Strings — if your lookup key is numeric (e.g. an account number), remember it'll be compared as a string, so formatting/leading zeros matter.
• These tasks are almost always used together with looping tasks and decision/branch tasks elsewhere in the BP — Collections tasks themselves don't iterate or branch, they just store and retrieve.

Use Case

See this Collection BP Tasks Use Case for an example of the Collection tasks in use. 

Tasks