Collections

Each WiseFood Data API resource is exposed on the client as a collection proxy — client.articles, client.guides, client.textbooks, client.fctables, and so on. A collection proxy behaves like a smart, lazily-loaded Python container.

All collection proxies subclass BaseCollectionProxy, so they share the access patterns below.

Length and iteration

len(client.articles)            # number of entities in the first index page

for article in client.articles: # one API fetch per entity
    print(article.title)

Note

len() and iteration are backed by an initial index page (default 100 identifiers). They reflect that page, not necessarily the entire remote collection. For large datasets, prefer search or explicit slicing with limit/offset.

Access by position

first = client.articles[0]      # entity at index 0 (from the cached index page)

Slicing → lazy proxies

A slice is translated into a limit/offset request and returns lazy proxies. No per-entity data is fetched until you read a field:

page = client.articles[0:10]    # one list request; 10 lazy proxies
page = client.articles[20:40]   # offset=20, limit=20

for a in page:
    print(a.title)              # each access fetches that single entity

Constraints (they raise rather than silently doing the wrong thing):

• Open-ended slices (client.articles[10:]) are not supported — always give a stop.

• Steps other than 1 (client.articles[0:10:2]) are not supported.

Access by identifier

Strings are resolved as a slug, a full URN, or a UUID, depending on the resource:

client.articles["some-article-slug"]          # slug for a URN-backed resource
client.articles["urn:article:some-slug"]       # full URN
client.artifacts["3f2c…-uuid"]                 # UUID for an ID-backed resource

If the identifier is already in the cached index it is reused; otherwise the proxy fetches it directly from the API. See Identifiers & URNs for how slugs, URNs, and UUIDs relate.

Explicit get with optional laziness

art = client.articles.get("some-article-slug")              # fetch now
art = client.articles.get("some-article-slug", lazy=True)   # defer until first field access

Tab-completion in IPython / Jupyter

Collection proxies expose slugs() and integrate with IPython’s completion protocol, so you can tab-complete known identifiers inside brackets:

client.articles.slugs()         # list of completion-friendly identifiers

# In IPython/Jupyter:
client.articles["med<TAB>"      # completes to matching article slugs

This requires IPython to be installed (see Installation).