wait_chain: reject empty strategy list with ValueError instead of IndexError

[HrachShah]

Summary

wait_chain() accepts a variadic list of wait strategies and stores them in self.strategies. When called with no strategies, self.strategies ends up as (). The first call to __call__ then evaluates min(max(1, 1), 0) == 0, so self.strategies[wait_func_no - 1] turns into self.strategies[-1] and raises an IndexError from inside a running retry.

This is opaque for what is really a programmer error at construction time. An empty wait_chain has no useful behaviour, so failing early is friendlier than letting it crash later from a min(max(1, 1), 0) corner case.

Fix

tenacity/wait.py: validate in __init__ and raise ValueError('wait_chain() requires at least one strategy') so the failure happens at the point where the bad configuration is created, not later when the wait function is first evaluated.

Tests

tests/test_tenacity.py:

• test_wait_chain_requires_at_least_one_strategy — asserts ValueError from the bare wait_chain() constructor and from the wrapped Retrying(wait=wait_chain()) path that previously surfaced the IndexError.

The change is one validation line plus the test. A release note is included in releasenotes/notes/.

[mergify]

Merge Queue Status

• ✅ Entered queue — 2026-07-03 08:05 UTC · Rule: default · triggered by rule autoqueue
• ✅ Checks passed · on draft merge queue: checking main (a4b93c0) and #647 together #650
• ✅ Merged — 2026-07-03 08:05 UTC · at ce53bc85245373a1b8b605ae78f5f2bc2de46fe2 · squash

This pull request spent 16 seconds in the queue, including 2 seconds running CI.

Required conditions to merge