Error Handling

Goals

  • Make failure paths explicit and understandable.
  • Catch exceptions only where you can recover.
  • Preserve error context for debugging.

Guidance

  • Prefer EAFP where it improves clarity, but keep failure scope tight.
  • Catch specific exceptions; avoid blanket catches.
  • Batch tasks can isolate per-item failures; pipelines often stop early.
  • When re-raising, keep the original context.

Examples

EAFP vs LBYL style checks:

if can_i_do_x():
    do_x()
else:
    handle_error()

try:
    do_x()
except SomeError:
    handle_error()

Raising and handling a specific error path:

def foo():
    # do something
    raise Exception("something wrong")

try:
    foo()
except Exception as e:
    # handle exception

Batch-style call site:

for bar in bars:
    foo(bar)

Sources & References