Skip to content

Exception handling

Avoid exception handling in conversions as long as possible, because it reduces clarity and can hide issues in the code.

c.try_

c.try_(conv).except_(exc_def, value, re_raise_if=None) conversion wrapper allows to handle exceptions:

  • conv: a conversion, which is expected to raise an exception
  • exc_def: exception class or tuple of exception classes to be caught
  • value: a value or conversion to return when exc_def is caught and re_raise_if is absent or evaluates to false
  • re_raise_if (optional): if it evaluates to true, the caught exception is re-raised

Both value and re_raise_if can work with the input data as usual, or reference the caught exception as c.EXCEPTION.

try_ only guards evaluation of the wrapped expression. Exceptions raised while iterating a generator produced inside it (for example c.iter(...)) escape the handler. Materialize inside the wrapper:

c.try_(c.iter(c.item("a")).as_type(list)).except_(KeyError, [])

from convtools import conversion as c

converter = (
    c.iter(
        c.try_(c.item(0) / c.item(1))
        .except_(ZeroDivisionError, value=c.this, re_raise_if=c.item(0) == 0)
        .except_(TypeError, None)
    )
    .as_type(list)
    .gen_converter()
)

assert converter([(1, 2), (3, 0), (4, "abc")]) == [0.5, (3, 0), None]