unicorn/catch-error-name

When one file catches e, the next err and a third ex, you need to look up the name every time you read or move a catch block. With this rule the caught error is always called error.

The rule checks catch clauses, the first parameter of a function passed to .catch(), and the first parameter of the second function passed to .then().

--fix renames the parameter and every use of it. It leaves the parameter alone if the code around it already uses the new name for another variable, because the rename would mix the two up.

Configure#

Off by default. To turn it on:

bunfig.toml

[lint.rules]
"unicorn/catch-error-name" = "error" # or "warn"

Every option, set to its default:

bunfig.toml

[lint.rules]
"unicorn/catch-error-name" = ["error", {
  name = "error",
  ignore = [], # list of strings
}]

Examples#

example.js3 problems

try {
  await save(user);
} catch (e) {
  console.log(e.message);
}

save(user).catch(err => console.log(err.message));

save(user).then(done, ex => console.log(ex.message));

example.jsNo problems

try {
  await save(user);
} catch (error) {
  console.log(error.message);
}

save(user).catch(error => console.log(error.message));

save(user).then(done, error => console.log(error.message));

A name that ends in error is fine, such as saveError, or error_ in a nested catch. So is a name that starts with _, as long as the error isn't used.

example.jsNo problems

try {
  await save(user);
} catch (saveError) {
  try {
    await rollback();
  } catch (error_) {
    report(saveError, error_);
  }
}

save(user).catch(_ => retry());

Options#

Option	Default	What it does
`name`	`"error"`	The name to require. A value that isn't a valid identifier is ignored.
`ignore`	`[]`	Regular expressions, as strings. A name that matches one of them is allowed.

example.jswith { "name": "err" }No problems

save(user).catch(err => console.log(err.message));

example.jswith { "ignore": ["^reason$"] }No problems

save(user).catch(reason => console.log(reason));