René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"`SyntaxError: parameter without a default follows parameter with a default` is inaccurate","articleBody":"# Bug report\n\n### Bug description:\n\n#### The problem\n\nTrying to define a function in which a positional parameter follows a parameter with a default value, we get a `SyntaxError` as expected:\n\n```pycon\n\u003e\u003e\u003e def f(a=1, b):\n  File \"\u003cpython-input-0\u003e\", line 1\n    def f(a=1, b):\n               ^\nSyntaxError: parameter without a default follows parameter with a default\n```\n\nBut the message that comes with it is not as accurate as it could be, as we can easily construct a valid function signature in which a parameter without a default follows a parameter with a default so long as both are keyword-only parameters:\n\n```pycon\n\u003e\u003e\u003e def f(*, a=1, b):\n...     pass\n...     \n\u003e\u003e\u003e \n```\n\n(I didn't use `a=1, *, b` for this counterexample because then one could argue that, if \"follows\" is interpreted as \"follows _immediately_\", we did resolve the situation described in the error message, making it formally accurate, albeit confusing.)\n\n#### Proposed fix\n\nSo, in my opinion, the message should be changed to something more like `positional parameter without a default follows parameter with a default`, reflecting the true nature of the problem.\n\n#### Other benefits\n\nIt's not _just_ a matter of accuracy for accuracy's sake, however: This change would also hint to people who don't know or have forgotten about keyword-only parameters that there is another possibility of resolving the issue than the ones implied by the current message.\n\n#### Additional context\n\n- As a result of https://github.com/python/cpython/issues/91210, the error message was changed from `non-default argument follows default argument` to the one above starting in Python 3.12 (via https://github.com/python/cpython/pull/95933). So there is precedent for changing this message to make it more accurate.\n- Keyword-only parameters were introduced in Python 3.0 with [PEP 3102](https://peps.python.org/pep-3102/), which makes me think I must be missing some obvious reason why their consequences can't be reflected in the error message.\n- `dataclasses` has an exception message with the same issue, but as that one also still has the \"argument\" vs \"parameter\" issue mentioned above, it's probably appropriate to create a separate ticket for that.\n\n### CPython versions tested on:\n\n3.14\n\n### Operating systems tested on:\n\nLinux\n\n\u003c!-- gh-linked-prs --\u003e\n### Linked PRs\n* gh-130937\n\u003c!-- /gh-linked-prs --\u003e\n","author":{"url":"https://github.com/smheidrich","@type":"Person","name":"smheidrich"},"datePublished":"2025-02-01T11:45:27.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":3},"url":"https://github.com/129545/cpython/issues/129545"}