| |
|
|
|
|
|
|
| |
Re-enabling native-comp surfaced a suite-wide fragility. When a test redefines a C primitive (or a native-compiled function), native-comp routes native callers through a trampoline that calls the mock with the primitive's maximum arity. A fixed-arity mock narrower than the primitive then throws wrong-number-of-arguments, intermittently, as the eln-cache fills.
I swept every arity-narrow subr mock to append &rest _ (188 sites, preserving any named args the body uses), and added tests/test-meta-subr-mock-arity.el, which fails make test on any subr mock too narrow for the primitive's arity. The rule isn't "never mock a subr". The suite mocks message and completing-read freely. It's "a subr mock must accept the primitive's arity."
Background, the three failure modes, and the research are in docs/native-comp-subr-mocking.org.
|
|
|
`cj/python-setup` and `cj/webdev-setup` are the per-buffer preference functions wired into `python-ts-mode` / `typescript-ts-mode` / `tsx-ts-mode` / `js-ts-mode`. Sibling tests covered the command builders and prettier/mypy wrappers but not these. 9 new tests across two files cover, for each setup:
- the setq-local pairs (fill-column, tab-width, standard-indent, indent-tabs-mode) land at the language's defaults (Python: 80/4/4/nil; webdev: 100/2/2/nil),
- the mode helpers (company, flyspell-prog, superword, electric-pair-local) all get called,
- LSP gating: starts when the language server binary is on PATH, skips otherwise.
Plus a check that `cj/webdev-keybindings` binds `C-; f` to the prettier wrapper.
External modes and `executable-find` / `lsp-deferred` are stubbed so the tests run without a Python or Node toolchain.
|
