The wcc test drivers ran `ww build <bare-/tmp src>` with no -o, so the
compiler's <stem>.sepwork scratch landed beside the source and was never
cleaned: unbounded /tmp growth (2195 stale dirs observed) that fills tmpfs
and fabricates phantom test failures + silent harness aborts, and for
in-repo fixture builds leaked .sepwork into the tracked tree.
Each leaking build now writes its source + output inside a per-invocation
tmpdir, passes -o <tmpdir>/<stem> so the .sepwork lands inside it, and
rm -rf's the tmpdir on every exit path -- including fopen-fail and the
expected-fail reject builds (scratch is mkdir'd before the build can fail).
`ww run` and explicit-`-o`/byte-id helpers are left as-is; the 990/993
byte-id comparison logic is byte-for-byte unchanged.
Two items filed separately (this commit holds the no-Makefile / no-main.c
rail):
- #13: a stale <src>.s byte-id readback (749) silently no-ops since
separate-compile emits .s to <ostem>.sepwork/__root.s; documented inline.
- #14: build-system Makefile recipes build selfhost/cmd/*/main.ww with no
-o and leak main.sepwork in-tree (bounded, gitignored; own commit).
One concern -- sepwork leak hygiene -- across 228 drivers; uniform
transform applied per-file and two-round reviewed. make test: all 402
passed, zero net-new /tmp scratch, zero test-driven in-repo .sepwork.
An overlong array literal (more initializers than the declared length) is
invalid -- cstage loud-rejects it everywhere -- but wwstage silently
accepted (and truncated) it in several positions; #9 wired only the decl
position. This folds the remaining three (one class: checkarrlitfits
over-fill coverage), all wwstage-only reject-align:
- #12a return fn f() [2]int = [1,2,3] -- silently accepted.
- #12b call-arg g([1,2,3]) -- louded only late via cgen #271.
- #106 alias type A=[2]int; let g: A = [1,2,3] -- silently truncated;
checkarrlitfits bailed on the N_TNAME alias without chasing.
Four inserts in check.ww: an alias-chase (resolvealias) at the top of
checkarrlitfits (makes all callers alias-aware), the over-fill check wired
into checkretassign (hoisted above the isassignable short-circuit) and
desugarcallargs, and the alias-let-global guard made alias-aware. cstage
unchanged (w6c md5 unchanged); reject-only, so no asm moves -- 990-997 8/8,
no lib byte-id pin flips. A 5th position (tuple-element overlong) is a
separate pre-existing hole, filed (#20). test/wcc/828 table-driven.
