You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Last tag: 0.4.0, cut 2026-07-06 (34 days ago). 6 commits on master since.
No CHANGELOG.md in this repo, so from the commit log:
Lua numbers are carried as doubles (carry Lua numbers as doubles #11). A Lua number is a C double, so push-double/get-double were added and registered against lua_pushnumber/lua_tonumber directly, along with the matching Luax helpers (set-double-global, get-double-field, …). push-float/get-float still work — they are now thin wrappers that convert — but they round to single precision and turn anything above the float range into an infinity, which the docstrings and README now say.
Proposed bump: minor → 0.5.0. New *-double API, additive: the float entry points keep their names and types, so existing code compiles unchanged (it just keeps the old precision loss, now documented).
Opened by the carpentry-org heartbeat agent (Claude). Veit has not reviewed this yet.
Last tag: 0.4.0, cut 2026-07-06 (34 days ago). 6 commits on
mastersince.No
CHANGELOG.mdin this repo, so from the commit log:push-double/get-doublewere added and registered againstlua_pushnumber/lua_tonumberdirectly, along with the matchingLuaxhelpers (set-double-global,get-double-field, …).push-float/get-floatstill work — they are now thin wrappers that convert — but they round to single precision and turn anything above the float range into an infinity, which the docstrings and README now say.get-string-global,val,eval-fileanddo-in(fix stack-slot leaks in get-string-global, val, eval-file, and do-in #10).Proposed bump: minor → 0.5.0. New
*-doubleAPI, additive: thefloatentry points keep their names and types, so existing code compiles unchanged (it just keeps the old precision loss, now documented).Opened by the carpentry-org heartbeat agent (Claude). Veit has not reviewed this yet.