router: classify_model heuristic + 31-case corpus (Phase 5 commit #1)
Phase 5 commit #1 per docs/PHASE5.md §11. Pure-Lua per-request model routing — no IO, no LLM probe in v1. router.classify_model(text, cfg) -> (model_name | nil, class_label): 1. classify_class(text) walks heuristics in priority order: code class: - triple-backtick fence anywhere - "traceback" / "stacktrace" / "stack trace" (ci) - "error:" / "exception:" in first 60 chars (ci) - path-with-code-extension token (.py/.lua/.c/.js/.go/.rs/.cpp/.h/.ts) - 5+ lines with indented content (looks like a paste) reasoning class (requires text >= 15 chars to skip bare keywords): - "explain" / "why " / "how does" / "compare" (ci) - "?" + length > 100 chars default class: everything else 2. Map class via cfg.routing.classes[class] → model name (or nil = keep current). 3. Return (model_name_or_nil, class_label). ALWAYS evaluates regardless of cfg.routing.auto — caller (repl.ask_ai in commit #3) gates on the flag. This separation lets `:route check` introspect the heuristic even when routing is off (N1). M._classify_class exposed for testing. Test corpus (test_router_model.lua, 31 cases): - 13 code-class positives (fence, traceback, paths, multi-line paste) - 6 reasoning-class positives (explain/why/how does/compare/?+length) - 8 default-class (short queries, bare keywords below 15-char threshold, non-code paths like .md/.txt) - 3 model-mapping cases (code→"deep", reasoning→"cloud", default→nil) - 1 R-N2 default test: classes.reasoning=nil → reasoning text yields nil model override (heuristic still fires, no swap) - All 31 pass; 15-char threshold catches "how does ASLR work?" without false-positive on bare "explain". Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
+81
-8
@@ -1,12 +1,9 @@
|
|||||||
-- router.lua — task classifier: meta / shell / AI.
|
-- router.lua — task classifier: meta / shell / AI / model-routing.
|
||||||
-- See docs/PHASE0.md §5.
|
-- See docs/PHASE0.md §5 and docs/PHASE5.md §4 for Phase 5 additions.
|
||||||
--
|
--
|
||||||
-- Pure function. Takes (line, config). Returns (kind, payload).
|
-- M.classify(line, config) → (kind, payload) for input dispatch (Phase 0).
|
||||||
-- kind : "meta" | "shell" | "ai"
|
-- M.classify_model(text, cfg) → name | nil for per-request model routing
|
||||||
-- payload: the (possibly stripped) line that the dispatcher should act on.
|
-- (Phase 5; pure-Lua heuristics, no IO).
|
||||||
--
|
|
||||||
-- Empty / whitespace-only lines are returned as ("ai", ""); the repl loop
|
|
||||||
-- skips them before dispatching.
|
|
||||||
|
|
||||||
local M = {}
|
local M = {}
|
||||||
|
|
||||||
@@ -63,4 +60,80 @@ function M.classify(line, config)
|
|||||||
return "ai", line
|
return "ai", line
|
||||||
end
|
end
|
||||||
|
|
||||||
|
-- ---------------------------------------------------------------- classify_model
|
||||||
|
-- Phase 5: per-request model routing heuristic. Pure-Lua, no IO.
|
||||||
|
-- Returns the NAME of a model preset (string) to switch to for this
|
||||||
|
-- request, or nil to keep the active model unchanged.
|
||||||
|
--
|
||||||
|
-- The mapping from class to model name lives in `cfg.routing.classes`.
|
||||||
|
-- A class with value `nil` means "keep current" — even though the
|
||||||
|
-- heuristic fires, no override happens (used by default for the
|
||||||
|
-- `reasoning` class per R-N2 cost-safety policy).
|
||||||
|
--
|
||||||
|
-- This function ALWAYS evaluates the heuristic regardless of
|
||||||
|
-- `cfg.routing.auto` — the caller (repl.ask_ai) gates on the flag.
|
||||||
|
-- This separation lets `:route check <text>` introspect the heuristic
|
||||||
|
-- even when routing is disabled (N1).
|
||||||
|
|
||||||
|
local function classify_class(text)
|
||||||
|
if not text or text == "" then return "default" end
|
||||||
|
|
||||||
|
-- ── Code class — looks like a paste or contains code markers
|
||||||
|
if text:find("```", 1, true) then return "code" end
|
||||||
|
local lower = text:lower()
|
||||||
|
if lower:find("traceback", 1, true)
|
||||||
|
or lower:find("stacktrace", 1, true)
|
||||||
|
or lower:find("stack trace", 1, true) then
|
||||||
|
return "code"
|
||||||
|
end
|
||||||
|
-- exception/error markers near beginning (first 60 chars)
|
||||||
|
if lower:sub(1, 60):find("error:", 1, true)
|
||||||
|
or lower:sub(1, 60):find("exception:", 1, true) then
|
||||||
|
return "code"
|
||||||
|
end
|
||||||
|
-- path with code-extension token
|
||||||
|
if text:match("[%./~][%w%-_/.]+%.([%w]+)") then
|
||||||
|
local ext = text:match("[%./~][%w%-_/.]+%.([%w]+)")
|
||||||
|
if ext == "py" or ext == "lua" or ext == "c"
|
||||||
|
or ext == "js" or ext == "go" or ext == "rs"
|
||||||
|
or ext == "cpp" or ext == "h" or ext == "ts" then
|
||||||
|
return "code"
|
||||||
|
end
|
||||||
|
end
|
||||||
|
-- multi-line + indented (looks like a code paste)
|
||||||
|
local nlines = 0
|
||||||
|
for _ in (text .. "\n"):gmatch("[^\n]*\n") do nlines = nlines + 1 end
|
||||||
|
if nlines > 4 and text:find("\n%s+%S") then return "code" end
|
||||||
|
|
||||||
|
-- ── Reasoning class
|
||||||
|
-- Min length 15 — catches "how does X work" but excludes bare "why" / "explain"
|
||||||
|
if #text >= 15 then
|
||||||
|
if lower:find("explain", 1, true)
|
||||||
|
or lower:find("why ", 1, true) -- trailing space (not "whyever")
|
||||||
|
or lower:find("how does", 1, true)
|
||||||
|
or lower:find("compare", 1, true) then
|
||||||
|
return "reasoning"
|
||||||
|
end
|
||||||
|
end
|
||||||
|
if text:find("?", 1, true) and #text > 100 then
|
||||||
|
return "reasoning"
|
||||||
|
end
|
||||||
|
|
||||||
|
return "default"
|
||||||
|
end
|
||||||
|
|
||||||
|
-- Public API.
|
||||||
|
function M.classify_model(text, cfg)
|
||||||
|
local class = classify_class(text)
|
||||||
|
local classes = (cfg and cfg.routing and cfg.routing.classes) or {}
|
||||||
|
local target = classes[class]
|
||||||
|
-- nil target = keep current (this is the R-N2 default for "reasoning")
|
||||||
|
if target == nil then return nil, class end
|
||||||
|
-- Caller may want the class label for the status line; return both.
|
||||||
|
return target, class
|
||||||
|
end
|
||||||
|
|
||||||
|
-- Exposed for `:route check` introspection (N1).
|
||||||
|
M._classify_class = classify_class
|
||||||
|
|
||||||
return M
|
return M
|
||||||
|
|||||||
@@ -0,0 +1,108 @@
|
|||||||
|
-- test_router_model.lua — Phase 5 commit #1 corpus for classify_model.
|
||||||
|
-- Run from repo root: `luajit test_router_model.lua` (exit 0 on pass).
|
||||||
|
|
||||||
|
package.path = "./?.lua;./vendor/?.lua;" .. package.path
|
||||||
|
local router = require("router")
|
||||||
|
|
||||||
|
local cfg = {
|
||||||
|
routing = {
|
||||||
|
auto = true,
|
||||||
|
classes = {
|
||||||
|
code = "deep",
|
||||||
|
reasoning = "cloud",
|
||||||
|
default = nil, -- nil → keep current
|
||||||
|
},
|
||||||
|
},
|
||||||
|
}
|
||||||
|
|
||||||
|
-- (text, expected_class)
|
||||||
|
local CASES = {
|
||||||
|
-- ── code class
|
||||||
|
{ "```python\ndef foo(): pass\n```", "code" },
|
||||||
|
{ "Traceback (most recent call last):", "code" },
|
||||||
|
{ "got a stack trace from my server", "code" },
|
||||||
|
{ "error: undefined reference to `foo'", "code" },
|
||||||
|
{ "exception: file not found", "code" },
|
||||||
|
{ "please look at ./src/main.lua", "code" },
|
||||||
|
{ "the issue is in ~/repos/foo/bar.py", "code" },
|
||||||
|
{ "check /usr/lib/python3/dist-packages/x.py", "code" },
|
||||||
|
{ "fix this:\n if x == 0:\n return\n else:\n pass\n", "code" },
|
||||||
|
|
||||||
|
-- ── reasoning class
|
||||||
|
{ "Explain how MMAP works on Linux", "reasoning" },
|
||||||
|
{ "why does my disk fill up so fast", "reasoning" },
|
||||||
|
{ "how does ASLR work?", "reasoning" },
|
||||||
|
{ "compare ZFS and btrfs in terms of snapshots", "reasoning" },
|
||||||
|
{ "Can you explain the difference between a process and a thread in detail?", "reasoning" },
|
||||||
|
{ "I have a long question with a question mark in it that goes well past one hundred characters does it route to reasoning?", "reasoning" },
|
||||||
|
|
||||||
|
-- ── default class — short queries, no markers
|
||||||
|
{ "hi", "default" },
|
||||||
|
{ "what time is it", "default" },
|
||||||
|
{ "ls /tmp", "default" },
|
||||||
|
{ "thanks", "default" },
|
||||||
|
{ "explain", "default" }, -- bare "explain" < 30 chars threshold
|
||||||
|
{ "why", "default" }, -- bare "why"
|
||||||
|
{ "?", "default" }, -- bare ?
|
||||||
|
{ "hello world", "default" },
|
||||||
|
|
||||||
|
-- ── edge: ambiguous — prefer false-positive into reasoning over false-negative
|
||||||
|
{ "How does it feel to be a robot? Just curious.", "reasoning" }, -- 47 chars + how does
|
||||||
|
-- ── edge: short error mention should still be code
|
||||||
|
{ "got error: foo", "code" },
|
||||||
|
|
||||||
|
-- ── edge: a non-code path-like (e.g. README.md, .txt) should NOT match
|
||||||
|
{ "see notes.md for details", "default" },
|
||||||
|
{ "lookup /tmp/x.txt", "default" },
|
||||||
|
}
|
||||||
|
|
||||||
|
local pass, fail = 0, 0
|
||||||
|
local fails = {}
|
||||||
|
for i, c in ipairs(CASES) do
|
||||||
|
local _model, class = router.classify_model(c[1], cfg)
|
||||||
|
if class == c[2] then
|
||||||
|
pass = pass + 1
|
||||||
|
else
|
||||||
|
fail = fail + 1
|
||||||
|
fails[#fails+1] = string.format(" [%2d] text=%q expected=%s got=%s",
|
||||||
|
i, c[1]:sub(1, 60), c[2], tostring(class))
|
||||||
|
end
|
||||||
|
end
|
||||||
|
|
||||||
|
print(string.format("router.classify_model: %d/%d pass", pass, pass+fail))
|
||||||
|
for _, f in ipairs(fails) do print(f) end
|
||||||
|
|
||||||
|
-- Verify model routing: code → "deep", reasoning → "cloud", default → nil
|
||||||
|
local cases_model = {
|
||||||
|
{ "Traceback", "deep", "code" },
|
||||||
|
{ "Explain in detail how X works", "cloud", "reasoning" },
|
||||||
|
{ "hi", nil, "default" },
|
||||||
|
}
|
||||||
|
print()
|
||||||
|
for _, c in ipairs(cases_model) do
|
||||||
|
local m, k = router.classify_model(c[1], cfg)
|
||||||
|
if m == c[2] and k == c[3] then
|
||||||
|
pass = pass + 1
|
||||||
|
else
|
||||||
|
fail = fail + 1
|
||||||
|
fails[#fails+1] = string.format(
|
||||||
|
" model: text=%q expected (%s,%s) got (%s,%s)",
|
||||||
|
c[1], tostring(c[2]), c[3], tostring(m), tostring(k))
|
||||||
|
end
|
||||||
|
end
|
||||||
|
|
||||||
|
-- Verify the R-N2 default: when classes.reasoning = nil, reasoning text → no override
|
||||||
|
local cfg_safe = { routing = { auto = true, classes = { code = "deep", reasoning = nil } } }
|
||||||
|
local m, k = router.classify_model("Explain in detail how X works", cfg_safe)
|
||||||
|
if m == nil and k == "reasoning" then
|
||||||
|
pass = pass + 1
|
||||||
|
else
|
||||||
|
fail = fail + 1
|
||||||
|
fails[#fails+1] = string.format(
|
||||||
|
" reasoning-nil-default: got (%s,%s); want (nil,reasoning)",
|
||||||
|
tostring(m), tostring(k))
|
||||||
|
end
|
||||||
|
|
||||||
|
print(string.format("\nTOTAL: %d/%d pass", pass, pass+fail))
|
||||||
|
for _, f in ipairs(fails) do print(f) end
|
||||||
|
os.exit(fail == 0 and 0 or 1)
|
||||||
Reference in New Issue
Block a user