golangLAKEHOUSE/cmd/matrixd/main.go

// matrixd is the matrix indexer service. Wraps internal/matrix's
// Retriever with HTTP routes per docs/SPEC.md §3.4.
//
// Routes:
//   POST /matrix/search     — multi-corpus retrieve+merge
//   GET  /matrix/corpora    — list known vectord indexes (proxy)
//   POST /matrix/relevance  — adjacency-pollution filter (CODE-aware;
//                              port of mcp-server/relevance.ts)
//   POST /matrix/downgrade  — strong-model auto-downgrade decision
//                              (port of mode.rs::execute pass5 gate)
//
// matrixd talks to embedd (for query-text embedding) and vectord
// (for per-corpus search) via HTTP. Both URLs come from
// [matrixd] config; gateway sets them to its own upstream URLs so
// matrixd inherits the same provider topology.
package main

import (
	"encoding/json"
	"errors"
	"flag"
	"log/slog"
	"net/http"
	"os"
	"strings"

	"github.com/go-chi/chi/v5"

	"git.agentview.dev/profit/golangLAKEHOUSE/internal/matrix"
	"git.agentview.dev/profit/golangLAKEHOUSE/internal/shared"
)

const maxRequestBytes = 4 << 20 // 4 MiB cap on request bodies

func main() {
	configPath := flag.String("config", "lakehouse.toml", "path to TOML config")
	flag.Parse()

	cfg, err := shared.LoadConfig(*configPath)
	if err != nil {
		slog.Error("config", "err", err)
		os.Exit(1)
	}
	if cfg.Matrixd.EmbeddURL == "" || cfg.Matrixd.VectordURL == "" {
		slog.Error("matrixd: embedd_url and vectord_url required in [matrixd]")
		os.Exit(1)
	}

	retriever := matrix.New(cfg.Matrixd.EmbeddURL, cfg.Matrixd.VectordURL)
	h := &handlers{r: retriever}

	if err := shared.Run("matrixd", cfg.Matrixd.Bind, h.register, cfg.Auth); err != nil {
		slog.Error("server", "err", err)
		os.Exit(1)
	}
}

type handlers struct {
	r *matrix.Retriever
}

func (h *handlers) register(r chi.Router) {
	r.Post("/matrix/search", h.handleSearch)
	r.Get("/matrix/corpora", h.handleCorpora)
	r.Post("/matrix/relevance", h.handleRelevance)
	r.Post("/matrix/downgrade", h.handleDowngrade)
}

func (h *handlers) handleSearch(w http.ResponseWriter, r *http.Request) {
	var req matrix.SearchRequest
	if !decodeJSON(w, r, &req) {
		return
	}
	resp, err := h.r.Search(r.Context(), req)
	if err != nil {
		writeMatrixError(w, err)
		return
	}
	writeJSON(w, http.StatusOK, resp)
}

// relevanceRequest is the POST /matrix/relevance body. Threshold
// defaults to matrix.DefaultRelevanceThreshold when zero.
type relevanceRequest struct {
	Focus     matrix.FocusFile        `json:"focus"`
	Chunks    []matrix.CandidateChunk `json:"chunks"`
	Threshold float64                 `json:"threshold,omitempty"`
}

func (h *handlers) handleRelevance(w http.ResponseWriter, r *http.Request) {
	var req relevanceRequest
	if !decodeJSON(w, r, &req) {
		return
	}
	if len(req.Chunks) == 0 {
		http.Error(w, "chunks must be non-empty", http.StatusBadRequest)
		return
	}
	threshold := req.Threshold
	if threshold == 0 {
		threshold = matrix.DefaultRelevanceThreshold
	}
	res := matrix.FilterChunks(req.Focus, req.Chunks, threshold)
	writeJSON(w, http.StatusOK, res)
}

// downgradeRequest is the POST /matrix/downgrade body. Mirrors
// matrix.DowngradeInput; ForceFullOverride is read from the
// LH_FORCE_FULL_ENRICHMENT env var when omitted from the body.
type downgradeRequest struct {
	Mode              string `json:"mode"`
	Model             string `json:"model"`
	ForcedMode        bool   `json:"forced_mode,omitempty"`
	ForceFullOverride *bool  `json:"force_full_override,omitempty"`
}

func (h *handlers) handleDowngrade(w http.ResponseWriter, r *http.Request) {
	var req downgradeRequest
	if !decodeJSON(w, r, &req) {
		return
	}
	if req.Mode == "" || req.Model == "" {
		http.Error(w, "mode and model are required", http.StatusBadRequest)
		return
	}
	in := matrix.NewDowngradeInputFromEnv(req.Mode, req.Model, req.ForcedMode)
	if req.ForceFullOverride != nil {
		// Explicit body override beats env, useful for tooling that
		// wants to ask "what would the gate do under these conditions"
		// without env pollution.
		in.ForceFullOverride = *req.ForceFullOverride
	}
	writeJSON(w, http.StatusOK, matrix.MaybeDowngrade(in))
}

func (h *handlers) handleCorpora(w http.ResponseWriter, r *http.Request) {
	names, err := h.r.Corpora(r.Context())
	if err != nil {
		slog.Error("matrix corpora", "err", err)
		http.Error(w, "vectord unavailable", http.StatusBadGateway)
		return
	}
	writeJSON(w, http.StatusOK, map[string]any{"corpora": names, "count": len(names)})
}

func decodeJSON(w http.ResponseWriter, r *http.Request, v any) bool {
	defer r.Body.Close()
	r.Body = http.MaxBytesReader(w, r.Body, maxRequestBytes)
	if err := json.NewDecoder(r.Body).Decode(v); err != nil {
		var maxErr *http.MaxBytesError
		if errors.As(err, &maxErr) || strings.Contains(err.Error(), "http: request body too large") {
			http.Error(w, "body too large", http.StatusRequestEntityTooLarge)
			return false
		}
		http.Error(w, "decode body: "+err.Error(), http.StatusBadRequest)
		return false
	}
	return true
}

func writeJSON(w http.ResponseWriter, code int, v any) {
	w.Header().Set("Content-Type", "application/json")
	w.WriteHeader(code)
	if err := json.NewEncoder(w).Encode(v); err != nil {
		slog.Warn("matrix write json", "err", err)
	}
}

// writeMatrixError maps internal/matrix sentinels to HTTP statuses.
// Corpus / embed failures bubble up as 502 (the upstream service is
// what's wrong); validation errors are 400.
func writeMatrixError(w http.ResponseWriter, err error) {
	switch {
	case errors.Is(err, matrix.ErrEmptyCorpora),
		errors.Is(err, matrix.ErrEmptyQuery):
		http.Error(w, err.Error(), http.StatusBadRequest)
	case errors.Is(err, matrix.ErrCorpus),
		errors.Is(err, matrix.ErrEmbed):
		slog.Warn("matrix upstream", "err", err)
		http.Error(w, err.Error(), http.StatusBadGateway)
	default:
		slog.Error("matrix", "err", err)
		http.Error(w, "internal", http.StatusInternalServerError)
	}
}