Fix rules CSV parsing and add rule text editing API

Strip a leading UTF-8 BOM before parsing rule CSVs (it was causing the
header row to be misread as a rule) and skip rows that clearly failed
CSV parsing (e.g. a whole line landing in the rule_number field after
a spreadsheet app double-encoded a quoted cell) instead of silently
loading garbage.

Add PUT /api/rules/{lang}/{number} and GET /api/rules/numbers so rule
text, suggested penalty and escalation mode can be edited per language
from the admin UI, always writing the CSV back out with correct
quoting via encoding/csv. Also fixes two bugs found while wiring this
up: rulesDir() ignored the configured rules_dir (falling back to a
RULES_DIR env var that's never set), and the CORS middleware didn't
allow PUT.

rules.go

@@ -1,12 +1,17 @@
 package main
 
 import (
+	"bufio"
 	"encoding/csv"
+	"encoding/json"
+	"log"
 	"net/http"
 	"os"
 	"path/filepath"
+	"regexp"
 	"strings"
 	"sync"
+	"unicode/utf8"
 )
 
 var (
@@ -14,6 +19,72 @@ var (
 	rules   = map[string]map[string]Rule{}
 )
 
+var langNameRe = regexp.MustCompile(`^[a-z]{2,5}$`)
+
+func rulesDir() string {
+	if configuredRulesDir != "" {
+		return configuredRulesDir
+	}
+	return "rules"
+}
+
+// readCSVFile reads a CSV file, transparently stripping a leading UTF-8 BOM
+// (common when the file has been round-tripped through a spreadsheet app).
+func readCSVFile(path string) ([][]string, error) {
+	f, err := os.Open(path)
+	if err != nil {
+		return nil, err
+	}
+	defer f.Close()
+	br := bufio.NewReader(f)
+	if bom, err := br.Peek(3); err == nil && bom[0] == 0xEF && bom[1] == 0xBB && bom[2] == 0xBF {
+		br.Discard(3)
+	}
+	r := csv.NewReader(br)
+	r.FieldsPerRecord = -1
+	return r.ReadAll()
+}
+
+// writeCSVFile writes records back out with correct quoting, which is what
+// prevents the "double-encoded row" corruption spreadsheet apps tend to cause.
+func writeCSVFile(path string, records [][]string) error {
+	f, err := os.Create(path)
+	if err != nil {
+		return err
+	}
+	w := csv.NewWriter(f)
+	if err := w.WriteAll(records); err != nil {
+		f.Close()
+		return err
+	}
+	w.Flush()
+	if err := w.Error(); err != nil {
+		f.Close()
+		return err
+	}
+	return f.Close()
+}
+
+// upsertRuleRecord updates the row for `number` in-place, or appends a new
+// row if it isn't present yet. records[0] is assumed to be the header.
+func upsertRuleRecord(records [][]string, number, text, penalty, escalation string) [][]string {
+	for i, rec := range records {
+		if i == 0 || len(rec) < 1 {
+			continue
+		}
+		if strings.TrimSpace(rec[0]) != number {
+			continue
+		}
+		for len(rec) < 4 {
+			rec = append(rec, "")
+		}
+		rec[1], rec[2], rec[3] = text, penalty, escalation
+		records[i] = rec
+		return records
+	}
+	return append(records, []string{number, text, penalty, escalation})
+}
+
 func loadRules(dir string) error {
 	entries, err := os.ReadDir(dir)
 	if err != nil {
@@ -25,14 +96,7 @@ func loadRules(dir string) error {
 			continue
 		}
 		lang := strings.TrimSuffix(e.Name(), ".csv")
-		f, err := os.Open(filepath.Join(dir, e.Name()))
-		if err != nil {
-			continue
-		}
-		r := csv.NewReader(f)
-		r.FieldsPerRecord = -1
-		records, err := r.ReadAll()
-		f.Close()
+		records, err := readCSVFile(filepath.Join(dir, e.Name()))
 		if err != nil {
 			continue
 		}
@@ -47,6 +111,10 @@ func loadRules(dir string) error {
 					continue
 				}
 			}
+			if strings.ContainsAny(rec[0], ",\"") || !utf8.ValidString(rec[0]) {
+				log.Printf("rules: skipping malformed row %d in %s: %q", i+1, e.Name(), rec[0])
+				continue
+			}
 			ru := Rule{
 				Number:           strings.TrimSpace(rec[0]),
 				Text:             strings.TrimSpace(rec[1]),
@@ -91,6 +159,8 @@ func getRules(lang string) []Rule {
 func registerRuleRoutes(mux *http.ServeMux) {
 	mux.HandleFunc("GET /api/rules", requireAuth(handleListRules))
 	mux.HandleFunc("GET /api/rules/languages", requireAuth(handleListRuleLanguages))
+	mux.HandleFunc("GET /api/rules/numbers", requireAuth(handleListRuleNumbers))
+	mux.HandleFunc("PUT /api/rules/{lang}/{number}", requireAdmin(handleUpdateRule))
 	mux.HandleFunc("POST /api/rules/reload", requireAdmin(handleReloadRules))
 }
 
@@ -114,13 +184,90 @@ func handleListRules(w http.ResponseWriter, r *http.Request) {
 }
 
 func handleReloadRules(w http.ResponseWriter, r *http.Request) {
-	dir := os.Getenv("RULES_DIR")
-	if dir == "" {
-		dir = "rules"
-	}
-	if err := loadRules(dir); err != nil {
+	if err := loadRules(rulesDir()); err != nil {
 		writeError(w, http.StatusInternalServerError, "load_error")
 		return
 	}
 	w.WriteHeader(http.StatusNoContent)
 }
+
+// handleListRuleNumbers returns the union of rule numbers known across every
+// loaded language, so the rules-editor UI can show a row even for a rule
+// that a given language's file hasn't been given a translation for yet.
+func handleListRuleNumbers(w http.ResponseWriter, r *http.Request) {
+	rulesMu.RLock()
+	seen := map[string]bool{}
+	out := make([]string, 0, 64)
+	for _, langMap := range rules {
+		for number := range langMap {
+			if !seen[number] {
+				seen[number] = true
+				out = append(out, number)
+			}
+		}
+	}
+	rulesMu.RUnlock()
+	writeJSON(w, http.StatusOK, out)
+}
+
+func handleUpdateRule(w http.ResponseWriter, r *http.Request) {
+	lang := r.PathValue("lang")
+	number := strings.TrimSpace(r.PathValue("number"))
+	if !langNameRe.MatchString(lang) {
+		writeError(w, http.StatusBadRequest, "invalid_language")
+		return
+	}
+	path := filepath.Join(rulesDir(), lang+".csv")
+	if _, err := os.Stat(path); err != nil {
+		writeError(w, http.StatusNotFound, "unknown_language")
+		return
+	}
+	if number == "" {
+		writeError(w, http.StatusBadRequest, "invalid_number")
+		return
+	}
+	var req struct {
+		Text             string `json:"text"`
+		SuggestedPenalty string `json:"suggested_penalty"`
+		EscalationMode   string `json:"escalation_mode"`
+	}
+	if err := json.NewDecoder(r.Body).Decode(&req); err != nil {
+		writeError(w, http.StatusBadRequest, "invalid_body")
+		return
+	}
+	req.Text = strings.TrimSpace(req.Text)
+	if req.Text == "" {
+		writeError(w, http.StatusBadRequest, "missing_text")
+		return
+	}
+	if req.EscalationMode == "" {
+		req.EscalationMode = "same"
+	}
+
+	records, err := readCSVFile(path)
+	if err != nil {
+		writeError(w, http.StatusInternalServerError, "read_error")
+		return
+	}
+	if len(records) == 0 {
+		records = [][]string{{"rule_number", "rule_text", "suggested_penalty", "escalation_mode"}}
+	}
+	records = upsertRuleRecord(records, number, req.Text, req.SuggestedPenalty, req.EscalationMode)
+	if err := writeCSVFile(path, records); err != nil {
+		writeError(w, http.StatusInternalServerError, "write_error")
+		return
+	}
+	if err := loadRules(rulesDir()); err != nil {
+		writeError(w, http.StatusInternalServerError, "reload_error")
+		return
+	}
+
+	rulesMu.RLock()
+	ru, ok := rules[lang][number]
+	rulesMu.RUnlock()
+	if !ok {
+		writeError(w, http.StatusInternalServerError, "reload_mismatch")
+		return
+	}
+	writeJSON(w, http.StatusOK, ru)
+}