fed-sx-m1: Step 2b — envelope:canonical_bytes/1 + 8 determinism tests

next/kernel/envelope.erl

@@ -1,5 +1,5 @@
 -module(envelope).
--export([validate_shape/1, get_field/2]).
+-export([validate_shape/1, get_field/2, canonical_bytes/1]).
 
 %% Activity envelope per design §3.1.
 %%
@@ -51,3 +51,35 @@ validate_signature_shape(Env) ->
         false ->
             {error, {bad_signature, not_a_proplist}}
     end.
+
+%% canonical_bytes/1 — the byte string the signature covers.
+%%
+%% Real fed-sx will use dag-cbor over a JSON-LD-canonicalised form
+%% (design §3.2). For milestone 1 we stand in for that with the host
+%% BIF `cid:to_string/1`, which produces a CIDv1 over the deterministic
+%% textual form of the term. Two prior steps make this work:
+%%   1. The signature pair is stripped (sig covers everything except
+%%      itself).
+%%   2. The top-level property list is sorted by key so field order in
+%%      the source envelope is not load-bearing.
+%%
+%% The result is an Erlang binary suitable as the sig-cover input.
+
+canonical_bytes(Env) when is_list(Env) ->
+    Stripped = strip_signature(Env),
+    Sorted = sort_pairs(Stripped),
+    cid:to_string(Sorted).
+
+strip_signature([]) -> [];
+strip_signature([{signature, _} | Rest]) -> strip_signature(Rest);
+strip_signature([P | Rest]) -> [P | strip_signature(Rest)].
+
+sort_pairs([]) -> [];
+sort_pairs([H | T]) -> insert_pair(H, sort_pairs(T)).
+
+insert_pair(P, []) -> [P];
+insert_pair({K1, V1}, [{K2, V2} | Rest]) ->
+    case K1 < K2 of
+        true  -> [{K1, V1}, {K2, V2} | Rest];
+        false -> [{K2, V2} | insert_pair({K1, V1}, Rest)]
+    end.

next/tests/envelope_canonical.sh

@@ -0,0 +1,105 @@
+#!/usr/bin/env bash
+# next/tests/envelope_canonical.sh — Step 2b acceptance test.
+#
+# Loads next/kernel/envelope.erl and checks canonical_bytes/1 contract:
+# returns a binary, deterministic across runs, invariant under
+# field-order permutation, invariant under signature changes, and
+# different for different covered content. 7 cases.
+
+set -uo pipefail
+cd "$(git rev-parse --show-toplevel)"
+
+SX_SERVER="${SX_SERVER:-hosts/ocaml/_build/default/bin/sx_server.exe}"
+if [ ! -x "$SX_SERVER" ]; then
+  SX_SERVER="/root/rose-ash/hosts/ocaml/_build/default/bin/sx_server.exe"
+fi
+if [ ! -x "$SX_SERVER" ]; then
+  echo "ERROR: sx_server.exe not found." >&2
+  exit 1
+fi
+
+VERBOSE="${1:-}"
+PASS=0; FAIL=0; ERRORS=""
+TMPFILE=$(mktemp); trap "rm -f $TMPFILE" EXIT
+
+cat > "$TMPFILE" <<'EPOCHS'
+(epoch 1)
+(load "lib/erlang/tokenizer.sx")
+(load "lib/erlang/parser.sx")
+(load "lib/erlang/parser-core.sx")
+(load "lib/erlang/parser-expr.sx")
+(load "lib/erlang/parser-module.sx")
+(load "lib/erlang/transpile.sx")
+(load "lib/erlang/runtime.sx")
+(load "lib/erlang/vm/dispatcher.sx")
+
+(epoch 2)
+(eval "(get (erlang-load-module (file-read \"next/kernel/envelope.erl\")) :name)")
+
+;; canonical_bytes returns a binary
+(epoch 10)
+(eval "(get (erlang-eval-ast \"is_binary(envelope:canonical_bytes([{id,1},{type,create},{actor,alice},{published,1000},{signature,whatever}]))\") :name)")
+
+;; Determinism: same envelope twice -> same bytes
+(epoch 11)
+(eval "(get (erlang-eval-ast \"envelope:canonical_bytes([{id,1},{type,create},{actor,alice}]) =:= envelope:canonical_bytes([{id,1},{type,create},{actor,alice}])\") :name)")
+
+;; Signature stripping: different signatures -> same canonical bytes
+(epoch 12)
+(eval "(get (erlang-eval-ast \"envelope:canonical_bytes([{id,1},{type,create},{actor,alice},{signature,sig_one}]) =:= envelope:canonical_bytes([{id,1},{type,create},{actor,alice},{signature,sig_two}])\") :name)")
+
+;; No signature vs some signature -> same canonical bytes
+(epoch 13)
+(eval "(get (erlang-eval-ast \"envelope:canonical_bytes([{id,1},{type,create},{actor,alice}]) =:= envelope:canonical_bytes([{id,1},{type,create},{actor,alice},{signature,whatever}])\") :name)")
+
+;; Key-order invariance: reordering top-level fields -> same bytes
+(epoch 14)
+(eval "(get (erlang-eval-ast \"envelope:canonical_bytes([{id,1},{type,create},{actor,alice}]) =:= envelope:canonical_bytes([{actor,alice},{type,create},{id,1}])\") :name)")
+
+;; Changing a covered field changes the bytes
+(epoch 15)
+(eval "(get (erlang-eval-ast \"envelope:canonical_bytes([{id,1},{type,create},{actor,alice}]) =/= envelope:canonical_bytes([{id,2},{type,create},{actor,alice}])\") :name)")
+
+;; Distinct envelopes -> distinct bytes
+(epoch 16)
+(eval "(get (erlang-eval-ast \"envelope:canonical_bytes([{id,1},{type,create},{actor,alice}]) =/= envelope:canonical_bytes([{id,1},{type,update},{actor,bob}])\") :name)")
+EPOCHS
+
+OUTPUT=$(timeout 120 "$SX_SERVER" < "$TMPFILE" 2>/dev/null)
+
+check() {
+  local epoch="$1" desc="$2" expected="$3"
+  local actual
+  actual=$(echo "$OUTPUT" | awk -v e="$epoch" '
+    $0 ~ "^\\(ok-len " e " " { getline; print; exit }
+    $0 ~ "^\\(ok " e " "     { print; exit }
+    $0 ~ "^\\(error " e " "  { print; exit }
+  ')
+  [ -z "$actual" ] && actual="<no output for epoch $epoch>"
+  if echo "$actual" | grep -qF -- "$expected"; then
+    PASS=$((PASS+1))
+    [ "$VERBOSE" = "-v" ] && echo "  ok $desc"
+  else
+    FAIL=$((FAIL+1))
+    ERRORS+="  FAIL [$desc] (epoch $epoch) expected: $expected | actual: $actual
+"
+  fi
+}
+
+check  2  "module load name"                "envelope"
+check 10  "canonical_bytes returns binary"  "true"
+check 11  "deterministic"                   "true"
+check 12  "signature stripped (changes)"    "true"
+check 13  "signature stripped (absent)"     "true"
+check 14  "key-order invariant"             "true"
+check 15  "covered field change visible"    "true"
+check 16  "distinct envelopes distinct"     "true"
+
+TOTAL=$((PASS+FAIL))
+if [ $FAIL -eq 0 ]; then
+  echo "ok $PASS/$TOTAL next/tests/envelope_canonical.sh passed"
+else
+  echo "FAIL $PASS/$TOTAL passed, $FAIL failed:"
+  echo "$ERRORS"
+fi
+[ $FAIL -eq 0 ]