fed-sx-m1: Step 4d — bootstrap:build_genesis/verify_genesis + cidhash helpers + 12 tests

next/kernel/bootstrap.erl

@@ -1,7 +1,9 @@
 -module(bootstrap).
 -export([read_genesis/0, read_genesis/1,
          read_section/2, sections/0, section_subdir/1,
-         default_base/0, ends_with_sx/1]).
+         default_base/0, ends_with_sx/1,
+         build_genesis/1, verify_genesis/2,
+         cidhash_path/1, write_cidhash/2, read_cidhash/1]).
 
 %% Genesis bundle reader per design §12.2.
 %%
@@ -94,3 +96,47 @@ read_one(DirPath, Name) ->
         {ok, Bytes} -> {Name, Bytes};
         {error, R}  -> {Name, {error, R}}
     end.
+
+%% ── Step 4d: bundle CID compute + verify ────────────────────────
+%%
+%% The bundle CID is the canonical content-address of everything in
+%% read_genesis/0's result. We delegate to the host `cid:to_string/1`
+%% BIF (Step 1b substrate): it walks the term via `er-format-value`,
+%% feeds the deterministic textual form into `cid-from-sx`, returns
+%% a CIDv1 (raw codec, sha2-256 multihash) as a binary.
+%%
+%% Design §12.3: at startup the kernel computes this CID and
+%% compares against a hardcoded value (here: a sibling `.cidhash`
+%% file). A mismatch is a hard refuse-to-start.
+
+build_genesis(ReadResult) ->
+    case ReadResult of
+        {ok, Sections} ->
+            Cid = cid:to_string({genesis_bundle, Sections}),
+            {ok, [{cid, Cid}, {sections, Sections}]};
+        Other ->
+            {error, {bad_read_result, Other}}
+    end.
+
+verify_genesis(ReadResult, ExpectedCid) ->
+    case build_genesis(ReadResult) of
+        {ok, [{cid, Cid}, _]} ->
+            case Cid =:= ExpectedCid of
+                true  -> ok;
+                false -> {error, {cid_mismatch, Cid, ExpectedCid}}
+            end;
+        Err -> Err
+    end.
+
+%% Sibling-file CID storage. "/.cidhash" appended to BasePath as
+%% an integer-segment binary (string-literal segments are broken).
+
+%% "/.cidhash" — 47='/' 46='.' c i d h a s h
+cidhash_path(BasePath) ->
+    <<BasePath/binary, 47, 46, 99, 105, 100, 104, 97, 115, 104>>.
+
+write_cidhash(BasePath, Cid) ->
+    file:write_file(cidhash_path(BasePath), Cid).
+
+read_cidhash(BasePath) ->
+    file:read_file(cidhash_path(BasePath)).

next/tests/bootstrap_build.sh

@@ -0,0 +1,127 @@
+#!/usr/bin/env bash
+# next/tests/bootstrap_build.sh — Step 4d acceptance test.
+#
+# Exercises bootstrap:build_genesis/1, verify_genesis/2,
+# cidhash_path/1, write_cidhash/2, read_cidhash/1. The bundle CID
+# is computed by delegating to the host cid:to_string BIF (Step 1b
+# substrate) over the read_genesis result. 11 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
+
+# Clean any stale .cidhash from previous runs before tests touch
+# the filesystem.
+rm -f next/genesis/.cidhash
+
+VERBOSE="${1:-}"
+PASS=0; FAIL=0; ERRORS=""
+TMPFILE=$(mktemp); trap "rm -f $TMPFILE; rm -f next/genesis/.cidhash" EXIT
+
+cat > "$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/bootstrap.erl\")) :name)")
+
+;; build_genesis returns {ok, [{cid, _}, {sections, _}]}
+(epoch 10)
+(eval "(erlang-eval-ast \"{ok, B} = bootstrap:build_genesis(bootstrap:read_genesis()), {Tag, _} = hd(B), Tag\")")
+
+;; The CID is a non-empty binary
+(epoch 11)
+(eval "(get (erlang-eval-ast \"{ok, [{cid, C}, _]} = bootstrap:build_genesis(bootstrap:read_genesis()), is_binary(C)\") :name)")
+(epoch 12)
+(eval "(get (erlang-eval-ast \"{ok, [{cid, C}, _]} = bootstrap:build_genesis(bootstrap:read_genesis()), byte_size(C) > 50\") :name)")
+
+;; build_genesis is deterministic across calls
+(epoch 13)
+(eval "(get (erlang-eval-ast \"{ok, [{cid, C1}, _]} = bootstrap:build_genesis(bootstrap:read_genesis()), {ok, [{cid, C2}, _]} = bootstrap:build_genesis(bootstrap:read_genesis()), C1 =:= C2\") :name)")
+
+;; build_genesis preserves the sections list
+(epoch 14)
+(eval "(erlang-eval-ast \"{ok, [_, {sections, S}]} = bootstrap:build_genesis(bootstrap:read_genesis()), length(S)\")")
+
+;; build_genesis rejects bad input shapes
+(epoch 15)
+(eval "(get (erlang-eval-ast \"case bootstrap:build_genesis({error, broken}) of {error, {bad_read_result, _}} -> ok; _ -> bad end\") :name)")
+
+;; verify_genesis returns ok when CID matches
+(epoch 20)
+(eval "(get (erlang-eval-ast \"{ok, [{cid, C}, _]} = bootstrap:build_genesis(bootstrap:read_genesis()), bootstrap:verify_genesis(bootstrap:read_genesis(), C) =:= ok\") :name)")
+
+;; verify_genesis returns {error, {cid_mismatch, _, _}} when CID doesn't match
+(epoch 21)
+(eval "(get (erlang-eval-ast \"case bootstrap:verify_genesis(bootstrap:read_genesis(), <<99,99,99>>) of {error, {cid_mismatch, _, _}} -> ok; _ -> bad end\") :name)")
+
+;; cidhash_path concatenation
+(epoch 22)
+(eval "(get (erlang-eval-ast \"bootstrap:cidhash_path(<<110,101,120,116>>) =:= <<110,101,120,116,47,46,99,105,100,104,97,115,104>>\") :name)")
+
+;; write_cidhash + read_cidhash round-trip the bundle CID
+(epoch 23)
+(eval "(get (erlang-eval-ast \"{ok, [{cid, C}, _]} = bootstrap:build_genesis(bootstrap:read_genesis()), Base = bootstrap:default_base(), ok = bootstrap:write_cidhash(Base, C), {ok, Stored} = bootstrap:read_cidhash(Base), Stored =:= C\") :name)")
+
+;; Full verify path against the persisted .cidhash
+(epoch 24)
+(eval "(get (erlang-eval-ast \"Base = bootstrap:default_base(), {ok, [{cid, C}, _]} = bootstrap:build_genesis(bootstrap:read_genesis()), ok = bootstrap:write_cidhash(Base, C), {ok, Stored} = bootstrap:read_cidhash(Base), bootstrap:verify_genesis(bootstrap:read_genesis(), Stored) =:= ok\") :name)")
+EPOCHS
+
+OUTPUT=$(timeout 180 "$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: $actual
+"
+  fi
+}
+
+check  2  "module load name"                "bootstrap"
+check 10  "build_genesis head tag"          "cid"
+check 11  "CID is a binary"                 "true"
+check 12  "CID length > 50"                 "true"
+check 13  "build_genesis deterministic"     "true"
+check 14  "sections preserved (7 entries)"  "7"
+check 15  "build_genesis rejects bad shape" "ok"
+check 20  "verify_genesis ok when match"    "true"
+check 21  "verify_genesis errs on mismatch" "ok"
+check 22  "cidhash_path concatenation"      "true"
+check 23  "write/read_cidhash round-trip"   "true"
+check 24  "verify against persisted hash"   "true"
+
+TOTAL=$((PASS+FAIL))
+if [ $FAIL -eq 0 ]; then
+  echo "ok $PASS/$TOTAL next/tests/bootstrap_build.sh passed"
+else
+  echo "FAIL $PASS/$TOTAL passed, $FAIL failed:"
+  echo "$ERRORS"
+fi
+[ $FAIL -eq 0 ]