fed-sx-m1: Step 4c — bootstrap:read_genesis/0,1 + 5 helpers + 15 read tests

next/kernel/bootstrap.erl

@@ -0,0 +1,96 @@
+-module(bootstrap).
+-export([read_genesis/0, read_genesis/1,
+         read_section/2, sections/0, section_subdir/1,
+         default_base/0, ends_with_sx/1]).
+
+%% Genesis bundle reader per design §12.2.
+%%
+%% read_genesis/0,1 walks the seven canonical section subdirectories
+%% under `next/genesis/`, filters .sx files, reads each file into a
+%% binary, and returns a structured snapshot:
+%%
+%%   {ok, [{Section :: atom,
+%%          [{FileName :: binary, FileBytes :: binary}, ...]},
+%%         ...]}
+%%
+%% Step 4d will compute the bundle CID by hashing the assembled
+%% byte string across all entries; Step 4e will register the parsed
+%% definitions in the kernel registry.
+%%
+%% Port note: this module does NOT parse the .sx contents. The
+%% Erlang-on-SX port has no in-Erlang path from binary bytes to SX
+%% structured terms (same substrate gap that parked Step 3b); the
+%% bundle CID needs only the raw bytes, and registry registration
+%% will happen via an SX-side helper that the kernel hands the
+%% binary contents to. read_genesis/1 ignores its arg in v1 except
+%% to swap the BasePath — `default_base/0` is "next/genesis".
+%%
+%% Port note 2: string-literal binary segments `<<"abc">>` truncate
+%% to one byte in this port, so all path constants are hand-spelled
+%% as integer-segment binaries.
+
+%% ── Public API ──────────────────────────────────────────────────
+
+%% "next/genesis"
+default_base() ->
+    <<110,101,120,116,47,103,101,110,101,115,105,115>>.
+
+read_genesis() ->
+    read_genesis(default_base()).
+
+read_genesis(BasePath) ->
+    {ok, lists:map(
+           fun (S) -> {S, read_section(BasePath, S)} end,
+           sections())}.
+
+sections() ->
+    [activity_types, object_types, projections,
+     validators, codecs, sig_suites, audience].
+
+%% "activity-types"
+section_subdir(activity_types) ->
+    <<97,99,116,105,118,105,116,121,45,116,121,112,101,115>>;
+%% "object-types"
+section_subdir(object_types) ->
+    <<111,98,106,101,99,116,45,116,121,112,101,115>>;
+%% "projections"
+section_subdir(projections) ->
+    <<112,114,111,106,101,99,116,105,111,110,115>>;
+%% "validators"
+section_subdir(validators) ->
+    <<118,97,108,105,100,97,116,111,114,115>>;
+%% "codecs"
+section_subdir(codecs) ->
+    <<99,111,100,101,99,115>>;
+%% "sig-suites"
+section_subdir(sig_suites) ->
+    <<115,105,103,45,115,117,105,116,101,115>>;
+%% "audience"
+section_subdir(audience) ->
+    <<97,117,100,105,101,110,99,101>>.
+
+read_section(BasePath, Section) ->
+    SubDir = section_subdir(Section),
+    %% 47 = '/'
+    Path = <<BasePath/binary, 47, SubDir/binary>>,
+    case file:list_dir(Path) of
+        {ok, Names} ->
+            SxNames = lists:filter(fun (N) -> ends_with_sx(N) end, Names),
+            lists:map(fun (Name) -> read_one(Path, Name) end, SxNames);
+        {error, _} ->
+            []
+    end.
+
+%% Suffix check on the .sx extension. 46='.' 115='s' 120='x'.
+ends_with_sx(<<46, 115, 120>>) -> true;
+ends_with_sx(<<>>) -> false;
+ends_with_sx(<<_, Rest/binary>>) -> ends_with_sx(Rest).
+
+%% ── Internal ────────────────────────────────────────────────────
+
+read_one(DirPath, Name) ->
+    Full = <<DirPath/binary, 47, Name/binary>>,
+    case file:read_file(Full) of
+        {ok, Bytes} -> {Name, Bytes};
+        {error, R}  -> {Name, {error, R}}
+    end.

next/tests/bootstrap_read.sh

@@ -0,0 +1,123 @@
+#!/usr/bin/env bash
+# next/tests/bootstrap_read.sh — Step 4c acceptance test.
+#
+# Exercises bootstrap:read_genesis/0, read_section/2, sections/0,
+# section_subdir/1, ends_with_sx/1. Verifies per-section file
+# counts match the manifest authored in Steps 4a/4b. 14 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/bootstrap.erl\")) :name)")
+
+;; sections/0 returns 7 atoms
+(epoch 10)
+(eval "(erlang-eval-ast \"length(bootstrap:sections())\")")
+
+;; ends_with_sx — positive on "create.sx", negative on "hello"
+(epoch 11)
+(eval "(get (erlang-eval-ast \"bootstrap:ends_with_sx(<<99,114,101,97,116,101,46,115,120>>)\") :name)")
+(epoch 12)
+(eval "(get (erlang-eval-ast \"bootstrap:ends_with_sx(<<104,101,108,108,111>>)\") :name)")
+(epoch 13)
+(eval "(get (erlang-eval-ast \"bootstrap:ends_with_sx(<<>>)\") :name)")
+
+;; Per-section file counts match the manifest (3/10/7/3/3/2/3)
+(epoch 20)
+(eval "(erlang-eval-ast \"length(bootstrap:read_section(bootstrap:default_base(), activity_types))\")")
+(epoch 21)
+(eval "(erlang-eval-ast \"length(bootstrap:read_section(bootstrap:default_base(), object_types))\")")
+(epoch 22)
+(eval "(erlang-eval-ast \"length(bootstrap:read_section(bootstrap:default_base(), projections))\")")
+(epoch 23)
+(eval "(erlang-eval-ast \"length(bootstrap:read_section(bootstrap:default_base(), validators))\")")
+(epoch 24)
+(eval "(erlang-eval-ast \"length(bootstrap:read_section(bootstrap:default_base(), codecs))\")")
+(epoch 25)
+(eval "(erlang-eval-ast \"length(bootstrap:read_section(bootstrap:default_base(), sig_suites))\")")
+(epoch 26)
+(eval "(erlang-eval-ast \"length(bootstrap:read_section(bootstrap:default_base(), audience))\")")
+
+;; read_genesis/0 returns {ok, [{Section, Entries}, ...]} with 7 entries
+(epoch 30)
+(eval "(erlang-eval-ast \"{ok, G} = bootstrap:read_genesis(), length(G)\")")
+
+;; First entry is {activity_types, [_,_,_]}
+(epoch 31)
+(eval "(get (erlang-eval-ast \"{ok, G} = bootstrap:read_genesis(), {S, Entries} = hd(G), S\") :name)")
+
+;; Each entry has the right number of files
+(epoch 32)
+(eval "(erlang-eval-ast \"{ok, G} = bootstrap:read_genesis(), {_, E} = hd(G), length(E)\")")
+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: $actual
+"
+  fi
+}
+
+check  2  "module load name"                "bootstrap"
+check 10  "sections/0 length"               "7"
+check 11  "ends_with_sx create.sx"          "true"
+check 12  "ends_with_sx hello"              "false"
+check 13  "ends_with_sx empty"              "false"
+check 20  "section activity_types count"    "3"
+check 21  "section object_types count"      "10"
+check 22  "section projections count"       "7"
+check 23  "section validators count"        "3"
+check 24  "section codecs count"            "3"
+check 25  "section sig_suites count"        "2"
+check 26  "section audience count"          "3"
+check 30  "read_genesis returns 7 sections" "7"
+check 31  "first section name"              "activity_types"
+check 32  "first section entry count"       "3"
+
+TOTAL=$((PASS+FAIL))
+if [ $FAIL -eq 0 ]; then
+  echo "ok $PASS/$TOTAL next/tests/bootstrap_read.sh passed"
+else
+  echo "FAIL $PASS/$TOTAL passed, $FAIL failed:"
+  echo "$ERRORS"
+fi
+[ $FAIL -eq 0 ]

plans/fed-sx-milestone-1.md

@@ -250,7 +250,7 @@ replay(LogState, InitAcc, Fun) -> ...
 - [x] **4b-proj** — Projections: activity-log, by-type, by-actor, by-object, actor-state, define-registry, audience-graph — 7 `DefineProjection` files + manifest updated + 9 new parse tests
 - [x] **4b-vld** — Validators: envelope-shape, signature, type-schema — 3 `DefineValidator` files + manifest updated + 5 new parse tests
 - [x] **4b-cod** — Codecs (dag-cbor, raw, dag-json) + sig-suites (rsa-sha256-2018, ed25519-2020) + audience predicates (Public, Followers, Direct) — 8 SX files + manifest fully populated + 14 new parse tests
-- [ ] **4c** — `bootstrap:read_genesis/1` in Erlang: walk the manifest, file-read each referenced .sx, return parsed forms
+- [x] **4c** — `bootstrap:read_genesis/0,1` + `read_section/2` + `sections/0` + `section_subdir/1` + `ends_with_sx/1` in Erlang: walk seven hardcoded section subdirs, filter `.sx` files via byte-pattern suffix match, read each into a binary. Returns `{ok, [{Section, [{Name, Bytes}, ...]}, ...]}`. Skips SX parsing — the substrate has no in-Erlang binary→SX-term path (same gap as Step 3b); bundle CID over raw bytes is enough for Step 4d. `next/tests/bootstrap_read.sh` (15 cases).
 - [ ] **4d** — `bootstrap:build_genesis/1` + `bootstrap:verify_genesis/1`: compute bundle CID over the read forms via the host `cid:to_string` substrate; verify against a stored `bundle.cidhash`
 - [ ] **4e** — `bootstrap:load_genesis/1`: register parsed definitions into the in-memory registry (depends on Step 5)
 
@@ -955,6 +955,7 @@ A few things still under-specified; resolve as work begins.
 Newest first. One line per sub-deliverable commit. Erlang conformance gate
 (`bash lib/erlang/conformance.sh`) must remain 729/729 on every entry.
 
+- **2026-05-27** — Step 4c: `next/kernel/bootstrap.erl` — Erlang module that enumerates the genesis bundle by walking seven hardcoded section subdirs via `file:list_dir/1`, filters `.sx` files via byte-pattern suffix match (`ends_with_sx/1`), reads each into a binary via `file:read_file/1`. Returns `{ok, [{Section, [{Name, Bytes}, ...]}]}`. Hits the same SX-parser substrate gap as Step 3b — kept the surface byte-only; parsing happens via SX-side helpers in later steps. Port gotchas: `fun name/arity` references unsupported (use anonymous fun wrappers); `<<"...">>` string-literal segments truncate to one byte (paths hand-spelled as integer-segment binaries). `next/tests/bootstrap_read.sh` 15/15. Erlang conformance 729/729.
 - **2026-05-27** — Step 4b-cod: bootstrap codecs + sig-suites + audience predicates complete. 3 `DefineCodec` files (dag-cbor + raw + dag-json, dag-cbor + dag-json deferring to host-codec primitive when wired), 2 `DefineSigSuite` files (rsa-sha256-2018 PEM-keyed, ed25519-2020 multibase-keyed, both :verify returning false as m2-deferred stand-in), 3 `DefineAudience` files (Public/Followers/Direct member-of predicates per design §16). Manifest now lists 26 bootstrap files across all eight sections; `next/tests/genesis_parse.sh` 50/50. Step 4b complete; remaining Step 4 is bundler code (4c–4e). Erlang conformance 729/729.
 - **2026-05-27** — Step 4b-vld: bootstrap validators complete — 3 `DefineValidator` SX files (envelope-shape mirroring Step 2a, signature stub delegating to envelope:verify_signature/2 per design §9.6, type-schema looking up the object-type schema from define-registry). Manifest `:validators` populated; `next/tests/genesis_parse.sh` 36/36. Erlang conformance 729/729.
 - **2026-05-27** — Step 4b-proj: bootstrap projections complete — 7 `DefineProjection` SX files authored (activity-log identity, by-type/by-actor/by-object indexes, actor-state with key history fold, define-registry meta-fold over Create{Define*}, audience-graph stub). Manifest `:projections` populated; `next/tests/genesis_parse.sh` 31/31. Erlang conformance 729/729.