deploy/n8n: Improve robustness of diff base resolution for ingestion

deploy/nexus/genome-raw-commit.sh

@@ -87,13 +87,32 @@ for key in "${!G_FILES[@]}"; do
   summary="${summary}${summary:+; }${G_NAME[$key]}:${short}"
 done
 
-# Pull in any remote advances (e.g. a merged wiki PR), then push
+# Push to origin/<base>. The vault is SCRATCH, so we never do an interactive rebase
+# (which can conflict when the same raw file is edited repeatedly). Strategy:
+#   try a fast-forward push; if origin moved, re-apply our raw changes on top of a
+#   fresh origin/<base> and push again. Deterministic, conflict-free.
 git fetch -q origin
-if git show-ref --verify --quiet "refs/remotes/origin/${GENOME_BASE}"; then
-  git rebase -q "origin/${GENOME_BASE}" \
-    || { git rebase --abort 2>/dev/null || true; printf '{"status":"error","reason":"rebase-conflict","genome":"%s"}\n' "$genome"; exit 1; }
+if ! git push -q "$clone_url" "HEAD:${GENOME_BASE}" 2>/dev/null; then
+  # origin advanced: capture our just-made tree for raw/, realign hard, re-apply, retry once.
+  tmp="$(mktemp -d)"
+  cp -a raw/. "$tmp"/ 2>/dev/null || true
+  git reset -q --hard "origin/${GENOME_BASE}"
+  git clean -q -fd
+  cp -a "$tmp"/. raw/ 2>/dev/null || true
+  rm -rf "$tmp"
+  git add -A -- raw/
+  git reset -q -- raw/.stignore raw/.stfolder 2>/dev/null || true
+  if git diff --cached --quiet; then
+    # our content already matches origin -> nothing to push, report ok-noop-after-realign
+    printf '{"status":"ok","genome":"%s","base":"%s","commits":0,"head":"%s","summary":"already in sync after realign","files":[]}\n' \
+      "$genome" "$GENOME_BASE" "$(git rev-parse --short HEAD)"
+    exit 0
+  fi
+  git commit -q --author="${DEFAULT_AUTHOR_NAME} <${DEFAULT_AUTHOR_EMAIL}>" \
+    -m "raw(${genome}): re-apply after realign" -- raw/ || true
+  git push -q "$clone_url" "HEAD:${GENOME_BASE}" \
+    || { printf '{"status":"error","reason":"push-failed-after-realign","genome":"%s"}\n' "$genome"; exit 1; }
 fi
-git push -q "$clone_url" "HEAD:${GENOME_BASE}"
 head="$(git rev-parse --short HEAD)"
 
 # `files` array: local (file://) and remote (Forgejo web) link for each committed raw

deploy/vm101/n8n-pi-wrap

@@ -72,9 +72,23 @@ case "$cmd" in
    set -a; . "${HOME}/.config/knowledge-genome.env"; set +a
    cd "${GENOMES_ROOT}/${genome}" 2>/dev/null || { echo '{"status":"error","reason":"unknown genome"}'; exit 1; }
    git fetch -q origin
-   # changed/added raw files (exclude deletions: nothing to ingest), only under raw/, skip Syncthing infra
-   files="$(git diff --name-only --diff-filter=d "${before}" "${after}" -- raw/ 2>/dev/null \
-            | grep -vE '(^|/)\.st(folder|ignore)' || true)"
+   # Resolve the diff base robustly:
+   #   - before all-zero (brand-new branch) or unreachable (force-push) -> fall back to after~1
+   #   - if even after~1 is missing (root commit) -> list all raw files in `after`
+   base="$before"
+   case "$before" in *[!0]*) : ;; *) base="" ;; esac          # all-zero -> empty
+   if [ -n "$base" ] && ! git cat-file -e "${base}^{commit}" 2>/dev/null; then base=""; fi
+   if [ -z "$base" ]; then
+     if git cat-file -e "${after}~1^{commit}" 2>/dev/null; then base="${after}~1"; else base=""; fi
+   fi
+   if [ -n "$base" ]; then
+     files="$(git diff --name-only --diff-filter=d "${base}" "${after}" -- raw/ 2>/dev/null \
+              | grep -vE '(^|/)\.st(folder|ignore)' || true)"
+   else
+     # no usable base: enumerate raw files present at `after`
+     files="$(git ls-tree -r --name-only "${after}" -- raw/ 2>/dev/null \
+              | grep -vE '(^|/)\.st(folder|ignore)' || true)"
+   fi
    # emit a JSON array via jq (safe escaping)
    printf '%s\n' "$files" | grep -c . >/dev/null 2>&1 || files=""
    if [ -z "$files" ]; then