@cerefox/memory 1.2.0 → 1.2.1

package/dist/bin/cerefox.js

@@ -7438,7 +7438,7 @@ var exports_meta = {};
 __export(exports_meta, {
   PKG_VERSION: () => PKG_VERSION
 });
-var PKG_VERSION = "1.2.0";
+var PKG_VERSION = "1.2.1";
 var init_meta = () => {};
 
 // ../../_shared/config/paths.ts
@@ -75699,7 +75699,7 @@ import { homedir as homedir6 } from "node:os";
 import { join as join9 } from "node:path";
 
 // ../../_shared/ef-meta/index.ts
-var EF_VERSION = "1.2.0";
+var EF_VERSION = "1.2.1";
 var EF_LAST_CHANGED = "1.2.0";
 
 // src/cli/util/checks.ts
@@ -76927,6 +76927,7 @@ class IngestionPipeline {
       text,
       title,
       source = "paste",
+      sourceLabel,
       sourcePath: sourcePathOpt,
       projectName,
       projectId,
@@ -76957,6 +76958,7 @@ class IngestionPipeline {
         text,
         title,
         source,
+        sourceLabel,
         projectIds: fullSetResolved,
         metadata,
         author,
@@ -76995,6 +76997,7 @@ class IngestionPipeline {
           text,
           title,
           source,
+          sourceLabel,
           projectIds: fullSetResolved,
           metadata,
           author,
@@ -77082,6 +77085,7 @@ class IngestionPipeline {
       text,
       title,
       source = "manual",
+      sourceLabel,
       projectId,
       projectIds,
       metadata,
@@ -77198,7 +77202,7 @@ class IngestionPipeline {
       contentFormat: CONTENT_FORMAT_BLIND_STITCH,
       author,
       authorType,
-      sourceLabel: source,
+      sourceLabel: sourceLabel ?? source,
       expectedContentHash: expectedContentHash ?? (forceRechunk && contentUnchanged ? existing.content_hash : null),
       lastWriteWins
     });
@@ -78375,7 +78379,8 @@ async function action28(options) {
         text: doc2.full_content,
         title: doc2.doc_title,
         documentId: id,
-        source: "migrate-format",
+        source: doc2.doc_source,
+        sourceLabel: "migrate-format",
         author,
         authorType,
         expectedContentHash: doc2.content_hash,

package/dist/server-assets/_shared/ef-meta/index.ts

@@ -18,7 +18,7 @@
  * doesn't touch `supabase/functions/` leaves it alone).
  */
 
-export const EF_VERSION = "1.2.0";
+export const EF_VERSION = "1.2.1";
 
 /**
  * The most recent version whose EF-side SOURCE actually changed (#127).

package/dist/server-assets/db/migrations/0020_ingest_preserves_source.sql

@@ -0,0 +1,59 @@
+-- 0020_ingest_preserves_source.sql — stop content updates from silently
+-- overwriting a document's provenance (#191, reported by @tdebasis).
+--
+-- `cerefox_ingest_document`'s UPDATE branch assigned the source column
+-- unconditionally:
+--
+--     source      = p_source                              -- unconditional
+--     source_path = COALESCE(p_source_path, source_path)  -- preserved
+--     metadata    = COALESCE(p_metadata, metadata)        -- preserved
+--
+-- with `p_source TEXT DEFAULT 'agent'` in the signature. The two columns either
+-- side of it already implement the "absent means keep" rule — metadata got it in
+-- v0.11.1, after content updates without metadata were found to be wiping tags.
+-- source was left out of that fix.
+--
+-- Two consequences, both silent:
+--
+--   * Any caller that updates a document without passing p_source rewrites that
+--     document's source to the parameter default 'agent'. Nothing in the RPC's
+--     output, the audit entry, or the version row records that it happened: the
+--     audit operation is 'update-content', and version rows carry their own
+--     source label rather than the document's prior value.
+--   * `cerefox server migrate-format` hit this at corpus scale. It hardcoded
+--     source: "migrate-format" for every document it converted, even though it
+--     reads each document first and cerefox_get_document returns doc_source. A
+--     format conversion is not a change of origin, so every converted document
+--     lost the label it came in with.
+--
+-- Reported impact on one store: 1,317 documents rewritten to 'migrate-format' in
+-- a single run, 201 of which carried no metadata.source_agent and so had no
+-- other provenance field to fall back on. A second store on the same instance
+-- independently reported 509 of 553. Recovery required a point-in-time dump from
+-- before the run.
+--
+-- Fix: p_source defaults to NULL and the UPDATE branch coalesces, matching
+-- metadata and source_path exactly. The CREATE path keeps 'agent' as its
+-- concrete fallback via COALESCE(p_source, 'agent'), so new documents are
+-- unchanged. An explicit value still relabels, so deliberate callers are
+-- unaffected.
+--
+-- Note the distinction this preserves: p_source is the document's origin, while
+-- p_source_label records how a particular write was triggered and is stored on
+-- the version row. migrate-format now passes the document's own source for the
+-- former and keeps "migrate-format" for the latter, so the version history still
+-- shows which run performed the conversion.
+--
+-- Lives in rpcs.sql, which `cerefox server deploy` re-applies. This migration
+-- exists so the schema version moves and operators are told to redeploy.
+--
+-- Idempotent: safe to re-run.
+
+DO $$
+BEGIN
+    RAISE NOTICE
+        'Migration 0020: cerefox_ingest_document now preserves a document''s '
+        'source when p_source is omitted. Before this, any content update '
+        'without an explicit source silently rewrote provenance to ''agent'', '
+        'and migrate-format relabelled every document it converted (#191).';
+END $$;

package/dist/server-assets/db/rpcs.sql

@@ -1258,7 +1258,12 @@ $$;
 --
 -- Parameters:
 --   p_document_id     : NULL for create, UUID for update
---   p_title, p_source, p_source_path, p_content_hash : document fields
+--   p_title, p_source_path, p_content_hash : document fields
+--   p_source          : origin label. NULL = "not provided" → create uses
+--                       'agent', update keeps the existing source (#191). Pass a
+--                       value explicitly to relabel. Distinct from
+--                       p_source_label, which records how THIS write was
+--                       triggered and is stored on the version row.
 --   p_metadata        : JSONB metadata. NULL = "not provided" → create uses '{}',
 --                       update keeps the existing metadata (v0.11.1). Pass '{}'
 --                       explicitly to clear all metadata.
@@ -1291,7 +1296,10 @@ DROP FUNCTION IF EXISTS cerefox_ingest_document(UUID, TEXT, TEXT, TEXT, TEXT, JS
 CREATE FUNCTION cerefox_ingest_document(
     p_document_id       UUID        DEFAULT NULL,
     p_title             TEXT        DEFAULT 'Untitled',
-    p_source            TEXT        DEFAULT 'agent',
+    -- NULL = "not provided": create uses 'agent', update KEEPS the existing
+    -- source (#191 — a content update without a source used to overwrite the
+    -- document's provenance). Pass a value to relabel deliberately.
+    p_source            TEXT        DEFAULT NULL,
     p_source_path       TEXT        DEFAULT NULL,
     p_content_hash      TEXT        DEFAULT '',
     -- NULL = "not provided": create uses '{}', update KEEPS existing metadata
@@ -1445,9 +1453,11 @@ BEGIN
 
         -- Update document record. metadata: NULL = keep existing (v0.11.1 —
         -- a content update without metadata must not wipe the document's tags).
+        -- source: same rule, same reason (#191 — a content update without a
+        -- source must not wipe the document's provenance).
         UPDATE cerefox_documents SET
             title = p_title,
-            source = p_source,
+            source = COALESCE(p_source, source),
             source_path = COALESCE(p_source_path, source_path),
             content_hash = p_content_hash,
             metadata = COALESCE(p_metadata, metadata),
@@ -1465,7 +1475,7 @@ BEGIN
             title, source, source_path, content_hash, metadata,
             chunk_count, total_chars, review_status
         ) VALUES (
-            p_title, p_source, p_source_path, p_content_hash, COALESCE(p_metadata, '{}'::JSONB),
+            p_title, COALESCE(p_source, 'agent'), p_source_path, p_content_hash, COALESCE(p_metadata, '{}'::JSONB),
             v_chunk_count, v_total_chars, v_status
         )
         RETURNING id INTO v_doc_id;
@@ -2333,7 +2343,7 @@ SET search_path = public, pg_catalog
 AS $$
     -- Keep in lockstep with the `@version:` marker in schema.sql (cut_release.ts
     -- enforces it). Bump whenever schema.sql OR rpcs.sql changes.
-    SELECT '0.10.5'::TEXT;
+    SELECT '0.10.6'::TEXT;
 $$;
 
 -- ── cerefox_content_format_stats ─────────────────────────────────────────────

package/dist/server-assets/db/schema.sql

@@ -5,7 +5,7 @@
 -- Requires extensions: vector (pgvector), uuid-ossp
 -- These are enabled at the top of db_deploy.py before this file is applied.
 --
--- @version: 0.10.5
+-- @version: 0.10.6
 -- The `@version` marker above is read by the schema-version-mismatch banner
 -- (see /api/v1/schema-version). Bump it whenever schema.sql OR rpcs.sql
 -- changes in a way that requires `cerefox server deploy` to be re-run —

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@cerefox/memory",
-  "version": "1.2.0",
+  "version": "1.2.1",
   "description": "Cerefox — user-owned shared memory for AI agents. CLI + stdio MCP server + web UI + ingestion for a knowledge base on your own Supabase project (or fully self-hosted with Cerefox Local).",
   "license": "Apache-2.0",
   "homepage": "https://github.com/fstamatelopoulos/cerefox",