From ea8cf18f61aac7709bac87ccd790248989cb805e Mon Sep 17 00:00:00 2001
From: Marco Beretta <81851188+berry-13@users.noreply.github.com>
Date: Wed, 29 Jul 2026 03:03:24 +0200
Subject: [PATCH] fix(import): report manifest shards the archive is missing

A shard listed in the manifest but absent from the zip was filtered out
silently, so the surviving shards were treated as the whole export:
inspection undercounted it and the job reported success having skipped
every conversation in the missing file.
---
 packages/api/src/import/manifest.spec.ts | 30 +++++++++++++++++++
 packages/api/src/import/manifest.ts      | 37 ++++++++++++++++++++----
 packages/api/src/import/service.ts       |  7 +++++
 3 files changed, 68 insertions(+), 6 deletions(-)

diff --git a/packages/api/src/import/manifest.spec.ts b/packages/api/src/import/manifest.spec.ts
index 48bf711792..1d5be186dd 100644
--- a/packages/api/src/import/manifest.spec.ts
+++ b/packages/api/src/import/manifest.spec.ts
@@ -255,3 +255,33 @@ describe('detectExportFormat', () => {
     expect(detectExportFormat('nope')).toBeNull();
   });
 });
+
+describe('resolveLayout missing shards', () => {
+  it('reports a manifest-listed shard the archive does not contain', () => {
+    const manifest = parseManifest(
+      Buffer.from(
+        JSON.stringify({
+          version: 1,
+          logical_files: {
+            'conversations.json': {
+              files: ['conversations-000.json', 'conversations-001.json'],
+              sharded: true,
+            },
+          },
+        }),
+      ),
+    );
+
+    const layout = resolveLayout([{ name: 'conversations-000.json', bytes: 2 }], manifest);
+
+    expect(layout.conversationShards).toEqual(['conversations-000.json']);
+    expect(layout.missingShards).toEqual(['conversations-001.json']);
+  });
+
+  it('reports nothing missing when the filename fallback is used', () => {
+    const layout = resolveLayout([{ name: 'conversations.json', bytes: 2 }], null);
+
+    expect(layout.conversationShards).toEqual(['conversations.json']);
+    expect(layout.missingShards).toEqual([]);
+  });
+});
diff --git a/packages/api/src/import/manifest.ts b/packages/api/src/import/manifest.ts
index 59710dae42..e7258e1623 100644
--- a/packages/api/src/import/manifest.ts
+++ b/packages/api/src/import/manifest.ts
@@ -21,6 +21,11 @@ export interface ExportManifest {
 export interface ExportLayout {
   version: number | null;
   conversationShards: string[];
+  /** Shards the manifest lists that the archive does not contain. Dropping
+   * them silently made a truncated export look complete: the remaining shards
+   * were treated as the whole thing, so inspection undercounted and the job
+   * reported success having skipped the missing conversations. */
+  missingShards: string[];
   assetNames: string | null;
   assetEntries: ArchiveEntry[];
 }
@@ -58,12 +63,25 @@ export function parseManifest(buffer: Buffer): ExportManifest | null {
   }
 }
 
-function shardsFromManifest(manifest: ExportManifest, present: Set<string>): string[] {
+function shardsFromManifest(
+  manifest: ExportManifest,
+  present: Set<string>,
+): { found: string[]; missing: string[] } {
   const logical = manifest.logical_files[CONVERSATIONS_LOGICAL];
   if (!logical?.files || !Array.isArray(logical.files) || logical.files.length === 0) {
-    return [];
+    return { found: [], missing: [] };
   }
-  return logical.files.filter((name) => present.has(name));
+
+  const found: string[] = [];
+  const missing: string[] = [];
+  for (const name of logical.files) {
+    if (present.has(name)) {
+      found.push(name);
+      continue;
+    }
+    missing.push(name);
+  }
+  return { found, missing };
 }
 
 function grokShards(entries: ArchiveEntry[]): string[] {
@@ -187,13 +205,20 @@ export function resolveLayout(
 ): ExportLayout {
   const present = new Set(entries.map((entry) => entry.name));
 
-  const fromManifest = manifest ? shardsFromManifest(manifest, present) : [];
-  const conversationShards =
-    fromManifest.length > 0 ? fromManifest : shardsFromFilenames(entries, present);
+  const fromManifest = manifest
+    ? shardsFromManifest(manifest, present)
+    : { found: [], missing: [] };
+  const useManifest = fromManifest.found.length > 0;
+  const conversationShards = useManifest
+    ? fromManifest.found
+    : shardsFromFilenames(entries, present);
 
   return {
     version: manifest?.version ?? null,
     conversationShards,
+    /** Only meaningful when the manifest was actually used: the filename
+     * fallback has no declared list to be missing from. */
+    missingShards: useManifest ? fromManifest.missing : [],
     assetNames: present.has(ASSET_NAMES_ENTRY) ? ASSET_NAMES_ENTRY : null,
     assetEntries: entries.filter((entry) => entry.name.endsWith('.dat')),
   };
diff --git a/packages/api/src/import/service.ts b/packages/api/src/import/service.ts
index 89f7bb9e82..b0d06a4091 100644
--- a/packages/api/src/import/service.ts
+++ b/packages/api/src/import/service.ts
@@ -223,6 +223,13 @@ export async function runImport(input: RunImportInput): Promise<ImportReport> {
     const manifest = hasManifest ? parseManifest(await archive.read(MANIFEST_ENTRY)) : null;
     const layout = resolveLayout(archive.entries, manifest);
 
+    /** A shard the manifest declares but the archive omits is missing data, not
+     * an absent feature: without this the run imports what survived and reports
+     * success, and the user never learns which conversations were skipped. */
+    for (const shard of layout.missingShards) {
+      report.errors.push(`${shard}: listed in the manifest but missing from the archive`);
+    }
+
     const providerRun: ProviderImportContext = {
       archive,
       shards: layout.conversationShards,