feat(ingest): Recapture mode — re-grab post bodies/links + localize on-disk inline images (#830)

A plain backfill gates post-body capture on the seen-ledger, so a post whose
media is already on disk AND whose post key is already seen never gets its body
recaptured (operator-flagged: Industrial Lust description missing). Recovery
recaptures unconditionally but re-downloads the whole source.

New 'recapture' walk mode (4th beside tick/backfill/recovery): bypasses the
post-record gate so EVERY post's body + external links are re-captured
(detail-fetching empty bodies) WITHOUT re-downloading on-disk media; and
surfaces already-present media via a separate non-deleting relink channel so the
importer backfills ImageRecord.source_filehash for inline-image localization.

- ingest_core: recapture mode + recapture_records gate bypass + relink collect
- patreon_downloader: recapture surfaces seen-on-disk as skipped_disk(path),
  never refetches seen-missing media, still downloads genuinely-new
- importer.relink_source_filehash: NULL-only sha256 backfill, never unlinks
- download_service: mode derivation + phase-3 relink loop + lifecycle clear
- source_service/api: start_recapture + backfill_recapture field + action
- frontend: Recapture kebab action + 'Recapturing' badge across SourceActions/
  Row/Card/SubscriptionsTab + sources store
- tests across ingester/downloader/importer/source_service/api/download_service

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

backend/app/api/sources.py

@@ -138,14 +138,16 @@ async def delete_source(source_id: int):
 
 @sources_bp.route("/<int:source_id>/backfill", methods=["POST"])
 async def set_backfill(source_id: int):
-    """Plan #693/#697: start/stop a run-until-done backfill, or start a recovery.
-    Body: `{"action": "start" | "stop" | "recover"}` (default "start"). 'start'
-    walks the full post history in time-boxed chunks until it reaches the bottom
-    (then the source shows 'complete'); 'recover' is the same walk but bypasses
-    the Patreon seen-ledger to re-fetch dropped-and-deleted near-dups under the
-    current pHash threshold; 'stop' cancels either back to tick mode. Returns the
-    updated source dict (incl. backfill_state / backfill_chunks /
-    backfill_bypass_seen)."""
+    """Plan #693/#697 + #830: start/stop a backfill, or start a recovery /
+    recapture. Body: `{"action": "start" | "stop" | "recover" | "recapture"}`
+    (default "start"). 'start' walks the full post history in time-boxed chunks
+    until it reaches the bottom (then the source shows 'complete'); 'recover' is
+    the same walk but bypasses the Patreon seen-ledger to re-fetch
+    dropped-and-deleted near-dups under the current pHash threshold; 'recapture'
+    re-grabs EVERY post's body + external links and localizes on-disk inline
+    images WITHOUT re-downloading media; 'stop' cancels any back to tick mode.
+    Returns the updated source dict (incl. backfill_state / backfill_chunks /
+    backfill_bypass_seen / backfill_recapture)."""
     from pathlib import Path
 
     from ..services.credential_service import CredentialService
@@ -157,10 +159,10 @@ async def set_backfill(source_id: int):
 
     payload = await request.get_json(silent=True) or {}
     action = payload.get("action", "start")
-    if action not in ("start", "stop", "recover"):
+    if action not in ("start", "stop", "recover", "recapture"):
         return _bad(
             "invalid_action",
-            detail="action must be 'start', 'stop', or 'recover'",
+            detail="action must be 'start', 'stop', 'recover', or 'recapture'",
         )
 
     # Pre-flight (plan #703 #2): before arming a deep walk on a native-ingester
@@ -170,7 +172,7 @@ async def set_backfill(source_id: int):
     # platforms: gallery-dl verify is a slow --simulate subprocess, too heavy for
     # an arm action. The credential read happens in a session that's CLOSED
     # before the verify network call (don't hold a DB conn across the request).
-    if action in ("start", "recover"):
+    if action in ("start", "recover", "recapture"):
         async with get_session() as session:
             rec = await SourceService(session).get(source_id)
             if rec is None:
@@ -200,6 +202,8 @@ async def set_backfill(source_id: int):
                 record = await svc.start_backfill(source_id)
             elif action == "recover":
                 record = await svc.start_recovery(source_id)
+            elif action == "recapture":
+                record = await svc.start_recapture(source_id)
             else:
                 record = await svc.stop_backfill(source_id)
         except LookupError:

backend/app/services/download_service.py

@@ -111,6 +111,10 @@ class DownloadService:
         # `_backfill_bypass_seen` flag rides alongside the running backfill state;
         # download mode is "recovery" when both are set.
         bypass_seen = bool(overrides.get("_backfill_bypass_seen"))
+        # #830 recapture: re-grab post bodies/links + localize on-disk inline
+        # images, WITHOUT re-downloading media. Rides alongside the backfill
+        # state like _backfill_bypass_seen; mutually exclusive with it.
+        recapture = bool(overrides.get("_backfill_recapture"))
         if in_backfill:
             skip_value: bool | str = BACKFILL_SKIP_VALUE
             source_config.timeout = BACKFILL_CHUNK_SECONDS
@@ -130,6 +134,8 @@ class DownloadService:
         if uses_native_ingester(ctx["platform"]):
             if in_backfill and bypass_seen:
                 mode = "recovery"
+            elif in_backfill and recapture:
+                mode = "recapture"
             elif in_backfill:
                 mode = "backfill"
             else:
@@ -396,6 +402,20 @@ class DownloadService:
 
             await loop.run_in_executor(None, _upsert)
 
+        # #830 recapture: backfill source_filehash on EXISTING on-disk images so
+        # their post-body inline <img src=CDN> remaps to the local copy. A
+        # SEPARATE non-deleting channel (NOT the import list — that would unlink
+        # the duplicate file). Empty outside recapture mode.
+        for rel_str, rel_url in getattr(dl_result, "relink_source_paths", None) or []:
+            rel_path = Path(rel_str)
+            if not rel_path.exists():  # noqa: ASYNC240
+                continue
+
+            def _relink(p=rel_path, u=rel_url):
+                return self.importer.relink_source_filehash(p, u, artist=artist)
+
+            await loop.run_in_executor(None, _relink)
+
         # Kick the off-platform file-host downloader for any links this run
         # recorded (mega/gdrive/…). Global + idempotent (only claims pending/
         # retryable rows); the beat sweep is the backstop. Lazy import dodges a
@@ -508,6 +528,8 @@ class DownloadService:
             # plan #697: a recovery walk shares this lifecycle; clear its bypass
             # flag on completion so the next routine tick honors the seen-ledger.
             new_overrides.pop("_backfill_bypass_seen", None)
+            # #830: same for the recapture flag.
+            new_overrides.pop("_backfill_recapture", None)
             src.config_overrides = new_overrides
             src.backfill_runs_remaining = 0
             return

backend/app/services/gallery_dl.py

@@ -153,6 +153,11 @@ class DownloadResult:
     # Importer.upsert_post_record (keyed on external_post_id → updates, never
     # doubles, the same Post a media import would create).
     post_record_paths: list[str] = field(default_factory=list)
+    # Native ingester, recapture mode only (#830): (on-disk path, CDN source_url)
+    # pairs for already-present media whose ImageRecord.source_filehash should be
+    # backfilled (inline-image localization) WITHOUT re-download or unlink. Empty
+    # on the gallery-dl path and outside recapture.
+    relink_source_paths: list[tuple[str, str]] = field(default_factory=list)
     stdout: str = ""
     stderr: str = ""
     return_code: int = 0

backend/app/services/importer.py

@@ -1005,6 +1005,35 @@ class Importer:
                         return sv.strip()
         return None
 
+    def relink_source_filehash(
+        self, path: Path, source_url: str, *, artist: Artist | None = None,
+    ) -> bool:
+        """#830 recapture: backfill source_url + source_filehash on an EXISTING
+        on-disk image's ImageRecord (matched by sha256) so its post-body inline
+        `<img src=CDN>` remaps to the local copy at render time — WITHOUT
+        re-downloading and WITHOUT unlinking the file (unlike the import path's
+        duplicate-hash branch). NULL-only, mirroring _apply_sidecar: never
+        clobbers an already-set filehash. Returns True only when a row was
+        updated. No-op when the file is gone, the url has no filehash, no record
+        matches the bytes, or the record already carries a filehash.
+        """
+        fh = filehash_from_url(source_url)
+        if not fh:
+            return False
+        try:
+            sha = _sha256_of(path)
+        except OSError:
+            return False
+        record = self.session.execute(
+            select(ImageRecord).where(ImageRecord.sha256 == sha)
+        ).scalar_one_or_none()
+        if record is None or record.source_filehash is not None:
+            return False
+        record.source_url = source_url
+        record.source_filehash = fh
+        self.session.commit()
+        return True
+
     def _apply_sidecar(
         self,
         record: ImageRecord,

backend/app/services/ingest_core.py

@@ -105,18 +105,28 @@ class Ingester:
     ) -> DownloadResult:
         """Walk + download for one source, returning a gallery-dl-shaped result.
 
-        `mode` is "tick" | "backfill" | "recovery". Recovery bypasses the tier-1
-        seen-ledger AND the dead-letter ledger (tier-2 disk still skips kept
-        files). The walk stops on:
+        `mode` is "tick" | "backfill" | "recovery" | "recapture". Recovery
+        bypasses the tier-1 seen-ledger AND the dead-letter ledger (tier-2 disk
+        still skips kept files). Recapture (#830) is the cheap "re-grab post
+        text" walk: it bypasses the post-record gate (so EVERY post's body +
+        external links are re-captured / detail-fetched) but KEEPS the media
+        seen-ledger — on-disk media is NOT re-downloaded, only surfaced so its
+        ImageRecord's source_filehash can be backfilled for inline-image
+        localization. The walk stops on:
           - budget exhaustion (time_budget_seconds)         → TIMEOUT / PARTIAL
           - tick early-out (seen_threshold contiguous seen)  → success
           - reaching the bottom of the feed                  → success (rc 0)
         A client-level failure (drift / auth / network) fails the whole run loud.
         """
         bypass_seen = mode == "recovery"
+        recapture = mode == "recapture"
+        # Both recovery and recapture re-capture EVERY post's body + links — they
+        # bypass the post-record seen-gate (recovery via bypass_seen, recapture
+        # explicitly). A plain backfill stays gated (capture once per post).
+        recapture_records = bypass_seen or recapture
         # Only deep walks checkpoint their cursor mid-flight (plan #705 #6); a
         # tick has no resumable backfill state.
-        checkpoint = mode in ("backfill", "recovery")
+        checkpoint = mode in ("backfill", "recovery", "recapture")
         ledger_key = self._ledger_key
         # Optional seams (Patreon native ingester): capture pure-text posts that
         # have NO downloadable media so the artist archive is complete. Absent on
@@ -129,6 +139,10 @@ class Ingester:
         written: list[str] = []
         post_records: list[str] = []
         quarantined_paths: list[str] = []
+        # #830 recapture: (on-disk path, CDN source_url) pairs for already-present
+        # media, so phase 3 can backfill the ImageRecord's source_filehash WITHOUT
+        # re-downloading or unlinking the file. Empty outside recapture mode.
+        relink: list[tuple[str, str]] = []
         downloaded = 0
         errors = 0
         quarantined = 0
@@ -165,6 +179,7 @@ class Ingester:
                 quarantined_paths=list(quarantined_paths),
                 written_paths=written,
                 post_record_paths=list(post_records),
+                relink_source_paths=list(relink),
                 stdout="\n".join(log_lines),
                 stderr="",
                 return_code=return_code,
@@ -241,7 +256,7 @@ class Ingester:
                     if rk is not None:
                         pkey, ppid = rk
                         already = (
-                            set() if bypass_seen
+                            set() if recapture_records
                             else self._seen_keys(source_id, [pkey])
                         )
                         if pkey not in already:
@@ -275,6 +290,7 @@ class Ingester:
                 outcomes = self.downloader.download_post(
                     post, media, artist_slug, is_seen=_is_skip,
                     should_stop=lambda: time.monotonic() - start >= time_budget_seconds,
+                    recapture=recapture,
                 )
 
                 to_mark: list[tuple[str, str]] = []
@@ -300,6 +316,12 @@ class Ingester:
                         to_clear.append(key)
                         skipped_count += 1
                         consecutive_seen += 1
+                        # #830 recapture: surface (on-disk path, CDN url) so phase
+                        # 3 can backfill source_filehash for inline-image
+                        # localization — a SEPARATE non-deleting channel, never the
+                        # import list (which would unlink the file, per above).
+                        if recapture and outcome.path is not None:
+                            relink.append((str(outcome.path), media_item.url))
                     elif outcome.status == "skipped_seen":
                         skipped_count += 1
                         consecutive_seen += 1

backend/app/services/patreon_downloader.py

@@ -198,6 +198,7 @@ class PatreonDownloader:
         *,
         is_seen: Callable[[object], bool] = lambda m: False,
         should_stop: Callable[[], bool] = lambda: False,
+        recapture: bool = False,
     ) -> list[MediaOutcome]:
         """Download every media item of one post; return per-item outcomes.
 
@@ -212,6 +213,11 @@ class PatreonDownloader:
         re-checks the budget between posts), so we honour the deadline mid-post
         and return the items done so far — the rest re-fetch next chunk (they
         were never marked seen). Bounds chunk overrun to one media download.
+
+        `recapture` (#830): don't re-download already-present media, but DO surface
+        on-disk media as `skipped_disk` (with its path) even when the seen-ledger
+        would tier-1 skip it — so the engine can backfill source_filehash for
+        inline-image localization. Genuinely-missing seen media is NOT refetched.
         """
         post_dir = self.images_root / artist_slug / "patreon" / _post_dir_name(post)
         outcomes: list[MediaOutcome] = []
@@ -221,7 +227,10 @@ class PatreonDownloader:
                 break
             try:
                 outcomes.append(
-                    self._download_one(post, media, post_dir, artist_slug, i, is_seen)
+                    self._download_one(
+                        post, media, post_dir, artist_slug, i, is_seen,
+                        recapture=recapture,
+                    )
                 )
             except Exception as exc:  # resilient: isolate one item's failure
                 log.warning(
@@ -243,9 +252,15 @@ class PatreonDownloader:
         artist_slug: str,
         index: int,
         is_seen: Callable[[object], bool],
+        *,
+        recapture: bool = False,
     ) -> MediaOutcome:
-        # tier-1: seen ledger (injected; no DB here).
-        if is_seen(media):
+        # tier-1: seen ledger (injected; no DB here). In recapture mode we DON'T
+        # short-circuit here — we fall through to the disk check so an on-disk
+        # seen file is surfaced as skipped_disk (with its path) for source_filehash
+        # backfill; a seen file that's NOT on disk is left alone (not refetched).
+        seen = is_seen(media)
+        if seen and not recapture:
             return MediaOutcome(media=media, status="skipped_seen", path=None, error=None)
 
         nn = f"{index:02d}"
@@ -265,6 +280,12 @@ class PatreonDownloader:
                     media=media, status="skipped_disk", path=existing, error=None
                 )
 
+        # recapture: a seen item that isn't on disk is NOT re-downloaded (that's
+        # recovery's job) — recapture only re-grabs post text + localizes existing
+        # files. Returns skipped_seen so the run-of-seen / counts stay consistent.
+        if seen:
+            return MediaOutcome(media=media, status="skipped_seen", path=None, error=None)
+
         post_dir.mkdir(parents=True, exist_ok=True)
 
         # Pace real downloads only (the skips above already returned). plan #703.

backend/app/services/source_service.py

@@ -70,6 +70,10 @@ class SourceRecord:
     # plan #697: a running deep-walk that bypasses the Patreon seen-ledger
     # (recovery) vs. a normal backfill. Lets the badge label it "Recovering".
     backfill_bypass_seen: bool
+    # #830: a running deep-walk in RECAPTURE mode (re-grab post bodies/links +
+    # localize on-disk inline images, no media re-download). Lets the badge label
+    # it "Recapturing".
+    backfill_recapture: bool
     # plan #704: cumulative posts processed across the walk's chunks — live
     # progress for the badge.
     backfill_posts: int
@@ -94,6 +98,7 @@ class SourceRecord:
             "backfill_state": self.backfill_state,
             "backfill_chunks": self.backfill_chunks,
             "backfill_bypass_seen": self.backfill_bypass_seen,
+            "backfill_recapture": self.backfill_recapture,
             "backfill_posts": self.backfill_posts,
         }
 
@@ -171,6 +176,7 @@ class SourceService:
             backfill_state=co.get("_backfill_state"),
             backfill_chunks=int(co.get("_backfill_chunks", 0)),
             backfill_bypass_seen=bool(co.get("_backfill_bypass_seen")),
+            backfill_recapture=bool(co.get("_backfill_recapture")),
             backfill_posts=int(co.get("_backfill_posts", 0)),
         )
 
@@ -359,6 +365,34 @@ class SourceService:
         await self.session.commit()
         return await self._row_to_record(source)
 
+    async def start_recapture(self, source_id: int) -> SourceRecord:
+        """#830: arm a RECAPTURE walk — a backfill that re-grabs EVERY post's body
+        + external links (detail-fetching empty bodies) and localizes already-on-
+        disk inline images, WITHOUT re-downloading media. Reuses the entire #693
+        backfill state machine plus a `_backfill_recapture` flag that flips
+        download mode to recapture. Distinct from recovery (which re-downloads the
+        whole source); the two flags are mutually exclusive, so arming recapture
+        clears bypass_seen. Clears prior cursor/chunk/stall state so it walks
+        fresh from the top. The flag is cleared on completion (download_service)
+        and on stop. Recapture is Patreon-only (the native ingester's post-record
+        capture); inert elsewhere. The UI gates the action to Patreon sources."""
+        source = (await self.session.execute(
+            select(Source).where(Source.id == source_id)
+        )).scalar_one_or_none()
+        if source is None:
+            raise LookupError(f"source id={source_id} not found")
+        co = dict(source.config_overrides or {})
+        co["_backfill_state"] = "running"
+        co["_backfill_recapture"] = True
+        co.pop("_backfill_bypass_seen", None)  # mutually exclusive with recovery
+        for k in ("_backfill_cursor", "_backfill_cursor_stalls", "_backfill_chunks",
+                  "_backfill_posts"):
+            co.pop(k, None)
+        source.config_overrides = co
+        source.backfill_runs_remaining = BACKFILL_MAX_CHUNKS
+        await self.session.commit()
+        return await self._row_to_record(source)
+
     async def stop_backfill(self, source_id: int) -> SourceRecord:
         """Plan #693: cancel an in-progress backfill — back to idle/tick mode.
         Clears the running state + cursor/chunk/stall bookkeeping."""
@@ -369,7 +403,8 @@ class SourceService:
             raise LookupError(f"source id={source_id} not found")
         co = dict(source.config_overrides or {})
         for k in ("_backfill_state", "_backfill_cursor", "_backfill_cursor_stalls",
-                  "_backfill_chunks", "_backfill_bypass_seen", "_backfill_posts"):
+                  "_backfill_chunks", "_backfill_bypass_seen", "_backfill_recapture",
+                  "_backfill_posts"):
             co.pop(k, None)
         source.config_overrides = co
         source.backfill_runs_remaining = 0

frontend/src/components/subscriptions/SourceActions.vue

@@ -21,7 +21,7 @@
       <v-icon>{{ running ? 'mdi-stop' : 'mdi-magnify-scan' }}</v-icon>
       <v-tooltip activator="parent" location="top">
         {{ running
-          ? (recovering ? 'Stop recovery' : 'Stop backfill')
+          ? (recovering ? 'Stop recovery' : (recapturing ? 'Stop recapture' : 'Stop backfill'))
           : 'Backfill — one-time walk of the FULL post history until complete' }}
       </v-tooltip>
     </v-btn>
@@ -38,6 +38,17 @@
           under the current similarity threshold
         </v-list-item-subtitle>
       </v-list-item>
+      <v-list-item
+        v-if="isPatreon && !running"
+        prepend-icon="mdi-text-box-search-outline"
+        @click="emit('recapture', source)"
+      >
+        <v-list-item-title>Recapture post text &amp; links</v-list-item-title>
+        <v-list-item-subtitle>
+          Re-grab every post's body + external links and localize inline images
+          already on disk — without re-downloading media
+        </v-list-item-subtitle>
+      </v-list-item>
       <v-divider v-if="isPatreon && !running" />
       <v-list-item
         base-color="error"
@@ -58,10 +69,11 @@ const props = defineProps({
   source: { type: Object, required: true },
   checking: { type: Boolean, default: false },
 })
-const emit = defineEmits(['check', 'backfill', 'recover', 'remove'])
+const emit = defineEmits(['check', 'backfill', 'recover', 'recapture', 'remove'])
 
 const running = computed(() => props.source.backfill_state === 'running')
 const recovering = computed(() => !!props.source.backfill_bypass_seen)
+const recapturing = computed(() => !!props.source.backfill_recapture)
 const isPatreon = computed(() => props.source.platform === 'patreon')
 </script>
 

frontend/src/components/subscriptions/SourceCard.vue

@@ -40,7 +40,7 @@
       <v-chip
         v-else-if="source.backfill_state === 'running'"
         size="x-small" color="info" variant="tonal" label
-      >{{ source.backfill_bypass_seen ? 'Recovering' : 'Backfilling'
+      >{{ source.backfill_bypass_seen ? 'Recovering' : (source.backfill_recapture ? 'Recapturing' : 'Backfilling')
         }}{{ source.backfill_posts
               ? ` · ${source.backfill_posts} posts`
               : (source.backfill_chunks ? ` (${source.backfill_chunks})` : '') }}</v-chip>
@@ -60,6 +60,7 @@
         @check="$emit('check', $event)"
         @backfill="$emit('backfill', $event)"
         @recover="$emit('recover', $event)"
+        @recapture="$emit('recapture', $event)"
         @remove="$emit('remove', $event)"
       />
     </div>
@@ -79,7 +80,7 @@ const props = defineProps({
   checking: { type: Boolean, default: false },
   warningThreshold: { type: Number, default: 5 },
 })
-const emit = defineEmits(['edit', 'remove', 'toggle', 'check', 'backfill', 'recover'])
+const emit = defineEmits(['edit', 'remove', 'toggle', 'check', 'backfill', 'recover', 'recapture'])
 
 function onToggleEnabled(value) {
   emit('toggle', { source: props.source, enabled: value })

frontend/src/components/subscriptions/SourceRow.vue

@@ -51,7 +51,7 @@
       <v-chip
         v-else-if="source.backfill_state === 'running'"
         size="x-small" color="info" variant="tonal" label
-      >{{ source.backfill_bypass_seen ? 'Recovering' : 'Backfilling'
+      >{{ source.backfill_bypass_seen ? 'Recovering' : (source.backfill_recapture ? 'Recapturing' : 'Backfilling')
         }}{{ source.backfill_posts
               ? ` · ${source.backfill_posts} posts`
               : (source.backfill_chunks ? ` (${source.backfill_chunks})` : '') }}</v-chip>
@@ -71,6 +71,7 @@
         @check="$emit('check', $event)"
         @backfill="$emit('backfill', $event)"
         @recover="$emit('recover', $event)"
+        @recapture="$emit('recapture', $event)"
         @remove="$emit('remove', $event)"
       />
     </td>
@@ -87,7 +88,7 @@ const props = defineProps({
   checking: { type: Boolean, default: false },
   warningThreshold: { type: Number, default: 5 },
 })
-const emit = defineEmits(['edit', 'remove', 'toggle', 'check', 'backfill', 'recover'])
+const emit = defineEmits(['edit', 'remove', 'toggle', 'check', 'backfill', 'recover', 'recapture'])
 
 function onToggleEnabled(value) {
   emit('toggle', { source: props.source, enabled: value })

frontend/src/components/subscriptions/SubscriptionsTab.vue

@@ -138,7 +138,7 @@
               :source="item.singleSource"
               :checking="store.checkingIds.has(item.singleSource.id)"
               @check="onCheck" @backfill="onBackfill"
-              @recover="onRecover" @remove="removeSource"
+              @recover="onRecover" @recapture="onRecapture" @remove="removeSource"
             />
             <v-btn icon size="small" variant="text" @click.stop="openEditSource(item.singleSource)">
               <v-icon>mdi-pencil</v-icon>
@@ -214,6 +214,7 @@
                     @check="onCheck"
                     @backfill="onBackfill"
                     @recover="onRecover"
+                    @recapture="onRecapture"
                   />
                   <tr v-if="item.sources.length === 0">
                     <td colspan="8" class="fc-subs__sources-empty">
@@ -291,6 +292,7 @@
             @check="onCheck"
             @backfill="onBackfill"
             @recover="onRecover"
+            @recapture="onRecapture"
           />
           <div v-if="item.sources.length === 0" class="fc-subs__sources-empty">
             No sources yet. Tap + to add one.
@@ -646,6 +648,22 @@ async function onRecover(source) {
   }
 }
 
+// #830: arm a recapture walk (Patreon-only) — re-grab every post's body +
+// external links and localize on-disk inline images, without re-downloading
+// media. Reuses the backfill lifecycle/badge; stop via the same Stop control.
+async function onRecapture(source) {
+  try {
+    await store.recaptureSource(source.id, source.artist_id)
+    toast({ text: `Recapture started for ${source.artist_name}`, type: 'success' })
+    // recaptureSource patches the source in place — no full reload.
+  } catch (e) {
+    toast({
+      text: `Recapture start failed: ${e?.body?.detail || e?.message || e}`,
+      type: 'error',
+    })
+  }
+}
+
 async function checkAll(group) {
   let ok = 0
   let conflict = 0

frontend/src/stores/sources.js

@@ -131,6 +131,14 @@ export const useSourcesStore = defineStore('sources', () => {
     _patchSource(body)
     return body
   }
+  // #830: arm a recapture walk (Patreon-only) — re-grab every post's body +
+  // external links and localize on-disk inline images, WITHOUT re-downloading
+  // media. Shares the backfill lifecycle/badge; stop via stopBackfill.
+  async function recaptureSource(id, artistIdHint = null) {
+    const body = await api.post(`/api/sources/${id}/backfill`, { body: { action: 'recapture' } })
+    _patchSource(body)
+    return body
+  }
 
   // Plan #708 B4: dry-run preview — count what a backfill WOULD download
   // (native platforms), without downloading. Read-only, so no cache invalidate.
@@ -167,6 +175,7 @@ export const useSourcesStore = defineStore('sources', () => {
     startBackfill,
     stopBackfill,
     recoverSource,
+    recaptureSource,
     previewSource,
     findOrCreateArtist, autocompleteArtist,
     loadScheduleStatus,

tests/test_api_sources.py

@@ -324,6 +324,32 @@ async def test_backfill_endpoint_recover_arms_bypass(client, artist, db):
     assert stopped_body["backfill_bypass_seen"] is False
 
 
+@pytest.mark.asyncio
+async def test_backfill_endpoint_recapture_arms_flag(client, artist, db):
+    """#830: action='recapture' arms a backfill that re-grabs post bodies/links +
+    localizes on-disk inline images; the response exposes backfill_recapture=True
+    (badge 'Recapturing') and NOT backfill_bypass_seen. Stop clears it."""
+    src = Source(
+        artist_id=artist.id, platform="patreon",
+        url="https://patreon.com/alice-recapture", enabled=True,
+    )
+    db.add(src)
+    await db.commit()
+    sid = src.id
+
+    resp = await client.post(f"/api/sources/{sid}/backfill", json={"action": "recapture"})
+    assert resp.status_code == 200
+    body = await resp.get_json()
+    assert body["backfill_state"] == "running"
+    assert body["backfill_recapture"] is True
+    assert body["backfill_bypass_seen"] is False
+
+    stopped = await client.post(f"/api/sources/{sid}/backfill", json={"action": "stop"})
+    stopped_body = await stopped.get_json()
+    assert stopped_body["backfill_state"] is None
+    assert stopped_body["backfill_recapture"] is False
+
+
 # --- Plan #703 #2: pre-flight credential verify on backfill/recover arm -----
 
 

tests/test_download_service.py

@@ -60,8 +60,10 @@ def _make_fake_dl_result(
     *, success=True, written_paths=None, quarantined_paths=None,
     files_downloaded=0, error_type=None, error_message=None,
     stdout="", stderr="", cursor=None, run_stats=None, posts_processed=0,
+    relink_source_paths=None,
 ):
     return SimpleNamespace(
+        relink_source_paths=relink_source_paths or [],
         success=success,
         url="https://patreon.com/alice",
         artist_slug="alice",
@@ -793,6 +795,31 @@ async def test_recovery_mode_selected_and_flag_cleared_on_complete(
     assert "_backfill_bypass_seen" not in co
 
 
+@pytest.mark.asyncio
+async def test_recapture_mode_selected_and_flag_cleared_on_complete(
+    db, db_sync, tmp_path, seed_artist_and_source,
+):
+    """#830: `_backfill_recapture` alongside a running backfill selects recapture
+    mode. A clean rc=0 walk completes the shared lifecycle AND clears the
+    recapture flag so the next tick is a plain tick again."""
+    _artist, source = seed_artist_and_source
+    source.config_overrides = {"_backfill_state": "running", "_backfill_recapture": True}
+    source.backfill_runs_remaining = 5
+    await db.commit()
+
+    svc, calls = _backfill_svc(db, db_sync, tmp_path, _make_fake_dl_result(
+        success=True, written_paths=[], files_downloaded=0,
+    ))
+    await svc.download_source(source.id)
+
+    assert calls[0]["mode"] == "recapture"
+    co = (await db.execute(
+        select(Source.config_overrides).where(Source.id == source.id)
+    )).scalar_one()
+    assert co.get("_backfill_state") == "complete"
+    assert "_backfill_recapture" not in co
+
+
 @pytest.mark.asyncio
 async def test_partial_error_type_maps_to_ok_status(
     db, db_sync, tmp_path, seed_artist_and_source,

tests/test_patreon_downloader.py

@@ -206,6 +206,54 @@ def test_skip_disk(tmp_path):
     assert existing.read_bytes() == b"already here"  # untouched
 
 
+def test_recapture_surfaces_seen_on_disk_without_download(tmp_path):
+    """#830 recapture: a seen item that IS on disk is surfaced as skipped_disk
+    (with its path, for source_filehash backfill) — NOT tier-1 short-circuited,
+    and never re-downloaded."""
+    session = _FakeSession()
+    dl = _downloader(tmp_path, session=session)
+    post_dir = tmp_path / "artist-x" / "patreon" / "2026-05-01_1001_My Post Title"
+    post_dir.mkdir(parents=True)
+    existing = post_dir / "01_media1.png"
+    existing.write_bytes(b"already here")
+
+    outcomes = dl.download_post(
+        _post(), [_img("media1.png")], "artist-x",
+        is_seen=lambda m: True, recapture=True,
+    )
+    assert outcomes[0].status == "skipped_disk"
+    assert outcomes[0].path == existing
+    assert session.calls == []                        # no re-download
+    assert existing.read_bytes() == b"already here"    # untouched
+
+
+def test_recapture_does_not_refetch_seen_missing(tmp_path):
+    """#830 recapture: a seen item that's NOT on disk is left alone (skipped_seen,
+    no download) — refetching is recovery's job, not recapture's."""
+    session = _FakeSession()
+    dl = _downloader(tmp_path, session=session)
+    outcomes = dl.download_post(
+        _post(), [_img("media1.png")], "artist-x",
+        is_seen=lambda m: True, recapture=True,
+    )
+    assert outcomes[0].status == "skipped_seen"
+    assert outcomes[0].path is None
+    assert session.calls == []                        # not refetched
+
+
+def test_recapture_still_downloads_genuinely_new(tmp_path):
+    """#830 recapture: media that is neither seen nor on disk still downloads — a
+    recapture walk is thorough, it just doesn't re-pull existing files."""
+    session = _FakeSession()
+    dl = _downloader(tmp_path, session=session)
+    outcomes = dl.download_post(
+        _post(), [_img("media1.png")], "artist-x",
+        is_seen=lambda m: False, recapture=True,
+    )
+    assert outcomes[0].status == "downloaded"
+    assert outcomes[0].path is not None
+
+
 def test_video_routes_to_ytdlp(tmp_path, monkeypatch):
     session = _FakeSession()
     dl = _downloader(tmp_path, session=session)

tests/test_patreon_ingester.py

@@ -88,16 +88,23 @@ class _FakeDownloader:
         self.post_records = 0
 
     def download_post(self, post, media_items, artist_slug, *, is_seen,
-                      should_stop=lambda: False):
+                      should_stop=lambda: False, recapture=False):
         outcomes = []
         for m in media_items:
             if should_stop():
                 break
-            if is_seen(m):
+            seen = is_seen(m)
+            # Recapture mirrors the real downloader: a seen item isn't tier-1
+            # short-circuited — it falls through so an on-disk file surfaces as
+            # skipped_disk (for source_filehash backfill); a seen file NOT on
+            # disk is left alone (skipped_seen, never refetched).
+            if seen and not recapture:
                 outcomes.append(MediaOutcome(media=m, status="skipped_seen", path=None, error=None))
             elif _ledger_key(m) in self.on_disk:
                 p = self.tmp_path / f"{m.post_id}_{m.filename}"
                 outcomes.append(MediaOutcome(media=m, status="skipped_disk", path=p, error=None))
+            elif seen:
+                outcomes.append(MediaOutcome(media=m, status="skipped_seen", path=None, error=None))
             elif _ledger_key(m) in self.error:
                 self.download_calls += 1
                 outcomes.append(MediaOutcome(media=m, status="error", path=None, error="boom"))
@@ -581,6 +588,78 @@ async def test_backfill_recaptures_body_for_already_downloaded_post(
     assert downloader.post_records == 1
 
 
+def _seed_seen(sync_engine, source_id, key, post_id=None):
+    factory = sessionmaker(sync_engine, expire_on_commit=False)
+    with factory() as s:
+        s.add(PatreonSeenMedia(source_id=source_id, filehash=key, post_id=post_id))
+        s.commit()
+
+
+@pytest.mark.asyncio
+async def test_backfill_skips_already_captured_post_but_recapture_forces_it(
+    source_id, sync_engine, tmp_path,
+):
+    """#830: a plain backfill GATES post-record capture on the seen-ledger, so a
+    post whose `post:<id>` key is already seen is NOT recaptured (the operator's
+    'Industrial Lust body missing' bug). RECAPTURE mode bypasses that gate and
+    re-captures the body/links even though the post key is seen — without
+    re-downloading the on-disk media."""
+    m1 = _media("p1", 1)
+    # Pre-seed BOTH the media key and the synthetic post key as already seen.
+    _seed_seen(sync_engine, source_id, _ledger_key(m1), post_id="p1")
+    _seed_seen(sync_engine, source_id, "post:p1", post_id="p1")
+
+    # 1) Plain backfill: post key is seen → gate skips body recapture.
+    client = _FakeClient([(None, [("p1", [m1])])])
+    downloader = _FakeDownloader(tmp_path, on_disk={_ledger_key(m1)})
+    ing = _ingester(sync_engine, tmp_path, client, downloader)
+    res_backfill = ing.run(
+        source_id=source_id, campaign_id="c1", artist_slug="ingest",
+        url="https://patreon.com/ingest", mode="backfill",
+    )
+    assert res_backfill.post_record_paths == []     # gated → not recaptured
+    assert downloader.post_records == 0
+
+    # 2) Recapture: gate bypassed → body recaptured; media NOT re-downloaded;
+    # the on-disk file is surfaced for source_filehash relink.
+    client2 = _FakeClient([(None, [("p1", [m1])])])
+    downloader2 = _FakeDownloader(tmp_path, on_disk={_ledger_key(m1)})
+    ing2 = _ingester(sync_engine, tmp_path, client2, downloader2)
+    res_recap = ing2.run(
+        source_id=source_id, campaign_id="c1", artist_slug="ingest",
+        url="https://patreon.com/ingest", mode="recapture",
+    )
+    assert len(res_recap.post_record_paths) == 1    # gate bypassed → recaptured
+    assert downloader2.post_records == 1
+    assert res_recap.files_downloaded == 0          # no media re-download
+    assert downloader2.download_calls == 0
+    # The on-disk media is surfaced as a (path, CDN url) relink pair.
+    assert len(res_recap.relink_source_paths) == 1
+    rel_path, rel_url = res_recap.relink_source_paths[0]
+    assert rel_url == m1.url
+
+
+@pytest.mark.asyncio
+async def test_recapture_does_not_refetch_seen_media_missing_from_disk(
+    source_id, sync_engine, tmp_path,
+):
+    """Recapture must NOT re-download a seen item that's absent from disk (that's
+    recovery's job) — it returns skipped_seen and contributes no relink pair."""
+    m1 = _media("p1", 1)
+    _seed_seen(sync_engine, source_id, _ledger_key(m1), post_id="p1")
+    client = _FakeClient([(None, [("p1", [m1])])])
+    # NOT in on_disk → seen + missing.
+    downloader = _FakeDownloader(tmp_path)
+    ing = _ingester(sync_engine, tmp_path, client, downloader)
+    result = ing.run(
+        source_id=source_id, campaign_id="c1", artist_slug="ingest",
+        url="https://patreon.com/ingest", mode="recapture",
+    )
+    assert result.files_downloaded == 0
+    assert downloader.download_calls == 0
+    assert result.relink_source_paths == []
+
+
 # --- dead-letter ledger (plan #705 #7) ------------------------------------
 
 

tests/test_sidecar_import.py

@@ -120,6 +120,44 @@ def test_sidecar_source_url_persists_filehash(importer, import_layout):
     assert rec.source_filehash == "0123456789abcdef0123456789abcdef"
 
 
+def test_relink_source_filehash_backfills_existing(importer, import_layout):
+    """#830 recapture (Part B): relink_source_filehash backfills source_url +
+    source_filehash on an existing on-disk image's ImageRecord (matched by
+    sha256) WITHOUT re-importing or unlinking the file. NULL-only — never
+    clobbers a filehash already set."""
+    import_root, _ = import_layout
+    m = import_root / "Alice" / "img.jpg"
+    _split(m, "v")
+    r = importer.import_one(m)
+    assert r.status == "imported"
+    rec = importer.session.get(ImageRecord, r.image_id)
+    assert rec.source_filehash is None
+
+    url = "https://cdn.test/p/0123456789ABCDEF0123456789abcdef/img.jpg"
+    assert importer.relink_source_filehash(m, url) is True
+    importer.session.refresh(rec)
+    assert rec.source_url == url
+    assert rec.source_filehash == "0123456789abcdef0123456789abcdef"
+    assert m.exists()  # never unlinked
+
+    # NULL-only: a second relink with a different url is a no-op.
+    other = "https://cdn.test/p/ffffffffffffffffffffffffffffffff/x.jpg"
+    assert importer.relink_source_filehash(m, other) is False
+    importer.session.refresh(rec)
+    assert rec.source_filehash == "0123456789abcdef0123456789abcdef"
+
+
+def test_relink_source_filehash_no_match_is_noop(importer, import_layout):
+    """A path whose bytes match no ImageRecord (never imported) is a clean no-op,
+    not an error."""
+    import_root, _ = import_layout
+    m = import_root / "Alice" / "ghost.jpg"
+    _split(m, "h")
+    assert importer.relink_source_filehash(
+        m, "https://cdn.test/p/0123456789abcdef0123456789abcdef/ghost.jpg"
+    ) is False
+
+
 def test_reimport_same_post_idempotent(importer, import_layout):
     import_root, _ = import_layout
     # Threshold 0: only an exact phash match collapses. Orthogonal splits

tests/test_source_service.py

@@ -267,6 +267,47 @@ async def test_start_recovery_arms_bypass_flag(db):
     assert "_backfill_bypass_seen" not in (co2 or {})
 
 
+@pytest.mark.asyncio
+async def test_start_recapture_arms_recapture_flag(db):
+    """#830: start_recapture arms the backfill state machine PLUS the
+    _backfill_recapture flag, surfaced on the record so the badge can label it
+    'Recapturing'. Mutually exclusive with recovery (clears bypass_seen); stop
+    clears it."""
+    from backend.app.services.source_service import BACKFILL_MAX_CHUNKS
+
+    artist = await _artist(db, "Alice")
+    svc = SourceService(db)
+    rec = await svc.create(
+        artist_id=artist.id, platform="patreon",
+        url="https://patreon.com/alice",
+    )
+    # Pre-arm recovery, then recapture must clear bypass_seen (mutual exclusion).
+    await svc.start_recovery(rec.id)
+
+    updated = await svc.start_recapture(rec.id)
+    assert updated.backfill_state == "running"
+    assert updated.backfill_recapture is True
+    assert updated.backfill_bypass_seen is False
+    assert updated.backfill_runs_remaining == BACKFILL_MAX_CHUNKS
+
+    co = (await db.execute(
+        select(Source.config_overrides).where(Source.id == rec.id)
+    )).scalar_one()
+    assert co.get("_backfill_recapture") is True
+    assert "_backfill_bypass_seen" not in co
+
+    # to_dict carries the new field for the API/UI.
+    assert updated.to_dict()["backfill_recapture"] is True
+
+    # Stop clears the recapture flag too.
+    stopped = await svc.stop_backfill(rec.id)
+    assert stopped.backfill_recapture is False
+    co2 = (await db.execute(
+        select(Source.config_overrides).where(Source.id == rec.id)
+    )).scalar_one()
+    assert "_backfill_recapture" not in (co2 or {})
+
+
 @pytest.mark.asyncio
 async def test_start_backfill_raises_when_source_missing(db):
     svc = SourceService(db)