feat(api): auto-link rehearsal sessions on watcher upload and nc-scan

parse_rehearsal_date() extracts YYMMDD / YYYYMMDD from the file path
and get_or_create() a RehearsalSession. Both the watcher nc-upload
endpoint and the nc-scan endpoint now set song.session_id when a
dated folder is detected. Existing songs without a session_id are
back-filled on the next import of the same folder.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

api/src/rehearsalhub/routers/internal.py

@@ -10,8 +10,10 @@ from sqlalchemy.ext.asyncio import AsyncSession
 from rehearsalhub.db.engine import get_session
 from rehearsalhub.repositories.audio_version import AudioVersionRepository
 from rehearsalhub.repositories.band import BandRepository
+from rehearsalhub.repositories.rehearsal_session import RehearsalSessionRepository
 from rehearsalhub.repositories.song import SongRepository
 from rehearsalhub.schemas.audio_version import AudioVersionCreate
+from rehearsalhub.services.session import parse_rehearsal_date
 from rehearsalhub.services.song import SongService
 
 log = logging.getLogger(__name__)
@@ -69,6 +71,15 @@ async def nc_upload(
     if event.nc_file_etag and await version_repo.get_by_etag(event.nc_file_etag):
         return {"status": "skipped", "reason": "version already registered"}
 
+    # Resolve or create rehearsal session from YYMMDD folder segment
+    session_repo = RehearsalSessionRepository(session)
+    rehearsal_date = parse_rehearsal_date(path)
+    rehearsal_session_id = None
+    if rehearsal_date:
+        rehearsal_session = await session_repo.get_or_create(band.id, rehearsal_date, nc_folder)
+        rehearsal_session_id = rehearsal_session.id
+        log.debug("nc-upload: linked to session %s (%s)", rehearsal_session_id, rehearsal_date)
+
     song_repo = SongRepository(session)
     song = await song_repo.get_by_nc_folder_path(nc_folder)
     if song is None:
@@ -76,6 +87,7 @@ async def nc_upload(
     if song is None:
         song = await song_repo.create(
             band_id=band.id,
+            session_id=rehearsal_session_id,
             title=title,
             status="jam",
             notes=None,
@@ -83,6 +95,8 @@ async def nc_upload(
             created_by=None,
         )
         log.info("nc-upload: created song '%s' for band '%s'", title, band_slug)
+    elif rehearsal_session_id and song.session_id is None:
+        song = await song_repo.update(song, session_id=rehearsal_session_id)
 
     # Use first member of the band as uploader (best-effort for watcher uploads)
     from sqlalchemy import select

api/src/rehearsalhub/routers/songs.py

@@ -12,10 +12,12 @@ from rehearsalhub.dependencies import get_current_member
 from rehearsalhub.repositories.audio_version import AudioVersionRepository
 from rehearsalhub.repositories.band import BandRepository
 from rehearsalhub.repositories.comment import CommentRepository
+from rehearsalhub.repositories.rehearsal_session import RehearsalSessionRepository
 from rehearsalhub.repositories.song import SongRepository
 from rehearsalhub.schemas.comment import SongCommentCreate, SongCommentRead
 from rehearsalhub.schemas.song import SongCreate, SongRead, SongUpdate
 from rehearsalhub.services.band import BandService
+from rehearsalhub.services.session import parse_rehearsal_date
 from rehearsalhub.services.song import SongService
 from rehearsalhub.storage.nextcloud import NextcloudClient
 
@@ -156,6 +158,7 @@ async def scan_nextcloud(
 
     nc = NextcloudClient.for_member(current_member)
     version_repo = AudioVersionRepository(session)
+    session_repo = RehearsalSessionRepository(session)
     song_svc = SongService(session)
 
     # dav_prefix to strip full WebDAV hrefs → user-relative paths
@@ -231,6 +234,13 @@ async def scan_nextcloud(
             skipped_count += 1
             continue
 
+        # Resolve rehearsal session from YYMMDD folder segment
+        rehearsal_date = parse_rehearsal_date(nc_file_path)
+        rehearsal_session_id = None
+        if rehearsal_date:
+            rs = await session_repo.get_or_create(band_id, rehearsal_date, nc_folder)
+            rehearsal_session_id = rs.id
+
         # Find or create song record
         song = await song_repo.get_by_nc_folder_path(nc_folder)
         if song is None:
@@ -239,14 +249,17 @@ async def scan_nextcloud(
             log.info("Creating new song '%s' (folder: %s)", song_title, nc_folder)
             song = await song_repo.create(
                 band_id=band_id,
+                session_id=rehearsal_session_id,
                 title=song_title,
                 status="jam",
-                notes=f"Rehearsal: {rehearsal_label}" if rehearsal_label else None,
+                notes=None,
                 nc_folder_path=nc_folder,
                 created_by=current_member.id,
             )
         else:
             log.info("Found existing song '%s' (id: %s)", song.title, song.id)
+            if rehearsal_session_id and song.session_id is None:
+                song = await song_repo.update(song, session_id=rehearsal_session_id)
 
         await song_svc.register_version(
             song.id,

api/src/rehearsalhub/services/session.py

@@ -0,0 +1,47 @@
+"""Helpers for parsing rehearsal session dates from Nextcloud paths."""
+
+from __future__ import annotations
+
+import re
+from datetime import date
+
+# Matches a YYMMDD or YYYYMMDD directory segment anywhere in the path.
+# e.g. "bands/my-band/231015/take.wav" → "231015" → date(2023, 10, 15)
+_YYMMDD_RE = re.compile(r"(?:^|/)(\d{6})(?:/|$)")
+_YYYYMMDD_RE = re.compile(r"(?:^|/)(\d{8})(?:/|$)")
+
+
+def parse_rehearsal_date(nc_file_path: str) -> date | None:
+    """
+    Extract a rehearsal date from a Nextcloud file path.
+
+    Supports YYMMDD (e.g. 231015 → 2023-10-15) and
+    YYYYMMDD (e.g. 20231015 → 2023-10-15) folder names.
+    Returns None if no date segment is found or the date is invalid.
+    """
+    # Try YYYYMMDD first (more specific)
+    m = _YYYYMMDD_RE.search(nc_file_path)
+    if m:
+        s = m.group(1)
+        try:
+            return date(int(s[:4]), int(s[4:6]), int(s[6:8]))
+        except ValueError:
+            pass
+
+    m = _YYMMDD_RE.search(nc_file_path)
+    if m:
+        s = m.group(1)
+        yy, mm, dd = int(s[0:2]), int(s[2:4]), int(s[4:6])
+        # Assume 2000s
+        try:
+            return date(2000 + yy, mm, dd)
+        except ValueError:
+            pass
+
+    return None
+
+
+def nc_folder_for_path(nc_file_path: str) -> str:
+    """Return the parent directory of a file path, with trailing slash."""
+    from pathlib import Path
+    return str(Path(nc_file_path).parent).rstrip("/") + "/"