ingest activities.csv
This commit is contained in:
Davide Scaini
+90
-55
@@ -538,73 +538,108 @@ def _file_suffix(name: str) -> str:
|
||||
|
||||
@app.post("/api/upload")
|
||||
async def upload_activity(
|
||||
file: UploadFile = File(...),
|
||||
files: list[UploadFile] = File(...),
|
||||
store_original: bool = Form(False),
|
||||
) -> JSONResponse:
|
||||
"""Accept a FIT/GPX/TCX file, extract it, update index.json, and re-merge."""
|
||||
"""Accept FIT/GPX/TCX files and/or activities.csv, extract, update index, re-merge.
|
||||
|
||||
activities.csv (Strava export format) can be included in the batch to:
|
||||
- Enrich activity files being uploaded in the same batch (matched by filename)
|
||||
- Retroactively update sidecars for existing activities (matched by strava_id)
|
||||
"""
|
||||
from bincio.extract.ingest import ingest_parsed
|
||||
from bincio.extract.parsers.factory import parse_file
|
||||
from bincio.extract.writer import make_activity_id
|
||||
from bincio.render.merge import merge_all
|
||||
|
||||
dd = _get_data_dir()
|
||||
|
||||
name = Path(file.filename or "upload.fit").name # strip any path components
|
||||
suffix = _file_suffix(name)
|
||||
if suffix not in _SUPPORTED_SUFFIXES:
|
||||
raise HTTPException(400, f"Unsupported file type '{Path(name).suffix}'. Expected FIT, GPX, or TCX.")
|
||||
|
||||
_MAX_UPLOAD_BYTES = 50 * 1024 * 1024 # 50 MB
|
||||
contents = await file.read()
|
||||
if len(contents) > _MAX_UPLOAD_BYTES:
|
||||
raise HTTPException(413, f"File too large ({len(contents)} bytes). Maximum is 50 MB.")
|
||||
|
||||
staging = dd / "_uploads"
|
||||
staging.mkdir(exist_ok=True)
|
||||
staged = staging / name
|
||||
staged.write_bytes(contents)
|
||||
|
||||
kept = False
|
||||
try:
|
||||
from bincio.extract.metrics import compute
|
||||
from bincio.extract.parsers.factory import parse_file
|
||||
from bincio.extract.writer import build_summary, make_activity_id, write_activity, write_index
|
||||
_MAX_UPLOAD_BYTES = 50 * 1024 * 1024 # 50 MB
|
||||
|
||||
activity = parse_file(staged)
|
||||
metrics = compute(activity)
|
||||
activity_id = make_activity_id(activity)
|
||||
|
||||
existing_json = dd / "activities" / f"{activity_id}.json"
|
||||
if existing_json.exists():
|
||||
raise HTTPException(409, f"Activity already exists: {activity_id}")
|
||||
|
||||
write_activity(activity, metrics, dd, privacy="public", rdp_epsilon=0.0001)
|
||||
summary = build_summary(activity, metrics, activity_id, "public")
|
||||
|
||||
# Read current index to preserve owner + existing summaries
|
||||
index_path = dd / "index.json"
|
||||
if index_path.exists():
|
||||
index_data = json.loads(index_path.read_text(encoding="utf-8"))
|
||||
# Separate CSV files from activity files
|
||||
csv_files: list[UploadFile] = []
|
||||
activity_files: list[UploadFile] = []
|
||||
for f in files:
|
||||
name = Path(f.filename or "").name.lower()
|
||||
if name.endswith(".csv"):
|
||||
csv_files.append(f)
|
||||
else:
|
||||
index_data = {"owner": {"handle": "unknown"}, "activities": []}
|
||||
owner = index_data.get("owner", {})
|
||||
existing = {s["id"]: s for s in index_data.get("activities", [])}
|
||||
existing[activity_id] = summary
|
||||
write_index(list(existing.values()), dd, owner)
|
||||
activity_files.append(f)
|
||||
|
||||
if store_original:
|
||||
originals_dir = dd / "originals"
|
||||
originals_dir.mkdir(exist_ok=True)
|
||||
staged.rename(originals_dir / name)
|
||||
kept = True
|
||||
# Build metadata from the first CSV found (activities.csv from Strava export)
|
||||
metadata = None
|
||||
if csv_files:
|
||||
from bincio.extract.strava_csv import StravaMetadata
|
||||
import tempfile
|
||||
csv_upload = csv_files[0]
|
||||
csv_bytes = await csv_upload.read()
|
||||
with tempfile.NamedTemporaryFile(suffix=".csv", delete=False) as tmp:
|
||||
tmp.write(csv_bytes)
|
||||
tmp_path = Path(tmp.name)
|
||||
try:
|
||||
metadata = StravaMetadata(tmp_path)
|
||||
finally:
|
||||
tmp_path.unlink(missing_ok=True)
|
||||
|
||||
from bincio.render.merge import merge_all
|
||||
results = []
|
||||
any_added = False
|
||||
|
||||
for file in activity_files:
|
||||
name = Path(file.filename or "upload.fit").name
|
||||
suffix = _file_suffix(name)
|
||||
if suffix not in _SUPPORTED_SUFFIXES:
|
||||
results.append({"name": name, "ok": False, "error": f"Unsupported file type '{Path(name).suffix}'"})
|
||||
continue
|
||||
|
||||
contents = await file.read()
|
||||
if len(contents) > _MAX_UPLOAD_BYTES:
|
||||
results.append({"name": name, "ok": False, "error": "File too large (max 50 MB)"})
|
||||
continue
|
||||
|
||||
staged = staging / name
|
||||
staged.write_bytes(contents)
|
||||
kept = False
|
||||
try:
|
||||
activity = parse_file(staged)
|
||||
|
||||
# Enrich with CSV metadata when available (matched by filename)
|
||||
if metadata is not None:
|
||||
metadata.enrich(name, activity)
|
||||
|
||||
activity_id = make_activity_id(activity)
|
||||
if (dd / "activities" / f"{activity_id}.json").exists():
|
||||
results.append({"name": name, "ok": False, "error": "duplicate"})
|
||||
continue
|
||||
|
||||
ingest_parsed(activity, dd, privacy="public")
|
||||
|
||||
if store_original:
|
||||
originals_dir = dd / "originals"
|
||||
originals_dir.mkdir(exist_ok=True)
|
||||
staged.rename(originals_dir / name)
|
||||
kept = True
|
||||
|
||||
results.append({"name": name, "ok": True, "id": activity_id})
|
||||
any_added = True
|
||||
except Exception:
|
||||
results.append({"name": name, "ok": False, "error": "Processing failed"})
|
||||
finally:
|
||||
if not kept:
|
||||
staged.unlink(missing_ok=True)
|
||||
|
||||
# Retroactively update sidecars for existing activities matched by strava_id
|
||||
csv_updates = 0
|
||||
if metadata is not None:
|
||||
from bincio.extract.strava_csv import apply_csv_to_data_dir
|
||||
csv_updates = apply_csv_to_data_dir(dd, metadata)
|
||||
|
||||
if any_added or csv_updates:
|
||||
merge_all(dd)
|
||||
|
||||
except HTTPException:
|
||||
raise
|
||||
except Exception as exc:
|
||||
raise HTTPException(422, f"Failed to process activity file: {type(exc).__name__}")
|
||||
finally:
|
||||
if not kept:
|
||||
staged.unlink(missing_ok=True)
|
||||
|
||||
return JSONResponse({"ok": True, "id": activity_id})
|
||||
added = [r for r in results if r["ok"]]
|
||||
return JSONResponse({"ok": True, "added": len(added), "csv_updates": csv_updates, "results": results})
|
||||
|
||||
|
||||
@app.post("/api/import-bas")
|
||||
|
||||
Reference in New Issue
Block a user