Enhance CSV metadata with error tracking for mosaics and tiles

This commit is contained in:
2026-04-25 16:06:54 -04:00
parent e8d3bf7180
commit ae37c06f15
10 changed files with 406 additions and 34 deletions
+7 -3
View File
@@ -153,13 +153,17 @@ Tile filenames encode position: `tile_r{row}_c{col}.jpg` where row increases wit
### Metadata files
**`scans.csv`** columns: `machine`, `machine_id`, `scan_id`, `name`, `scan_time`, `start_x`, `start_y`, `end_x`, `end_y`, `dx`, `dy`, `nx`, `ny`, `total_tiles`, `scan_lines`, `scan_mode`, `start_datetime`, `end_datetime`, `status`, `user`, `disk_space_mb`, `mosaic_url`, `mosaic_local_path`, `mosaic_on_disk`
**`scans.csv`** columns: `machine`, `machine_id`, `scan_id`, `name`, `scan_time`, `start_x`, `start_y`, `end_x`, `end_y`, `dx`, `dy`, `nx`, `ny`, `total_tiles`, `scan_lines`, `scan_mode`, `start_datetime`, `end_datetime`, `status`, `user`, `disk_space_mb`, `mosaic_url`, `mosaic_local_path`, `mosaic_on_disk`, `mosaic_download_status`, `mosaic_error`, `mosaic_error_code`, `mosaic_error_class`
- `mosaic_on_disk`: `True` if `mosaic.jpg` exists on disk at row-write time, regardless of which run downloaded it. Useful for inventory — reflects actual archive state rather than what happened in the current run.
- `mosaic_download_status`: one of `downloaded`, `failed`, `already_done`, `dry_run`, `skipped_metadata_only` (in `--metadata-only` mode). Failed attempts are still written so you can see missing server-side images in the same CSV.
- `mosaic_error` / `mosaic_error_code` / `mosaic_error_class`: set when the URL was tried and the file was not stored successfully. **`mosaic_error_class`** is a coarse hint: `permanent_missing` for HTTP 404/410, `transient` for 5xx or common network/timeout-style failures, and `unknown` for other cases (including a 200 with an empty body). **Rows are append-only;** a failed download leaves an audit record without overwriting prior runs history. Delete or rotate the CSVs if you need a new header (see `spruce.settings.SCANS_CSV_FIELDS` / `TILES_CSV_FIELDS`).
**`tiles.csv`** columns: `machine`, `machine_id`, `scan_id`, `scan_time`, `row_index`, `col_index`, `x_mm`, `y_mm`, `url`, `local_path`, `downloaded_at`, `file_size_bytes`
**`tiles.csv`** columns: `machine`, `machine_id`, `scan_id`, `scan_time`, `row_index`, `col_index`, `x_mm`, `y_mm`, `url`, `local_path`, `status`, `error`, `error_code`, `error_class`, `downloaded_at`, `file_size_bytes`
- `downloaded_at`: ISO 8601 UTC timestamp of when the tile was fetched. Empty if the download failed.
- `status`: `downloaded`, `failed`, or `dry_run` (if `--dry-run`). Failed rows are kept for the same reason as mosaics.
- `error` / `error_code` / `error_class`: same rough semantics as the mosaic fields (`permanent_missing` / `transient` / `unknown`). `error_code` is the HTTP status when available.
- `downloaded_at`: ISO 8601 UTC timestamp when the tile was fetched. Empty on failure.
---
+9 -1
View File
@@ -313,11 +313,19 @@ def _print_summary(
row(
"Mosaics failed:",
str(totals.mosaics_failed),
"0 bytes or HTTP error; see log above",
"0 bytes or HTTP error; see scans.csv and logs",
)
)
if not metadata_only and not mosaic_only:
log.info(row("Tiles downloaded:", str(totals.tiles_downloaded)))
if not dry_run and not metadata_only:
log.info(
row(
"Mosaic & tile errors:",
"",
f"{SCANS_CSV_FILENAME} & {TILES_CSV_FILENAME} (error_class, error, error_code)",
)
)
if metadata_only:
log.info(row("Metadata written:", str(totals.metadata_written), "new JSON files"))
log.info(sep)
+63
View File
@@ -0,0 +1,63 @@
"""
Structured HTTP download result and error classification.
"""
from __future__ import annotations
from dataclasses import dataclass
import requests
# public constants for error_class
PERMANENT_MISSING = "permanent_missing"
TRANSIENT = "transient"
UNKNOWN = "unknown"
OK = "" # success (no error class)
@dataclass(frozen=True)
class DownloadResult:
"""Result of a streaming download (after all retries if applicable)."""
size: int
status_code: int | None
error: str | None
error_class: str
@property
def ok(self) -> bool:
return self.size > 0 and self.error is None
def classify_http_error(
status_code: int | None, exc: BaseException | None
) -> str:
"""
404/410 => likely gone forever.
5xx and transport/timeouts => retry may help.
"""
if status_code in (404, 410):
return PERMANENT_MISSING
if status_code is not None and 500 <= status_code < 600:
return TRANSIENT
if exc is not None:
if isinstance(
exc,
(
requests.Timeout,
requests.ConnectTimeout,
requests.ReadTimeout,
),
):
return TRANSIENT
if isinstance(exc, (requests.exceptions.ConnectionError, OSError)):
return TRANSIENT
if isinstance(exc, requests.exceptions.ChunkedEncodingError):
return TRANSIENT
return UNKNOWN
def error_code_str(status_code: int | None) -> str:
if status_code is None:
return ""
return str(status_code)
+51 -20
View File
@@ -5,10 +5,12 @@ High-level scrape orchestration: drives the per-machine and per-scan loops.
import json
import logging
from concurrent.futures import ThreadPoolExecutor, as_completed
from dataclasses import dataclass, field
from dataclasses import dataclass
from pathlib import Path
from typing import Any
from spruce.download_result import error_code_str
@dataclass
class RunStats:
@@ -46,6 +48,17 @@ log = logging.getLogger(__name__)
# ---------------------------------------------------------------------------
@dataclass
class MosaicAttempt:
"""Outcome of a mosaic download attempt (for scans.csv and RunStats)."""
downloaded: bool
csv_status: str
error: str
error_code: str
error_class: str
def _download_mosaic(
sess: MachineSession,
scan_meta: dict[str, Any],
@@ -55,17 +68,19 @@ def _download_mosaic(
machine: dict[str, Any],
config: dict[str, Any],
dry_run: bool,
) -> bool:
"""Download the scan mosaic if not already done. Returns True if downloaded."""
) -> MosaicAttempt:
"""Download the scan mosaic if not already done."""
url = sess.mosaic_url(scan_id)
if progress.is_done(url):
return False
return MosaicAttempt(
False, "already_done", "", "", ""
)
if dry_run:
log.info("[DRY-RUN] Mosaic: %s%s", url, mosaic_path)
return False
return MosaicAttempt(False, "dry_run", "", "", "")
log.info("[%s] Downloading mosaic for scan %d", machine["label"], scan_id)
size = sess.download_file(url, mosaic_path)
if size:
res = sess.download_file(url, mosaic_path)
if res.ok:
if config.get("write_exif", True):
mmeta: dict[str, Any] | None = config.get("machine_metadata", {}).get(
machine["label"]
@@ -79,10 +94,16 @@ def _download_mosaic(
"[%s] Mosaic saved: %s (%.1f MB)",
machine["label"],
mosaic_path,
size / 1e6,
res.size / 1e6,
)
return MosaicAttempt(True, "downloaded", "", "", "")
return MosaicAttempt(
False,
"failed",
res.error or "",
error_code_str(res.status_code),
res.error_class,
)
return True
return False
def _download_tiles_for_scan(
@@ -161,8 +182,8 @@ def _download_tiles_for_scan(
) as pbar:
for future in as_completed(futures):
result = future.result()
if result.get("file_size_bytes"):
batch.append(result)
if result.get("status") == "downloaded":
progress.mark_done(result["url"])
downloaded += 1
pbar.update(1)
@@ -291,9 +312,9 @@ def process_scan(
mosaic_url = sess.mosaic_url(scan_id)
mosaic_already_done = progress.is_done(mosaic_url)
if metadata_only:
mosaic_just_downloaded = False
mosaic_attempt: MosaicAttempt | None = None
else:
mosaic_just_downloaded = _download_mosaic(
mosaic_attempt = _download_mosaic(
sess,
scan_meta,
scan_id,
@@ -303,16 +324,22 @@ def process_scan(
config,
dry_run,
)
if not metadata_only and mosaic_just_downloaded:
if not metadata_only and mosaic_attempt:
if mosaic_attempt.downloaded:
stats.mosaics_downloaded += 1
elif (
not metadata_only
and not dry_run
and not mosaic_already_done
and not mosaic_just_downloaded
):
elif not dry_run and mosaic_attempt.csv_status == "failed":
stats.mosaics_failed += 1
if metadata_only:
mds, mer, mco, mcl = "skipped_metadata_only", "", "", ""
elif mosaic_attempt is not None:
mds = mosaic_attempt.csv_status
mer = mosaic_attempt.error
mco = mosaic_attempt.error_code
mcl = mosaic_attempt.error_class
else:
mds, mer, mco, mcl = "", "", "", ""
# Write scan-level CSV row only if this scan hasn't been recorded before.
if mosaic_already_done and not metadata_only:
log.debug(
@@ -347,6 +374,10 @@ def process_scan(
"mosaic_url": mosaic_url,
"mosaic_local_path": str(mosaic_path),
"mosaic_on_disk": mosaic_path.exists(),
"mosaic_download_status": mds,
"mosaic_error": mer,
"mosaic_error_code": mco,
"mosaic_error_class": mcl,
}
)
+46 -7
View File
@@ -12,6 +12,13 @@ from typing import Any
import requests
from bs4 import BeautifulSoup
from spruce.download_result import (
OK,
UNKNOWN,
DownloadResult,
classify_http_error,
error_code_str,
)
from spruce.parsers import parse_scan_row, parse_scan_view, _grid_values
log = logging.getLogger(__name__)
@@ -205,8 +212,14 @@ class MachineSession:
# Downloads
# ------------------------------------------------------------------
def download_file(self, url: str, dest: Path, retries: int = 3) -> int:
"""Stream-download url to dest with retries. Returns bytes written (0 on failure)."""
def download_file(
self, url: str, dest: Path, retries: int = 3
) -> DownloadResult:
"""
Stream-download url to dest with retries.
Returns a DownloadResult (bytes, optional HTTP code, final error, class).
"""
dest.parent.mkdir(parents=True, exist_ok=True)
backoff = 5.0
for attempt in range(1, retries + 1):
@@ -221,8 +234,21 @@ class MachineSession:
if chunk:
fh.write(chunk)
size += len(chunk)
return size
if size == 0:
return DownloadResult(
0,
resp.status_code,
"0 bytes in response body",
UNKNOWN,
)
return DownloadResult(size, resp.status_code, None, OK)
except Exception as exc:
sc: int | None = None
if (
isinstance(exc, requests.HTTPError)
and exc.response is not None
):
sc = exc.response.status_code
if attempt < retries:
log.debug(
"Attempt %d/%d failed %s: %s — retrying in %.0fs",
@@ -241,7 +267,9 @@ class MachineSession:
url,
exc,
)
return 0
cl = classify_http_error(sc, exc)
return DownloadResult(0, sc, str(exc), cl)
return DownloadResult(0, None, "download_file: exhausted", UNKNOWN)
def download_tile(
self, tile: dict[str, Any], dest: Path, dry_run: bool
@@ -260,11 +288,22 @@ class MachineSession:
"local_path": str(dest),
"downloaded_at": "",
"file_size_bytes": "",
"status": "",
"error": "",
"error_code": "",
"error_class": "",
}
if dry_run:
row["status"] = "dry_run"
return row
size = self.download_file(tile["url"], dest)
if size:
res = self.download_file(tile["url"], dest)
if res.ok:
row["status"] = "downloaded"
row["downloaded_at"] = datetime.now(timezone.utc).isoformat()
row["file_size_bytes"] = size
row["file_size_bytes"] = res.size
else:
row["status"] = "failed"
row["error"] = res.error or ""
row["error_code"] = error_code_str(res.status_code)
row["error_class"] = res.error_class
return row
+8
View File
@@ -47,6 +47,10 @@ SCANS_CSV_FIELDS: list[str] = [
"mosaic_url",
"mosaic_local_path",
"mosaic_on_disk",
"mosaic_download_status",
"mosaic_error",
"mosaic_error_code",
"mosaic_error_class",
]
TILES_CSV_FIELDS: list[str] = [
@@ -60,6 +64,10 @@ TILES_CSV_FIELDS: list[str] = [
"y_mm",
"url",
"local_path",
"status",
"error",
"error_code",
"error_class",
"downloaded_at",
"file_size_bytes",
]
+36
View File
@@ -0,0 +1,36 @@
"""Tests for spruce.download_result classification."""
import requests
from spruce.download_result import (
PERMANENT_MISSING,
TRANSIENT,
UNKNOWN,
classify_http_error,
error_code_str,
)
def test_classify_404_permanent():
assert classify_http_error(404, None) == PERMANENT_MISSING
def test_classify_410_permanent():
assert classify_http_error(410, None) == PERMANENT_MISSING
def test_classify_503_transient():
assert classify_http_error(503, None) == TRANSIENT
def test_classify_timeout_transient():
assert classify_http_error(None, requests.Timeout()) == TRANSIENT
def test_classify_unknown_4xx():
assert classify_http_error(403, None) == UNKNOWN
def test_error_code_str():
assert error_code_str(None) == ""
assert error_code_str(404) == "404"
+85
View File
@@ -0,0 +1,85 @@
"""Mosaic download outcomes for scans.csv (RunStats / MosaicAttempt)."""
from pathlib import Path
from unittest.mock import MagicMock
from spruce.download_result import (
DownloadResult,
PERMANENT_MISSING,
TRANSIENT,
UNKNOWN,
)
from spruce.orchestrator import _download_mosaic
_MACHINE = {"label": "M", "option_value": "v", "machine_id": "1"}
_CONFIG: dict = {
"write_exif": False,
"timeout": 10,
"request_delay": 0.0,
"machine_metadata": {},
}
def test_download_mosaic_404_permanent_class():
sess = MagicMock()
url = "http://img/RootView_Database/9/mosaic.jpg"
sess.mosaic_url.return_value = url
sess.download_file.return_value = DownloadResult(0, 404, "404", PERMANENT_MISSING)
progress = MagicMock()
progress.is_done.return_value = False
mpath = Path("/tmp/mosaic_404.jpg")
out = _download_mosaic(
sess, {}, 9, mpath, progress, _MACHINE, _CONFIG, dry_run=False
)
assert out.csv_status == "failed"
assert out.error_class == PERMANENT_MISSING
assert out.error_code == "404"
progress.mark_done.assert_not_called()
def test_download_mosaic_503_transient_class():
sess = MagicMock()
sess.mosaic_url.return_value = "http://x/m.jpg"
sess.download_file.return_value = DownloadResult(0, 503, "err", TRANSIENT)
progress = MagicMock()
progress.is_done.return_value = False
out = _download_mosaic(
sess,
{},
1,
Path("/tmp/m.jpg"),
progress,
_MACHINE,
_CONFIG,
dry_run=False,
)
assert out.error_class == TRANSIENT
assert out.error_code == "503"
def test_download_mosaic_empty_body_unknown():
sess = MagicMock()
sess.mosaic_url.return_value = "http://x/m.jpg"
sess.download_file.return_value = DownloadResult(
0, 200, "0 bytes in response body", UNKNOWN
)
progress = MagicMock()
progress.is_done.return_value = False
out = _download_mosaic(
sess,
{},
1,
Path("/tmp/m.jpg"),
progress,
_MACHINE,
_CONFIG,
dry_run=False,
)
assert out.error_class == UNKNOWN
assert "bytes" in out.error or out.error
+75
View File
@@ -0,0 +1,75 @@
"""HTTP download result wiring in MachineSession."""
from pathlib import Path
import pytest
import requests
from spruce.download_result import PERMANENT_MISSING, TRANSIENT
from spruce.session import MachineSession
def _minimal_config() -> dict:
return {
"base_url": "http://127.0.0.1:8010/",
"image_base_url": "http://127.0.0.1:8011/",
"username": "u",
"password": "p",
"timeout": 10,
"request_delay": 0.0,
"workers": 2,
}
def _machine() -> dict:
return {
"label": "T [AMR-0]",
"option_value": "x",
"machine_id": "0",
}
def test_download_file_http_404_result(tmp_path: Path):
dest = tmp_path / "f.bin"
sess = MachineSession(_machine(), _minimal_config())
def fail_get(*_a, **_k):
r = requests.Response()
r.status_code = 404
r.url = "http://example/mosaic.jpg"
r.reason = "Not Found"
r.raise_for_status()
sess.http.get = fail_get # type: ignore[method-assign]
res = sess.download_file("http://example/mosaic.jpg", dest, retries=1)
assert res.size == 0
assert res.status_code == 404
assert res.error_class == PERMANENT_MISSING
assert "404" in (res.error or "")
def test_download_tile_row_on_failure(tmp_path: Path):
dest = tmp_path / "t.jpg"
tile = {
"scan_id": 1,
"row_index": 0,
"col_index": 0,
"x_mm": 0.0,
"y_mm": 0.0,
"url": "http://example/tile",
}
sess = MachineSession(_machine(), _minimal_config())
def fail_get(*_a, **_k):
r = requests.Response()
r.status_code = 500
r.url = "http://example"
r.reason = "Server Error"
r.raise_for_status()
sess.http.get = fail_get # type: ignore[method-assign]
row = sess.download_tile(tile, dest, dry_run=False)
assert row["status"] == "failed"
assert row["error_code"] == "500"
assert row["error_class"] == TRANSIENT
+23
View File
@@ -7,6 +7,8 @@ import yaml
from spruce.settings import (
MAX_SAFE_WORKERS,
SCANS_CSV_FIELDS,
TILES_CSV_FIELDS,
_clamp_workers,
load_config,
)
@@ -91,3 +93,24 @@ def test_load_config_missing_password_exits(tmp_path):
path.write_text(yaml.dump({"username": "x"}))
with pytest.raises(SystemExit):
load_config(str(path))
# ---------------------------------------------------------------------------
# CSV schemas (failure columns)
# ---------------------------------------------------------------------------
def test_scans_csv_fields_includes_mosaic_failure_columns():
for name in (
"mosaic_download_status",
"mosaic_error",
"mosaic_error_code",
"mosaic_error_class",
):
assert name in SCANS_CSV_FIELDS
def test_tiles_csv_fields_includes_status_and_error_columns():
for name in ("status", "error", "error_code", "error_class"):
assert name in TILES_CSV_FIELDS
assert TILES_CSV_FIELDS.index("status") < TILES_CSV_FIELDS.index("downloaded_at")