Improve randomisation of answers

This commit is contained in:
2026-07-14 12:15:43 +02:00
parent 91c6c16348
commit 650ab37158
3 changed files with 79 additions and 14 deletions
+12 -10
View File
@@ -480,9 +480,10 @@ amateurfunk-anki [--data DIR] [--out DIR] [--seed STR] [--epoch INT]
- `--seed STR` — deterministic seed for answer shuffling. Omit it
for the normal study-deck behavior: a fresh seed is generated for
each build, so answers are reshuffled every time decks are rebuilt.
- `--epoch INT` — override the package timestamp epoch. By default we
derive it from the manifest's `fetched_at`; this flag is mainly for
tests and explicit rebuilds.
- `--epoch INT` — override the package and note-modification timestamp
epoch. By default we use the current rebuild time so Anki recognizes
reshuffled notes as newer on re-import; this flag is mainly for tests
and explicit reproducible builds.
Exit codes: `0` success, `1` configuration / catalog / build error.
There is no Stage-2 equivalent of the fetcher's `EXIT_BAD_STATE`
@@ -602,13 +603,14 @@ same `.apkg` bytes out. Determinism rests on three things:
2. **Stable shuffle.** `randomized_answers()` builds a per-question
`random.Random` seeded from SHA-256 of
`f"{cli_seed}:{question_number}"`.
3. **Stable timestamps.** Every `now` value in the collection (the
`mod` columns, the JSON config blob timestamps) is fixed to
`build_epoch` — derived from the manifest's `fetched_at`, or
overridden via `--epoch`. ZIP member timestamps are also fixed
via `ZipInfo(name, zip_datetime(build_epoch))`. Without this
last step, the inner SQLite would be identical but the
archive's per-entry mtimes would still vary between runs.
3. **Controlled timestamps.** Every `now` value in the collection (the
`mod` columns and JSON config blob timestamps) uses `build_epoch`.
A normal rebuild sets it to the current time because the fresh
shuffle changes note HTML and Anki only updates an existing note
when the imported copy is newer. `--epoch` fixes it explicitly for
reproducible builds. ZIP member timestamps use the same value via
`ZipInfo(name, zip_datetime(build_epoch))`; otherwise the inner
SQLite could be identical while the archive bytes still differed.
The combined effect: two runs with the same `data/`, `--seed`, and
timestamp inputs produce byte-identical sha256 on each `.apkg`.
+18 -4
View File
@@ -37,6 +37,7 @@ import secrets
import shutil
import sqlite3
import sys
import time
import unicodedata
import zipfile
from pathlib import Path
@@ -1790,6 +1791,19 @@ def resolve_shuffle_seed(seed):
return secrets.token_hex(16)
def build_epoch_for_rebuild(override_epoch=None):
"""Return the modification epoch for a multiple-choice deck build.
A normal rebuild reshuffles the answer HTML, so its notes must carry
the current time for Anki to recognize the imported copies as newer.
`--epoch` remains the explicit escape hatch for byte-reproducible
builds, where the caller intentionally fixes every timestamp.
"""
if override_epoch is not None:
return int(override_epoch)
return int(time.time())
def build_all(data_dir, out_dir, seed, override_epoch=None, explanations_path=None):
"""Build every category's `.apkg` and return their result dicts.
@@ -1798,8 +1812,8 @@ def build_all(data_dir, out_dir, seed, override_epoch=None, explanations_path=No
`.apkg` each. Raises `AnkiBuildError` on configuration / catalog
/ explanation-schema problems.
"""
edition_dir, manifest, catalog = load_latest_catalog(data_dir)
build_epoch = build_epoch_from_manifest(manifest, override_epoch)
edition_dir, _manifest, catalog = load_latest_catalog(data_dir)
build_epoch = build_epoch_for_rebuild(override_epoch)
explanations = load_explanations(
explanations_path if explanations_path is not None
else DEFAULT_EXPLANATIONS_PATH
@@ -1866,8 +1880,8 @@ def _parse_args(argv):
type=int,
default=None,
help=(
"override the package timestamp epoch; by default this is "
"derived from manifest.json's fetched_at"
"override the package and note modification timestamp epoch; "
"by default the current rebuild time is used"
),
)
parser.add_argument(
+49
View File
@@ -127,6 +127,19 @@ class TestResolveShuffleSeed(unittest.TestCase):
self.assertNotEqual(first, second)
class TestBuildEpochForRebuild(unittest.TestCase):
"""Multiple-choice rebuilds must look newer to Anki by default."""
def test_default_uses_current_time(self):
with patch("amateurfunk_anki.time.time", return_value=1777777777.9):
self.assertEqual(aa.build_epoch_for_rebuild(), 1777777777)
def test_explicit_epoch_is_preserved_for_reproducible_builds(self):
with patch("amateurfunk_anki.time.time") as current_time:
self.assertEqual(aa.build_epoch_for_rebuild(1234567890), 1234567890)
current_time.assert_not_called()
class TestAnkiBuild(unittest.TestCase):
def setUp(self):
self.tmp = tempfile.TemporaryDirectory()
@@ -662,6 +675,42 @@ class TestAnkiBuild(unittest.TestCase):
]
self.assertFalse(all(byte_equal))
def test_cli_default_rebuild_stamps_notes_as_newer(self):
out_a = self.root / "anki-current-a"
out_b = self.root / "anki-current-b"
common = [
"--data", str(self.data_dir),
"--explanations", str(self.explanations_path),
]
with (
patch(
"amateurfunk_anki.resolve_shuffle_seed",
side_effect=["fresh-a", "fresh-b"],
),
patch(
"amateurfunk_anki.time.time",
side_effect=[1777777777.9, 1777777788.1],
),
):
aa.main([*common, "--out", str(out_a)])
aa.main([*common, "--out", str(out_b)])
def note_mods(out_dir):
apkg = out_dir / "amateurfunk-technische-kenntnisse-n.apkg"
with tempfile.TemporaryDirectory() as tmp:
db_path, _media, _names = extract_collection(apkg, Path(tmp))
conn = sqlite3.connect(db_path)
try:
return {
row[0] for row in conn.execute("select mod from notes")
}
finally:
conn.close()
self.assertEqual(note_mods(out_a), {1777777777})
self.assertEqual(note_mods(out_b), {1777777788})
class TestIntrinsicSvgSize(unittest.TestCase):
"""Parsing of native figure dimensions — SVG only."""