11 commits

Author	SHA1	Message	Date
bot-backend	4309a12c67	fix(llm): word-boundary INN match, tighten edge cases (#1640 follow-up) ... Replace (?<!\d)/(?!\d) lookarounds on the digit block in _INN_RE with (?<!\w)/(?!\w). The old (?<!\d) did not block alpha-prefixed tokens (e.g. «ИНН ref7707083893»), and (?!\d) did not block alpha-suffixed tokens (e.g. «ИНН 7707083893more»). \b is unsuitable here because Python \w covers both letters and digits, so there is no \b boundary between an alpha char and a digit char. The (?<!\w)/(?!\w) pair correctly anchors the 10/12-digit INN block to non-word boundaries on both sides. Context anchor and checksum gate from #1640 are unchanged. Adds 6 regression tests covering: alpha-prefix, alpha-suffix, embedded mid-token, 12-digit alpha-prefix, punctuation separator (should match), and end-of-string (should match).	2026-06-17 21:24:59 +03:00
bot-backend	5c5be890ab	fix(llm): INN redaction requires context/checksum, stop eating large numbers (#1640) ... _INN_RE now requires an explicit ИНН/inn keyword anchor (case-insensitive) within ~20 chars before a 10/12-digit block. Bare digit sequences without the keyword are no longer candidates — eliminates false-positives on large monetary amounts such as 1 200 000 000 (which coincidentally passes the ФНС checksum). Checksum validation is kept as a second gate to avoid redacting e.g. «ИНН 1234567890» with bad digits. _inn_repl updated to use match.group(1) (digit-only capture group) instead of the full match that now also includes the keyword prefix. 7 new regression tests in test_redaction.py: bare large numbers not redacted, keyword-cued real INNs (10/12 digit) still redacted, bad-checksum + keyword left intact, latin «inn:» accepted.	2026-06-17 20:41:39 +03:00
bot-backend	20efc3bf64	merge: resolve conflicts with main (#1660) into week-review-finish ... 5 conflicts resolved: - redaction.py: merged INN checksum + _PHONE_LOCAL_RE + _Repl alias from main - recommendation.py: kept commercial_sell_through_pct rename (#1635) - report_pdf.py: kept _scenario_deficit_cell + int|None return type (#1590) - parcel.py: kept main's detailed docstring (functionally identical) - alembic/env.py: kept HEAD (app.models import in __init__ already covers all)	2026-06-17 20:21:06 +03:00
bot-backend	14f3ef2019	fix(week-review): backend-аудит v2 — 82 фиксов (#1660) ... Co-authored-by: bot-backend <bot-backend@gendsgn.local> Co-committed-by: bot-backend <bot-backend@gendsgn.local>	2026-06-17 17:13:38 +00:00
Light1YT	249182b678	fix(week-review): дожим ревью — #1569, #1590, #1642 + feat #801 MSW-preview ... Финиш-волна по cross-file/partial остаткам аудита + a11y-харнесс. Полностью (4): #1569 alembic регистрирует все ORM-модели (+убран фантомный Parcel) · #1590 «Индекс дефицита» больше не лжёт на fallback-горизонте (report_pdf+report_md) · #1642 thumbs mtime-freshness (нет устаревшей миниатюры) · #801 /__preview/estimate mock-render для axe/lighthouse (env-gated bypass). Частично (остаток cross-file/домен, открыты): #1593 #1635 #1640. needs-leha (открыт): #1650 (нужна новая миграция вьюхи supply-слоёв). Verify: tsc --noEmit 0; py_compile OK. Closes #1569 Closes #1590 Closes #1642 Closes #801	2026-06-17 11:24:34 +05:00
Light1YT	86e9ea2937	fix(week-review): автофиксы код-ревью — 169 issue (label «week ревью 1») ... Многоагентный аудит + имплементация: один воркер на файл, точечные правки. Верификация: py_compile (47/47 .py) + tsc --noEmit (0 ошибок). Unit-тесты не прогонялись (окружение не поднято: rollup native dep / нет pytest-venv). Полностью исправлено (169): #1336, #1337, #1339, #1340, #1341, #1342, #1343, #1345, #1346, #1348, #1349, #1350, #1351, #1354, #1356, #1358, #1359, #1360, #1362, #1364, #1365, #1366, #1367, #1368, #1369, #1370, #1371, #1372, #1373, #1374, #1375, #1376, #1377, #1378, #1379, #1380, #1381, #1382, #1384, #1385, #1386, #1387, #1388, #1389, #1390, #1391, #1392, #1394, #1395, #1396, #1397, #1399, #1400, #1401, #1402, #1403, #1404, #1408, #1409, #1410, #1411, #1412, #1413, #1414, #1415, #1416, #1417, #1418, #1420, #1423, #1425, #1426, #1427, #1428, #1429, #1430, #1431, #1432, #1433, #1434, #1435, #1437, #1438, #1439, #1440, #1441, #1442, #1443, #1444, #1445, #1446, #1447, #1448, #1449, #1450, #1451, #1452, #1453, #1454, #1455, #1456, #1457, #1458, #1459, #1460, #1461, #1462, #1463, #1464, #1465, #1466, #1467, #1468, #1469, #1471, #1472, #1473, #1474, #1476, #1478, #1479, #1481, #1482, #1483, #1484, #1485, #1487, #1488, #1489, #1490, #1491, #1492, #1493, #1494, #1495, #1496, #1497, #1499, #1500, #1501, #1502, #1504, #1505, #1506, #1507, #1510, #1514, #1515, #1516, #1517, #1518, #1519, #1521, #1522, #1523, #1524, #1525, #1526, #1527, #1528, #1529, #1531, #1532, #1533, #1534, #1535, #1536, #1537, #1538 Частично (9, in-file часть, остаток cross-file): #1361, #1419, #1422, #1424, #1470, #1475, #1477, #1480, #1498 Требуют cross-file (3, не тронуты): #1338, #1363, #1421 Пропущено (1): #1539 Не входило в партию: 22 needs-Leha issue (нужны решения владельца). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>	2026-06-15 20:21:11 +05:00
Light1YT	c29f4aef57	fix(llm): cap серверного Retry-After через _MAX_BACKOFF_S=30s (#1209) ... В complete-loop'е min(...,30) применялся ТОЛЬКО к exponential backoff'у (else-ветка). Серверный Retry-After уходил в time.sleep как есть. _parse_retry_after принимает любое число секунд ("86400".isdigit() → True), а provider'ы шлют Retry-After до 86400с при quota-exhaustion (OpenAI, CDN-503). Эффект: time.sleep блокирует поток anyio-threadpool на часы (до 48ч при llm_max_retries=2). Async-консьюмер (chat.py) мостится через run_in_threadpool — поток держит токен пула (~40 потоков, общий с sync Depends(get_db)) и DB-сессию → пул исчерпывается → стопор приложения. Patch: вынес кап на module-level _MAX_BACKOFF_S=30s, применяю к ОБЕИМ веткам (Retry-After И exp.backoff). raw_wait логируется отдельно для наблюдаемости (видно когда провайдер просил больше). Новый юнит-тест test_rate_limited_retry_after_capped: provider шлёт retry_after=86400 → time.sleep вызывается с 30 (не 86400). 14/14 client тестов + 55/55 LLM-suite зелёные. ruff clean. Closes #1209	2026-06-13 06:07:46 +00:00
Light1YT	f12c03588d	fix(llm): close §19 redaction regex gaps for bare phone / CAPS name / bare SNILS (#1207) ... `_PHONE_RE` требовал префикс +7/8 + хотя бы один разделитель → copy-paste форма «79221234567»/«89221234567» (самая частая) уходила в OpenAI как есть. `_FULLNAME_RE` требовал строчные после первой буквы → стандарт ЕГРН/паспортных выписок «ИВАНОВ ИВАН ИВАНОВИЧ» не матчил. СНИЛС без разделителей не покрывался вообще (`_SNILS_RE` ждёт формат «NNN-NNN-NNN NN»). Для chat-пути scrub — единственный барьер перед OpenAI (SafePayload.text без allowlist'а), при llm_enabled=True PII покидает РФ (нарушение §19/152-ФЗ). Patch: - _PHONE_BARE_RE: (?<!\d)[78]\d{10}(?!\d) — 11-значные с префиксом 7/8. - _SNILS_BARE_RE: (?<!\d)\d{11}(?!\d) — любые 11-значные без разделителей. - _FULLNAME_RE: альтернатива в каждом слове — Titlecase | CAPS-2+. - Конфликт phone-bare vs SNILS-bare разрешён порядком в _PII_PATTERNS: phone раньше SNILS (7/8-префикс семантически точнее). - INN (ровно 10/12) с SNILS-bare не пересекается по длине. 8 новых юнит-тестов покрывают все три дыры + конфликт-резолюцию. 24/24 redaction тестов + 55/55 LLM-suite зелёные. Closes #1207	2026-06-13 09:56:58 +05:00
Light1YT	645b3c14a3	fix(chat): clean квартирография rendering in chat_system prompt (v2, #957) ... Live chat rendered the recommended unit-mix as a garbled rank-hyphen-type list ("2-1-к", "3-2-к" = "rank 2: 1-к"). Add rule 6 to chat_system@v2: list квартирография/segments by clean TYPE (студия / 1-к / 2-к / 3-к / 80+ м²) + share %, no ordinal-number-hyphen-type concatenation. Bumped prompt version 1->2 (versioned-prompt convention). Test locks v2 + the rule. Refs #957	2026-06-09 15:07:14 +05:00
Light1YT	fceaaf9a2c	feat(chat): LLM tool-loop + §19 redaction wiring for #957 (Step 2+3) ... Add the LLM prose-composition path for the parcel-forecast chat, layered over the deterministic Step-1 fallback which stays the safety net. - chat/tools.py: 5 read-only section tools (exec_summary, product_recommendation, forecast, risks, scenarios) — pure slices of the loaded report dict, no DB/ recompute, graceful on missing sections. market_now (raw analyze blob) and meta are deliberately NOT exposed -> highest-PII data cannot reach the LLM. - chat/safe_payload.py: the §19 gate — single place that builds the outbound SafePayload from a section-aggregate allowlist; honors is_confidential hard-block. - chat/orchestrator.py: manual tool-call loop with call-cap/termination, real grounded_in provenance; any LLMResult.ok=False (disabled/timeout/rate_limited/ redaction_refused/call_cap/provider_error/empty) degrades to the deterministic answer. - llm/prompts.py: versioned chat_system@v1 — answer only from sections, never fabricate numbers, advisory tone, decline out-of-scope. - api/v1/chat.py: branch on settings.llm_enabled; sync complete bridged via run_in_threadpool. Default-off -> deterministic path, no provider built. - Tests: fake provider only (no network), planted-secret redaction-boundary + per-reason fallback + call-cap + numbers-from-report coverage. Refs #957	2026-06-08 17:45:01 +05:00
Light1YT	4af7ba5a40	feat(llm): foundational LLM infra package with §19 redaction + deterministic fallback (#960) ... Optional external-OpenAI layer over the deterministic forecasting engine. Gated by llm_enabled (default False) so prod makes no network calls until deliberately enabled. Allowlist-first SafePayload contract + is_confidential hard-block + RU-PII regex scrub (mandatory on the external path). Abstract LLMProvider seam (is_external) for a future RU-hosted provider. Sync httpx core (Celery-friendly); tool/function-calling pass-through; timeout + bounded 429/5xx retry + per-request call cap, all degrading to fallback. Raw httpx (no openai SDK -> no pyproject/lock drift). 47 tests, ruff + mypy clean. Refs #960	2026-06-08 15:44:16 +05:00