gendesign/tradein-mvp/deploy/import-rosreestr.sh
bot-backend 752cb1044b
All checks were successful
CI Trade-In / changes (pull_request) Successful in 7s
CI / changes (pull_request) Successful in 7s
CI Trade-In / frontend-checks (pull_request) Has been skipped
CI / backend-tests (pull_request) Has been skipped
CI / frontend-tests (pull_request) Has been skipped
CI / openapi-codegen-check (pull_request) Has been skipped
CI Trade-In / backend-tests (pull_request) Successful in 51s
fix(tradein/ingest): раз-хардкод rosreestr DKP импорта на всю обл. 66
- scheduler.py import_rosreestr_dkp: снят фильтр city ILIKE '%катеринбург%', address из реального city источника, deals.region_code + новая deals.city заполняются (было: хардкод 'Екатеринбург,' + region_code NULL)
- migration 177: deals.city + индекс + бэкфилл существующих EKB-строк region_code=66/city
- guard city IS NOT NULL → address не NULL для ~15 null-city строк источника
- sync deploy/import-rosreestr.sh (ops-fallback) под тот же oblast-scope
- split dedup-теста: 077 (историческая) хранит EKB-фильтр, живой импорт — нет

Открывает +47183 не-ЕКБ сделок region 66, уже сидящих в источнике, ранее резавшихся на импорте.
2026-07-12 18:11:47 +03:00

90 lines
4.5 KiB
Bash
Executable file
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

#!/usr/bin/env bash
# Импорт реальных сделок Росреестра из gendesign-БД в tradein deals.
#
# Источник: gendesign-postgres-1 / rosreestr_deals (6.8М строк, партиц.).
# Берём ЕКБ квартиры (realestate_type_code=002001003000) за последние ~18 мес.
# rooms выводим из площади (Росреестр не отдаёт кол-во комнат).
# Координаты — NULL, проставляются отдельно (geocode по street).
#
# Только ДКП (вторичка) — ДДУ застройщиков скёюят median, отделены сознательно. См. PR-A.
#
# Запуск на прод-хосте: ./import-rosreestr.sh
# Повторяемо: дедуп по dedup_hash, новый запуск подтянет свежие кварталы.
set -euo pipefail
SRC_PG="${SRC_PG:-gendesign-postgres-1}"
DST_PG="${DST_PG:-tradein-postgres}"
SRC_DB="${SRC_DB:-gendesign}"
SRC_USER="${SRC_USER:-gendesign}"
SINCE="${SINCE:-2024-01-01}"
echo "[$(date -u +%H:%M:%S)] import-rosreestr: ЕКБ квартиры с $SINCE"
# 1. Staging-таблица в tradein.
docker exec "$DST_PG" psql -U tradein -d tradein -v ON_ERROR_STOP=on -c "
-- DROP перед CREATE: страхует от устаревшей формы staging-таблицы, если прошлый
-- запуск упал между CREATE и финальным DROP под старым кодом (без source_id).
DROP TABLE IF EXISTS deals_ros_staging;
CREATE TABLE deals_ros_staging (
dedup_hash text PRIMARY KEY, source_id text, address text, region_code int, city text,
rooms int, area_m2 numeric,
floor int, year_built int, price_rub bigint, price_per_m2 int, deal_date date
);
"
# 2. Трансформируем на gendesign-стороне, пайпим CSV в staging.
docker exec "$SRC_PG" psql -U "$SRC_USER" -d "$SRC_DB" -v ON_ERROR_STOP=on -c "
COPY (
SELECT
'ros:dkp:' || id::text AS dedup_hash,
id::text AS source_id,
trim(city) || ', ' || trim(street) AS address,
region_code AS region_code,
trim(city) AS city,
CASE
WHEN area < 30 THEN 0 WHEN area < 44 THEN 1
WHEN area < 62 THEN 2 WHEN area < 85 THEN 3
ELSE 4 END AS rooms,
round(area, 2) AS area_m2,
LEAST(NULLIF(substring(floor from '[0-9]+'), '')::int, 100) AS floor,
year_build AS year_built,
round(deal_price)::bigint AS price_rub,
round(price_per_sqm)::int AS price_per_m2,
period_start_date AS deal_date
FROM rosreestr_deals
WHERE region_code = 66
AND city IS NOT NULL AND trim(city) <> ''
AND realestate_type_code = '002001003000'
AND area BETWEEN 18 AND 200
AND deal_price BETWEEN 1000000 AND 100000000
AND street IS NOT NULL AND trim(street) <> ''
AND doc_type = 'ДКП'
AND period_start_date >= '$SINCE'
) TO STDOUT WITH CSV
" | docker exec -i "$DST_PG" psql -U tradein -d tradein -v ON_ERROR_STOP=on -c "
COPY deals_ros_staging FROM STDIN WITH CSV
"
STAGED=$(docker exec "$DST_PG" psql -U tradein -d tradein -tAc "SELECT count(*) FROM deals_ros_staging")
echo "[$(date -u +%H:%M:%S)] staged: $STAGED строк"
# 3. Удаляем старые синтетические сделки, вставляем реальные (дедуп).
docker exec "$DST_PG" psql -U tradein -d tradein -v ON_ERROR_STOP=on -c "
DELETE FROM deals WHERE address = 'Екатеринбург, реальная сделка';
INSERT INTO deals (
source, dedup_hash, source_id, address, region_code, city, rooms, area_m2, floor,
year_built, price_rub, price_per_m2, deal_date
)
SELECT 'rosreestr', dedup_hash, source_id, address, region_code, city, rooms, area_m2, floor,
year_built, price_rub, price_per_m2, deal_date
FROM deals_ros_staging
ON CONFLICT (dedup_hash) DO NOTHING;
DROP TABLE deals_ros_staging;
"
TOTAL=$(docker exec "$DST_PG" psql -U tradein -d tradein -tAc "SELECT count(*) FROM deals")
echo "[$(date -u +%H:%M:%S)] deals всего: $TOTAL"
echo "Координаты проставит geocode-deals (отдельный шаг)."