perfect-postcode/Taskfile.yml
2026-01-31 13:07:09 +00:00

201 lines
5.3 KiB
YAML

version: '3'
vars:
DATA_DIR: data
ARCGIS_OUTPUT: "{{.DATA_DIR}}/arcgis_data.parquet"
PRICE_PAID_OUTPUT: "{{.DATA_DIR}}/price-paid-complete.parquet"
IOD_OUTPUT: "{{.DATA_DIR}}/IoD2025_Scores.parquet"
POIS_RAW_OUTPUT: "{{.DATA_DIR}}/uk_pois.parquet"
POIS_FILTERED_OUTPUT: "{{.DATA_DIR}}/filtered_uk_pois.parquet"
POI_PROXIMITY_OUTPUT: "{{.DATA_DIR}}/poi_proximity.parquet"
EPC_PP_OUTPUT: "{{.DATA_DIR}}/epc_pp.parquet"
WIDE_OUTPUT: "{{.DATA_DIR}}/wide.parquet"
EPC_CSV: "{{.DATA_DIR}}/epc/certificates.csv"
JOURNEY_TIMES: "{{.DATA_DIR}}/journey_times_bank_checkpoint.parquet"
tasks:
install:
desc: Install dependencies
cmds:
- uv sync
- cd frontend && npm install
download:arcgis:
internal: true
desc: Download and convert ArcGIS postcode data
generates:
- "{{.ARCGIS_OUTPUT}}"
cmds:
- uv run python -m pipeline.download.arcgis --output {{.ARCGIS_OUTPUT}}
download:price-paid:
internal: true
desc: Download and convert Land Registry price-paid data
generates:
- "{{.PRICE_PAID_OUTPUT}}"
cmds:
- uv run python -m pipeline.download.price_paid --output {{.PRICE_PAID_OUTPUT}}
download:deprivation:
internal: true
desc: Download and convert Index of Deprivation data
generates:
- "{{.IOD_OUTPUT}}"
cmds:
- uv run python -m pipeline.download.deprivation_data --output {{.IOD_OUTPUT}}
download:pois:
internal: true
desc: Download and extract POIs from OpenStreetMap
generates:
- "{{.POIS_RAW_OUTPUT}}"
cmds:
- uv run python -m pipeline.download.pois --output {{.POIS_RAW_OUTPUT}}
transform:pois:
internal: true
desc: Transform raw POIs to filtered version with friendly names
deps:
- download:pois
sources:
- "{{.POIS_RAW_OUTPUT}}"
generates:
- "{{.POIS_FILTERED_OUTPUT}}"
cmds:
- uv run python -m pipeline.transform.transform_poi --input {{.POIS_RAW_OUTPUT}} --output {{.POIS_FILTERED_OUTPUT}}
transform:epc-pp:
internal: true
desc: Fuzzy join EPC and Price Paid data
deps:
- download:price-paid
sources:
- "{{.PRICE_PAID_OUTPUT}}"
- "{{.EPC_CSV}}"
generates:
- "{{.EPC_PP_OUTPUT}}"
cmds:
- uv run python -m pipeline.transform.join_epc_pp --epc {{.EPC_CSV}} --price-paid {{.PRICE_PAID_OUTPUT}} --output {{.EPC_PP_OUTPUT}}
transform:poi-proximity:
internal: true
desc: Compute POI proximity counts per postcode
deps:
- download:arcgis
- transform:pois
sources:
- "{{.ARCGIS_OUTPUT}}"
- "{{.POIS_FILTERED_OUTPUT}}"
generates:
- "{{.POI_PROXIMITY_OUTPUT}}"
cmds:
- uv run python -m pipeline.transform.poi_proximity --arcgis {{.ARCGIS_OUTPUT}} --pois {{.POIS_FILTERED_OUTPUT}} --output {{.POI_PROXIMITY_OUTPUT}}
prepare:
desc: Build wide property dataframe with all joins
deps:
- join:epc-pp
- download:arcgis
- download:deprivation
- transform:poi-proximity
sources:
- "{{.EPC_PP_OUTPUT}}"
- "{{.ARCGIS_OUTPUT}}"
- "{{.IOD_OUTPUT}}"
- "{{.POI_PROXIMITY_OUTPUT}}"
generates:
- "{{.WIDE_OUTPUT}}"
cmds:
- uv run python -m pipeline.transform.merge --epc-pp {{.EPC_PP_OUTPUT}} --arcgis {{.ARCGIS_OUTPUT}} --iod {{.IOD_OUTPUT}} --poi-proximity {{.POI_PROXIMITY_OUTPUT}} --journey-times {{.JOURNEY_TIMES}} --output {{.WIDE_OUTPUT}}
test:
cmds:
- uv run -m pipeline.utils.test_fuzzy_join
- uv run pytest pipeline/utils/test_haversine.py
- uv run pytest pipeline/utils/test_poi_counts.py
dev:server:
desc: Run Rust backend on port 8001
dir: server-rs
cmds:
- cargo run --release -- {{.WIDE_OUTPUT}}
dev:frontend:
desc: Run frontend dev server on port 3030 (proxies /api to :8001)
dir: frontend
cmds:
- npm run dev
build:server:
desc: Build server for production
dir: frontend
cmds:
- cargo build --release
build:frontend:
desc: Build frontend for production
dir: frontend
cmds:
- npm run typecheck
- npm run build
lint:
desc: Lint all code (Python, TypeScript, and Rust)
cmds:
- task: lint:python
- task: lint:frontend
- task: lint:rust
lint:python:
desc: Lint Python code with ruff
cmds:
- uv run ruff check .
lint:frontend:
desc: Lint frontend TypeScript code
dir: frontend
cmds:
- npm run lint
- npm run format:check
lint:rust:
desc: Lint Rust code with clippy and check formatting
dir: server-rs
cmds:
- cargo clippy -- -D warnings
- cargo fmt --check
format:
desc: Format all code (Python, TypeScript, and Rust)
cmds:
- task: format:python
- task: format:frontend
- task: format:rust
format:python:
desc: Format Python code with ruff
cmds:
- uv run ruff check --fix .
- uv run ruff format .
format:frontend:
desc: Format frontend TypeScript code
dir: frontend
cmds:
- npm run lint:fix
- npm run format
format:rust:
desc: Format Rust code with cargo fmt
dir: server-rs
cmds:
- cargo fmt --all
check:
desc: Run all checks (lint, typecheck, build)
cmds:
- task: lint
- task: build:server
- task: build:frontend
- task: test