How biltang/meituan_data works
This repository is a small, early-stage Meituan courier/order data-processing project, not a dispatch service. Its implemented workflow transforms waybill rows into courier and order datasets, with utilities for time-based partitioning and overlap analysis; matching is only a stub. The empty README offers no further capability claims, and the provided excerpts show no notebook execution flow, so the graph focuses on the implemented pipeline and its documented input/output files.
Main entry points: data.py Upstream extraction · courier.csv Courier dataset · orders.csv Orders dataset · all_waybill_info_meituan_0322.csv.zip Waybill data
Components 10
Data pipeline
- Upstream extraction
src/meituan/pipeline/data.py - Courier extraction
src/meituan/pipeline/data.py - Order extraction
src/meituan/pipeline/data.py
Data artifacts
- Courier dataset
processed_data/base/courier.csv - Orders dataset
processed_data/base/orders.csv - Parquet timeblocks
src/meituan/pipeline/data.py
Time analysis
- Timeblock utilities
src/meituan/pipeline/data.py - ID overlap analysis
src/meituan/pipeline/data.py - Courier-order matching
src/meituan/pipeline/data.py
Other components
- Waybill data
all_waybill_info_meituan_0322.csv.zip
Connections 9
- Waybill data → Upstream extraction: provides rows
src/meituan/pipeline/data.py - Upstream extraction → Courier extraction: calls
src/meituan/pipeline/data.py - Upstream extraction → Order extraction: calls
src/meituan/pipeline/data.py - Upstream extraction → Courier dataset: writes CSV
src/meituan/pipeline/data.py - Upstream extraction → Orders dataset: writes CSV
src/meituan/pipeline/data.py - Courier extraction → Courier dataset: produces rows
src/meituan/pipeline/data.py - Order extraction → Orders dataset: produces rows
src/meituan/pipeline/data.py - Timeblock utilities → Parquet timeblocks: writes parquet
src/meituan/pipeline/data.py - Timeblock utilities → ID overlap analysis: supports analysis
src/meituan/pipeline/data.py