Data and code for: How stable urban service networks emerge from unstable informal mobility

This deposit contains the data and code for "How stable urban service networks emerge from unstable informal mobility" (Wan, Tian, Cai, Chen & Xing, submitted to Nature Cities). It includes: a 24-hour field census of 4,556 street-vending stalls in Beilin District, Xi'an, China (March–June 2024); derived market-level and grid-level tables; a multimodal travel-cost field; Python code that reproduces every number, table and figure of the paper with one script (run_all.sh). The analysis implements a flow–radius framework with one flow term plus service, attraction and disturbance radii. It covers coverage, temporal succession, resilience under random and targeted node removal, a Huff-type attraction model, ordinal disturbance grading and a random-forest model of built-environment conditions. No personal information was collected. Raw Amap POI data cannot be redistributed; per-cell counts and derived outputs are included. See README.md for the data dictionary, licences and third-party sources. Version: 1.0 Language: English Keywords: street vending; informal economy; urban mobility; everyday infrastructure; service coverage; network resilience; accessibility; Xi'an; China