GTFS Feed Architecture & Fundamentals
The relational data model, validation rules, timezone semantics, coordinate reference systems, and versioning practices — the architectural foundations every GTFS pipeline rests on.
Read the overview
Python · GTFS · Public Transit
Production-grade Python for GTFS & public transit data automation.
Parse, normalize, and validate fragmented GTFS static feeds. Extract route geometries, calculate headways, and map stop networks. Integrate GTFS-RT real-time streams with static schedules and build scalable data pipelines and dashboards.
A reference site for transit analysts, urban tech developers, Python GIS engineers, and mobility platform teams who treat transit data as a real data product — versioned, validated, and observable.
Recommended first reads
New to GTFS or the site? These six guides give you the strongest foundation before diving into implementation detail.
Python Parsing & Data Normalization
End-to-end ingestion patterns with pandas, partridge, polars, and gtfs-kit; memory-efficient batch processing, error logging, frequency expansion, and schedule harmonization.
Read the overviewValidation Rules & Schema Errors
Common GTFS validation failures, their remediation patterns, and how to wire validation into ingestion and CI/CD pipelines with Python.
See validation guidesTimezone Handling & Schedule Normalization
IANA timezone resolution, DST transitions, 24+ hour GTFS time semantics, and correct UTC alignment for multi-agency schedule aggregation.
Tackle timezonesCalendar & Service Exception Modeling
Polars streaming, Dask distributed DataFrames, and PyArrow-backed Parquet partitioning — handling multi-gigabyte metropolitan feeds on commodity hardware.
Work out what runs todayGTFS-Realtime Integration
Decode protobuf feeds and merge VehiclePositions, TripUpdates, and ServiceAlerts with static schedules — live vehicle tracking, delay propagation, and arrival prediction.
Go real-timeFour areas most GTFS pipelines skip
Service calendars, fares and station interiors, the two kinds of trip linking, and whether a real-time feed is actually working. Each is a full topic with runnable Python and the failure modes real agency feeds produce.
Calendar & Service Exception Modeling
Expanding calendar.txt patterns into real dates, applying
calendar_dates.txt exceptions in the right order, and finding the
gaps and expiry dates that make a feed stop describing service.
Fare Rules & Pathways Modeling
The v1 zone rules, what Fares v2 can express that they cannot, and the
pathways.txt graph that turns a station from a point into a
navigable interior with honest interchange times.
Transfers & Trip Linking
transfers.txt describes what a rider may do; block_id
describes what a vehicle does. Conflating them charges riders for changes they
never made and invents connections nobody can make.
Monitoring Realtime Feed Quality
Staleness, trip match rate, fleet coverage and outage detection — the four measurements that tell you a live feed has quietly stopped updating while its endpoint keeps returning 200.
Check it is really liveArchitecture, Python implementation, and real-time — end to end.
Twenty-four topics and eighty-four pages in total. Each track collects in-depth guides plus focused how-tos — start at an overview and follow the links into specific implementation patterns.
GTFS Feed Architecture & Fundamentals
Relational structure, validation rules, timezone normalization, and coordinate systems — the architectural fundamentals behind every reliable GTFS pipeline.
- Agency Metadata and Feed Versioning Practices
- Calendar and Service Exception Modeling
- Coordinate Reference Systems for Transit Data
- Fare Rules and Pathways Modeling
- GTFS Validation Rules and Common Schema Errors
- Mastering stops.txt and stop_times.txt Relationships in GTFS
- Spatial Analysis and Route Geometry
- Timezone Handling and Schedule Normalization in GTFS
- Understanding GTFS Static Feed Structure
Python Parsing & Data Normalization
Production-grade ingestion with pandas, partridge, polars, and gtfs-kit. Memory-efficient batch processing, schedule normalization, and automated quality checks.
- Agency Metadata and Feed Versioning Practices in Python
- Automating Feed Updates with gtfs-kit
- Batch Processing Strategies for Multi-Agency GTFS Feeds
- Error Logging and Data Quality Categorization for GTFS Pipelines
- Exporting GTFS to Databases and Warehouses
- Handling Frequency-Based vs Timetable Schedules in GTFS
- Mastering stops.txt and stop_times.txt Relationships in Python
- Memory-Efficient Processing for Large GTFS Feeds
- Parsing GTFS with Pandas and Partridge
- Transfers and Trip Linking
GTFS-Realtime Integration
Decode GTFS-Realtime protobuf feeds and merge VehiclePositions, TripUpdates, and ServiceAlerts with static schedules — live tracking, delay propagation, and arrival prediction in Python.
- Decoding GTFS-Realtime Protobuf Feeds
- Handling GTFS Service Alerts
- Matching Realtime to Static Schedules
- Monitoring Realtime Feed Quality
- Processing TripUpdates and Delays
- Tracking Vehicle Positions in Realtime
Implementation guides worth bookmarking
These step-by-step tutorials go beyond theory — runnable Python, edge cases from real agency feeds, and production-tested remediation patterns.
How to Validate a GTFS Feed with Python
Step-by-step: schema checks, FK validation, and wiring results into a CI pipeline.
Converting Local Transit Times to UTC in Python
Correct handling of 24+ hour GTFS times, DST transitions, and IANA timezone resolution.
Optimizing Pandas Memory Usage for Transit Feeds
Dtype downcast, categorical encoding, and chunked reading — bring 2 GB feeds into memory safely.
Converting frequencies.txt to Exact Departure Times
Materialize headway-based service into concrete departure timestamps for routing and analytics.
Arrival Predictions from GTFS-RT TripUpdates
Apply real-time TripUpdate delays to scheduled stop_times.txt and propagate them across a trip's remaining stops.
Snapping GTFS Stops to Shapes in Python
Linear-referencing with shapely to compute shape_dist_traveled and each stop's off-shape distance.
Expanding GTFS Calendar to Service Dates
Turn weekly patterns and exception rows into the date-indexed service map every other calendar question resolves against.
Reconstructing Vehicle Blocks from block_id
Order a block's trips per service day and catch the overlaps and teleports no schema validator reports.
Measuring GTFS-Realtime Feed Staleness
Catch the feed that keeps answering and stops updating — the failure HTTP monitoring cannot see.