GTFS-GO

QGIS Plugin to extract GTFS-data as GeoJSON and render routes and stops on the Map.

Usage

General

Select datasource

local zipfile
download preset datasource

Processing

Simple parsing - routes and stops

GTFS tables has routes-data and stops-data but tables-structure is somewhat complicated.
This plugin can parse them into simple routes and stops GeoJSON files, also set a beautiful style on layers.

Traffic frequency aggregation

GTFS also has service time-table information. This plugin can aggregate traffic frequency, how many times do each PATH used. PATH means lines between two stops.
In addition, it is possible to unify SIMILAR stops - having same parent_stop or same prefix or same stop_name and near to each.

numbers along with lines indicate a frequency of each lines, set on left side towards direction of path (UK traffic style)
larger number of frequency, lines become bolder
result.csv is a table comparing before and after unified stops.

unifying algorithm

You can see similar stops unified into one stop.

before
after

stops unifying rules

Smaller number of rules is prefered.

parent_stops
- if stops have parent_stops value, unifying them into parent station
- new stop_id is parent's one
stop_id prefix
- by defining delimiter, split stop_name into prefix and suffix, group same prefix stops
- new stop_id is the first stop's one in grouped stops ordered by stop_id ascending.
stop_name and distance
- unifying stops having same stop_name and near to each in certain extent - 0.003 degree in terms of lonlat-plane
- new stop_id is the first stop's one in grouped stops ordered by stop_id ascending.

unifying result

In result.csv, you can see stops unifying result.

Processing Toolbox

The conversions and the repository search are also available as Processing algorithms under the GTFS-GO provider, so they can be used in the Processing Toolbox, Graphical Modeler, batch processing and processing.run(). Input of the conversions is a local GTFS zip file.

Algorithm	ID	Outputs
Extract routes and stops	`gtfsgo:extractroutesandstops`	routes, stops
Aggregate traffic frequency	`gtfsgo:aggregatefrequency`	aggregated routes, aggregated stops, stop relations (table)
Search [Japan]GTFS data repository	`gtfsgo:searchjapandpf`	GTFS feeds (table, `file_url` is the URL of the GTFS zip)

processing.run(
    "gtfsgo:aggregatefrequency",
    {
        "INPUT": "/path/to/gtfs.zip",
        "UNIFY_STOPS": True,
        "DELIMITER": "",
        "DATE": QDate(2024, 4, 1),  # optional
        "BEGIN_TIME": "07:00:00",  # optional, requires END_TIME
        "END_TIME": "09:00:00",
        "OUTPUT_ROUTES": "TEMPORARY_OUTPUT",
        "OUTPUT_STOPS": "TEMPORARY_OUTPUT",
        "OUTPUT_STOP_RELATIONS": "TEMPORARY_OUTPUT",
    },
)

processing.run(
    "gtfsgo:searchjapandpf",
    {
        "TARGET_DATE": QDate(2024, 4, 1),
        "EXTENT": "141.0,144.0,42.0,44.0 [EPSG:4326]",  # optional
        "PREF": 1,  # optional, prefecture code (1: 北海道 ... 47: 沖縄県), 0: any
        "OUTPUT": "TEMPORARY_OUTPUT",
    },
)

Acknowledgements

Version2.0.0, in which the frequency aggregating function is added, got technically and financially supported by Toyota Mobility Foundation and Traffic Brain. Thank you for great contributions!

Contribution

Translation

Translations are JSON dictionaries keyed by the English source text (i18n/<locale>.json), not Qt .ts/.qm files. See i18n/README.md.

wrap a new string with i18n.tr("...") in Python (strings in .ui files are picked up automatically, except notr="true")
run python3 i18n/extract.py --locale ja (and --locale fr) to add the new keys with empty translations
fill the translations in i18n/ja.json / i18n/fr.json (no compilation needed)

new data sources

Some data sources can be added from here however you need to check they have all the required .txt files

Tests

needs pandas

pip install pandas

cd GTFS-GO
python -m unittest discover gtfs_parser/tests

MIERUNE/GTFS-GO

README