GPS Logger Parser

A small Python library and CLI for parsing GPS logger files from multiple device formats and exporting their data to a standardized Parquet (or CSV) format.

Features

• Detects and parses many GPS logger formats (see src/gps_logger_parser/gps)
• CLI and Python API for parsing single files or batching
• Outputs parquet (default) or CSV via the parser API

Installation This project uses uv/uvx for environment and package management. Recommended development setup:

uv sync --dev

Run the CLI directly from the source (no install required):

uvx --from git+https://github.com/NINAnor/gps-logger-parser gps-logger-parser parse path/to/logger_file.txt -o ./out

Command-line usage After installation the package exposes the gps-logger-parser command (entry point).

Parse a file and write a Parquet file into ./out:

gps-logger-parser parse path/to/logger_file.txt -o ./out

(Alternative) run the CLI module directly without installing:

uv run gps-logger-parser parse path/to/logger_file.txt -o ./out

By default the parser writes a file named like the original input with a .parquet suffix.

Python API Use the detect_file helper to obtain a parser instance and write output programmatically:

from pathlib import Path
from gps_logger_parser.parser import detect_file

p = detect_file(Path("test_data/success/gps_gpx/example.gpx"))
p.write_parquet(Path("out"))

Parser instances expose write_parquet(path, filename=None) and helper methods to access the parsed data as a PyArrow table via as_table().

Project layout & tests

• Parsers are implemented under src/gps_logger_parser/gps, accelerometer, tdr, and other_sensor.
• Example test data is available in the test_data directory. -- Run tests with pytest (via uvx after provisioning dev env):

uv sync --dev
uv run pytest

Contributing Contributions and additional parsers are welcome. Please open an issue or a pull request with sample files and expected outputs so a parser can be added or improved.

License This project is licensed under the MIT License.