This is a simple function called convert() created for conversion vector files between different formats. Currently, it works for 6 different formats and acquire features like giving specific path for saving a new file and choosing filename for a new file.
Python 3.6+ ; required packages/modules:
- geopandas
- pandas
- numpy
- json
- shapely
Convert.py file contains function convert() and this is the function that converts files. Here is a structure of this function:
convert(from_file, to_format, to_file=None, epsg=None, csv_geo_columns=None)Parameters:
- from_file: str, path object or file-like object
File one's uses for conversion. - to_format: str
Format one's wants to convert a primary file. - to_file: str, desired path object or/and new file name, default
None
An optional argument. New name for converted file. By default, new file will be saved in the same place, where python file (in which conversion happens) currently is and will have the same name as input file (but with diffrent extension). For choosing a specific place on your device or specific filename, additionally put in desired path in this argument. - epsg: int, default
None
An optional argument. EPSG is an id that represents coordinate systems - check http://epsg.io/ for more examples! Also note that some formats, like shp, might contain informations about epsg in metadata. In such cases there is no need to fill epsg argument manually. If there is no epsg stated by user and algorythm cannot read it from metadata, output will be saved in default EPSG:4326 coordinates ( http://epsg.io/4326 ). Notice that Convert() DO NOT perform any transformations, which means you state here the epsg of input file. EPSG of input and output files should be always the same. - csv_geo_columns: list, default
None
An optional argument, yet needed for converting CSV format to any other. List of column name(s) containing geometry. Works for having coordinates in two columns (X, Y) or one (WKT). You CANNOT read csv file without pointing to spatial column(s) in this argument.
Currently, function convert() works with 6 formats:
- ESRI Shapefile
- Geography Markup Language (GML)
- GeoJSON
- TopoJSON
- GeoPackage
- Comma-separated values (CSV)
You can use the convert() function by cloning or downloading files from this repository and opening it in desired IDE or code editor.
Here are some examples of use:
# Simple shapefile (stored in local path input_data/) to GeoJSON conversion - without any optional arguments. This will save the file in working directory as 'file.geojson'
# EPSG will be either read from .prj or, if such file won't be find, will be set to default 4326.
convert(from file = 'input_data/file.shp', to_format = 'geojson')
# Same shapefile (stored in local path input_data/) to GeoJSON conversion, but with arguments - we want to save new file as "awesomefile.geojson" in output_data/ path
convert(from file = 'input_data/file.shp', to_format = 'geojson', to_file='output_data/awesomefile', epsg = 2180)
# CSV to GML conversion, notice the usage of csv_geo_columns argument - lon and lat are colnames in csv containing spatial data
convert(from file = 'data.csv', to_format = 'gml', to_file='newdata', epsg = 2180 , csv_geo_columns = ['lon', 'lat'])This project is: in progress. Feel free to test this function and give us feedback.
If you want to convert data containing information about date, most likely convert() function will change date data type to a string. Please have that in mind if dates in your data are important for further work.