Table to points

Draft: written, not yet checked by a person Operation · table-to-points

Makes a point layer from a table with coordinate columns, such as a CSV of longitudes and latitudes.

Pitfall. Longitude and latitude swapped puts points in the wrong hemisphere or off the map.

Pitfall. Coordinates stored as text, or with commas as decimal separators, are read as empty.

About this operation
Also called	XY Table To Point, Create points layer from table, CSV to points, Display XY Data, points from xy
Group	`conversion/points`
Checks	`crs.present, crs.area_of_use`
Pitfalls	2 listed above
Columbus tool	Not stated
Notes	Not stated
Status	active
Broader	Not stated
Narrower	Not stated
Related	Not stated

Parameters

Name	Type	Required	Default	Values	Note
`x_field`	field	yes	Not stated	Not stated	Longitude or easting column.
`y_field`	field	yes	Not stated	Not stated	Latitude or northing column.
`crs`	crs	yes	Not stated	Not stated	The CRS the coordinates are in, not the one wanted.

Direction	Name	Type	Geometry	Note
Input	`table`	table	Not stated	Not stated
Output	`output`	vector	point	Not stated

Engine	Tool	Parameters	Differences
`apache-sedona`	`sedona.sql.ST_Point`	`x_field` → `x` `y_field` → `y` `crs` Set it with ST_SetSRID; ST_Point gives the point no CRS.	A Spark SQL function: arguments are columns or column names, run over a Spark DataFrame.
`gdal`	`gdal vector make-point`	`x_field` → `x` `y_field` → `y` `crs` → `output-crs` : The CRS given to the new points. `table` → `input` `output` → `output`	None noted
`geopandas`	`geopandas.points_from_xy`	`x_field` → `x` `y_field` → `y` `crs` → `crs`	Returns the point geometries only; build the GeoDataFrame from the table and them.
`postgis`	`st_makepoint`	`x_field` → `#1` `y_field` → `#2` `crs` Wrap the point in ST_SetSRID(point, srid); ST_MakePoint gives it no CRS.	None noted
`qgis`	`native:createpointslayerfromtable`	`x_field` → `XFIELD` `y_field` → `YFIELD` `crs` → `TARGET_CRS` `table` → `INPUT` `output` → `OUTPUT`	TARGET_CRS defaults to EPSG:4326 if not set, which is wrong for projected coordinates.
`sf`	`sf::st_as_sf`	`x_field` coords = c(x_field, y_field), passed through ... `y_field` coords = c(x_field, y_field), passed through ... `crs` crs = ..., passed through ...; without it the points have no CRS. `table` → `x`	None noted
`shapely`	`shapely.points`	`x_field` → `coords` `y_field` → `y` `crs` Shapely geometries carry no CRS; keep it alongside (a GeoSeries or GeoDataFrame).	Takes the coordinate values themselves (arrays), not column names; geopandas.points_from_xy does the same with a CRS.