Utility functions

utilities in geometry and processing, with parts that help on data preparation and several additional related utilitary functions.

Data Preparations Functions

NetworkSegmentIntersections func

NetworkSegmenIntersections(df, dfi=None, EndPoints=True, tol=1e-3)

if EndPoints True (default), returns

ndf : GeoDataFrame of segmented network pts : GeoDataFrame of endpoints and intersections

if EndPoints False, returns

ndf : GeoDataFrame of segmented network

Segments network lines on each intersection points as a part of data preparation for GraphSims class.

result pts

this result compiles all end points of segmented network lines, appending a new field named 'JnctCnt' as interger, with values of number of connected edges. To derive/detect dead ends, find JnctCnt that have 1 as value.

indexing

The process will explode multilinestrings and split lines. Index will be reset, but other fields from the original feature are kept.

processing duration

function does not have built in multithread processing, but as most of the process are vectorized numpy operations and sindex, on small to medium models, runtime remains acceptable (<10). Larger models will have somewhat exponential duration. i.e., osm roadlines with 42k features took about 50s on a decent computer.

Parameters

df : GeoDataFrame required

Geopandas GeoDataFrame object, of unsegmented network.

EndPoints : Boolean default:True

Switch for the function to calculate and return pts. If False the pts result will return an empty list.

tol : int default:3

Tolerance level in power of 10^-1, with the default value of 3 will round/collect connected edges within 0.001 of projection unit.

Use Example

import SNAPy as sna
import geopandas as gpd

dfNetwork = gpd.read_file('testdata\\Network.gpkg') # network dataframe

dfNetworkSg, IxPts = sna.NetworkSegmentIntersections(dfNetwork)

dfNetworkSg.to_file("SegmentedNetwork.gpkg", layer="Network", crs="EPSG:32748", driver='GPKG') # geoDataFrame can be saved to GIS files
dfNetworkSg.to_file("SegmentedNetwork.gpkg", layer="Points", crs="EPSG:32748", driver='GPKG')

nwSim = sna.GraphSims(dfNetworkSg, dfEntries, settings) # segmented lines to Graphsims

NetworkSegmentDistance func

NetworkSegmentDistance(df, dist:float=50.0)

returns

ndf : GeoDataFrame of segmented network

Segments network lines to an approximate length according to projection units. I.e. a "meter" unit projection, where a line is 150m, with "dist" distance variable of 50m, it will seperate the line into 3 segments.

Minimum length and segmentation rounding

line lengths less than 1.5x than the segment distance will not be segmented. Segmentation rounding uses modulus operation to determine number of segments.

Does not segment intersection

Recommended to use this function AFTER using NetworkSegmentIntersection. Note that segmenting network into smaller segments will cause some/major computing time for any analysis.

Parameters

df : GeoDataFrame required

Geopandas GeoDataFrame object, of a network.

dist : float default:50

float distance for base network segment distance.

Use Example

import SNAPy as sna
import geopandas as gpd

dfNetwork = gpd.read_file('testdata\\Network.gpkg') # network dataframe

dfNetworkSg, IxPts = sna.NetworkSegmenIntersections(dfNetwork)

dfNetworkSg2 = sna.NetworkSegmentDistance(dfNetworkSg, 100) # Segmented by target distance of 100m

dfNetworkSg2.to_file("SegmentedNetwork.gpkg", layer="Network", crs="EPSG:32748", driver='GPKG') # geoDataFrame can be saved to GIS files

nwSim = sna.GraphSims(dfNetworkSg, dfEntries, settings) # segmented lines to Graphsims

@September2024