Spatial networks have been an important subject of many studies in quantitative geography and sociology (Barthelemy, 2010). In practical applications, they are traditionally managed in a geospatial vector format, in which spatial entities are recorded as polylines that are spaghetti collections of 2D/3D geospatial coordinates (George & Shekhar, 2008). On the other hand, a network can be regarded as a set of nodes, occupying particular positions and joined in pairs by physical or ephemeral constructs (Gastner & Newman, 2006). In this sense, networks are visualized in a link-node mode, termed as graph data model. Due to its great advantages in simplifying the  epresentation of a network and facilitating related computations powered by graph theory, this model has been widely used as
one of the central tools in spatial analysis (Hostis, 2007). Based on the R platform, a considerable number of packages have emerged for processing both kinds of objects. On the R-spatial website (Bivand, 2007), numerous packages are available for processing spatial data, especially like sp and maptools. As for graph, there are two popular packages to create and manipulate undirected and directed graphs, i.e. graph and igraph. However, there is no such thing to convert a spatial object to a graph-class object. This work aims to fill this gap