> From: Luc?a Mart?nez <[email protected]> > Hello Ernesto, > I was also testing every option of the tool. > I am afraid but we do not have enough documentation to determine the > exact behavior of some of the options. > As you say, they may need some adjustments.
Hello, I have created a set of layers almost complete, suitable to perform tests on topological rules used in spatial query, implemented in ArcGIS Desktop and in gvSIG. The layers can be downloaded from url [1]. [1] http://sit.provincia.agrigento.it/scaricabili/temp/ The names of the layers are self explanatory (I hope): inputLines, inputPolygons and inputPoints are the layers from which we want to select features performing the spatial query; maskPoints, maskLines and maskPolygons are the layers where we select features to be used as comparison (example: select features from layer inputPolygons that INTERSECT WITH selected features from layer maskLines). The input theme's attribute table contains, among the other, fields R1, R2, ... , R32. From R1 to R13 they refer to the following ArcGIS Desktop rules: R1 = intersect R2 = are within a distance of R3 = contain R4 = completely contain R5 = contain (Clementini) R6 = are identical to R7 = touch the boundary of R8 = share a line segment with R9 = are crossed by the outline of R10= have their centroid in R11= are within R12= are completely within R13= are within (Clementini) >From R14 to R21 fields refer to the following gvSIG rules: R14 = are the same as R15 = are disjoint from R16 = intersect with R17 = touch R18 = cross over R19 = contain R20 = are contained in R21 = are on the top of The other fields refer to the rules implemented in AdB-Toolbox (developed from OpenJump), but I have not yet tested rules for this software. The meaning of the value for each rule is: 1: the feature from the input layer is selected when the rule is applied; 0: the feature from the input layer is NOT selected when the rule is applied; 9: depends on effective value of distance specified (i.e, not tested for rule R2) -99: the rule cannot be applied for the geometries involved in the comparison (for example polygons cannot be contained in points or in polylines). The values have been obtained using the test I have performed on data, therefore they are liable to contain mistakes. I have found sometimes a strange or senseless behaviour applying some of the rules implemented in gvSIG, so that actually I think that spatial query in gvSIG is not yet quite reliable. Developers and volunteers can perform their own tests to verify or to rebut my assertions. The data can be used to improve documentation, too. Best regards -- ing. Ernesto Sferlazza responsabile nodo provinciale di Agrigento del SITR Sicilia resp. gruppo Sistema Informativo Territoriale Provincia regionale di Agrigento sede istituzionale: piazza Aldo Moro, 1 - 92100 AGRIGENTO sede operativa: via Acrone, 27 - 92100 AGRIGENTO cell. 3392718463 (personale) 3316874296 (aziendale) tel. uff 0922593752 _______________________________________________ Gvsig_internacional mailing list [email protected] http://listserv.gva.es/cgi-bin/mailman/listinfo/gvsig_internacional
