<html><head></head><body><div style="color:#000; background-color:#fff; font-family:verdana, helvetica, sans-serif;font-size:16px"><div id="yui_3_16_0_1_1454438917798_1037680">Those are more issues with neighbouring polygons - you can also get (depending on source data) unclosed polygons & self overlaps - butterfly or figure 8 polygons - which are also issues in polygon topology.</div><div id="yui_3_16_0_1_1454438917798_1037869"><br></div><div id="yui_3_16_0_1_1454438917798_1037871">Unclosed polygons are formed when non-touching linestrings are used to build a topology from, & the topology snapping tolerance is smaller than the gap between the line vertices.<br></div><div id="yui_3_16_0_1_1454438917798_1037557"><span></span></div>A self intersection is usually a digitising error. Like you , I use an external program to create topologies - you suggested GRASS, in my case I use Postgis, & once the topology structure is built from the input linestrings, you can extract/derive the resulting polygons.<br> <div id="yui_3_16_0_1_1454438917798_1037884" class="qtdSeparateBR"><div id="yui_3_16_0_1_1454438917798_1039138"><br></div><div id="yui_3_16_0_1_1454438917798_1040342" dir="ltr">I have not had this problem in QGIS when digitising with snapping is on - provided it is on for points & vertices, all the relevant layers & the tolerance is set appropriately.</div><div id="yui_3_16_0_1_1454438917798_1040725" dir="ltr"><br></div><div id="yui_3_16_0_1_1454438917798_1040374" dir="ltr">Postgis has tools to identify the locations of errors, zooming in to that location with QGIS with snapping on has let me fix them manually. I usually create the underlying linestrings, then the topology, then the polygons. Any errors are addressed in the original linestrings, then the derived data recreated. Repeat until it is acceptable. <br></div><div id="yui_3_16_0_1_1454438917798_1040596" dir="ltr"><br></div><div id="yui_3_16_0_1_1454438917798_1040499" dir="ltr">I have a minor issue with some auto fix tools, as they typically involve moving one point (or vertex) to the same location as another. However - you do not always have a say in which point of the two gets moved - & it can be the wrong one. <br></div><div id="yui_3_16_0_1_1454438917798_1040372" dir="ltr"><br></div><div id="yui_3_16_0_1_1454438917798_1040371" dir="ltr">Brent Wood<br></div><br></div><div style="display: block;" id="yui_3_16_0_1_1454438917798_1037878" class="yahoo_quoted"> <div id="yui_3_16_0_1_1454438917798_1037877" style="font-family: verdana, helvetica, sans-serif; font-size: 16px;"> <div id="yui_3_16_0_1_1454438917798_1037876" style="font-family: HelveticaNeue, Helvetica Neue, Helvetica, Arial, Lucida Grande, sans-serif; font-size: 16px;"> <div id="yui_3_16_0_1_1454438917798_1037875" dir="ltr"> <font id="yui_3_16_0_1_1454438917798_1037874" face="Arial" size="2"> <hr id="yui_3_16_0_1_1454438917798_1037873" size="1"> <b><span style="font-weight:bold;">From:</span></b> Bernhard Ströbl <bernhard.stroebl@jena.de><br> <b><span style="font-weight: bold;">To:</span></b> qgis-user@lists.osgeo.org <br> <b><span style="font-weight: bold;">Sent:</span></b> Tuesday, February 9, 2016 8:44 PM<br> <b><span style="font-weight: bold;">Subject:</span></b> Re: [Qgis-user] Repairing Topology Errors<br> </font> </div> <div id="yui_3_16_0_1_1454438917798_1037882" class="y_msg_container"><br>Hi,<br clear="none"><br clear="none">when looking at topology errors I can imagine three types<br clear="none">1) borders are not identical, i.e. you missed one vertex during digitizing<br clear="none">2) overlaps<br clear="none">3) gaps<br clear="none"><br clear="none">ad 1) in my experience it is easy to miss a vertex when digitizing along <br clear="none">an existing edge, the auto-trace plugin may help in this respect and <br clear="none">also "avoid intersections" (although you seem to have had problems with <br clear="none">it). Snapping must be activated, of course. You cannot snap without it <br clear="none">because you never hit the exact same point twice.<br clear="none">So for your edit layer you should snap to vertex with a reasonable <br clear="none">snapping distance and "avoid intersections". Thus you should be set to <br clear="none">not miss any vertices.<br clear="none">fix: only manually AFAIK<br clear="none">2) have "avoid intersections" active<br clear="none">fix: I use GRASS with its topological data model to identify all <br clear="none">averlaps and use Processing's "eliminate sliver polygons" to get rid of them<br clear="none">3) no way to avoid them because IMHO they could be intended<br clear="none">fix: same as 2)<br clear="none"><br clear="none">Bernhard<br clear="none"><br clear="none"><br clear="none">Am 08.02.2016 um 20:26 schrieb Springfield Harrison:<br clear="none">> Hello Group,<br clear="none">><br clear="none">> I have been manually creating a few simple polygons, always intending to<br clear="none">> have no gaps, slivers or overlaps. To do this I use snapping and<br clear="none">> occasionally the Autotrace feature.<br clear="none">><br clear="none">> However, when I run the topology checker, it often flags a variety of<br clear="none">> errors although I am careful to create polygons by only snapping to<br clear="none">> existing vertices of adjacent polygons. It does a great job of finding<br clear="none">> the errors, but I can find no instructions on methodology for repairing<br clear="none">> them.<br clear="none">><br clear="none">> Any suggesting on repairing topology errors easily?<br clear="none">><br clear="none">> Thank you, Cheers . . . . . . . . Spring Harrison<br clear="none">><br clear="none">><br clear="none">><br clear="none">> At 03-02-2016 07:51 Wednesday, Carlos Cerdán wrote:<br clear="none">>> Hi Springfield:<br clear="none">>><br clear="none">>> I've had similar trouble some time ago, as a QGIS beginner user.<br clear="none">>> Previous given advices are for not too-beginner users, but I think my<br clear="none">>> experience can be useful for the list (or even for you?):<br clear="none">>><br clear="none">>> That time my error was my Snapping options: I had selected two or<br clear="none">>> three layers and it was selected the "avoid intersections" option,<br clear="none">>> then, when I added new digitized polygons, they were splitted or<br clear="none">>> deleted (as you tell) because new polygons were intersecting marked<br clear="none">>> layers in snapping options. It fixed when new digitizing shape was the<br clear="none">>> one marked (of course to avoid intersections between her elements) .<br clear="none">>><br clear="none">>> Yes it was something simple, my foul. Is it not same trouble?. If so,<br clear="none">>> Check it in "Configuration --> Snapping options --> Change the<br clear="none">>> Self-assembling mode to "advanced" and be sure that "avoid<br clear="none">>> intersections" is marked in your target layer only.<br clear="none">>><br clear="none">>> Cheers,<br clear="none">>><br clear="none">>> Carlos Cerdán<br clear="none">>><br clear="none">>> 2016-02-02 20:57 GMT-05:00 Springfield Harrison <<a shape="rect" ymailto="mailto:stellargps@gmail.com" href="mailto:stellargps@gmail.com">stellargps@gmail.com</a><br clear="none">>> <mailto:<a shape="rect" ymailto="mailto:stellargps@gmail.com" href="mailto:stellargps@gmail.com">stellargps@gmail.com</a>> >:<br clear="none">>><br clear="none">>> Hello Group,<br clear="none">>><br clear="none">>>         Am using QGIS 2.10.1 and have done so without<br clear="none">>> significant problems for many months. I am attempting to create<br clear="none">>> simple polygons in a shapefile by direct vertex entry or copying<br clear="none">>> polygons from another shapefile layer and pasting them into the<br clear="none">>> active layer.<br clear="none">>><br clear="none">>>         In almost every case, these features disappear.Â<br clear="none">>> There is an entity entry in the attribute table but there are no<br clear="none">>> polygons created. In one case a polygon was created but it was<br clear="none">>> not the shape that I digitized; no more polygons could be created<br clear="none">>> although the process for doing so appears to be working. No error<br clear="none">>> messages were created. The CRS for the project and the shapefile<br clear="none">>> are the same.<br clear="none">>><br clear="none">>>         The shapefile can be opened in another GIS but has<br clear="none">>> no content.<br clear="none">>>        Â<br clear="none">>>         Here are some product details: QGIS versionÂ<br clear="none">>> 2.10.1-Pisa QGIS code revision d20c5b7<br clear="none">>> <<a shape="rect" href="https://github.com/qgis/QGIS/commit/d20c5b7" target="_blank">https://github.com/qgis/QGIS/commit/d20c5b7</a>><br clear="none">>><br clear="none">>>         Thanks very much for any help you can render,<br clear="none">>> Cheers . . . . . . . . Springfield Harrison<br clear="none">>><br clear="none">>> _______________________________________________<br clear="none">>> Qgis-user mailing list<br clear="none">>> <a shape="rect" ymailto="mailto:Qgis-user@lists.osgeo.org" href="mailto:Qgis-user@lists.osgeo.org">Qgis-user@lists.osgeo.org</a> <mailto:<a shape="rect" ymailto="mailto:Qgis-user@lists.osgeo.org" href="mailto:Qgis-user@lists.osgeo.org">Qgis-user@lists.osgeo.org</a>><br clear="none">>> List info: <a shape="rect" href="http://lists.osgeo.org/mailman/listinfo/qgis-user" target="_blank">http://lists.osgeo.org/mailman/listinfo/qgis-user</a><br clear="none">>> Unsubscribe: <a shape="rect" href="http://lists.osgeo.org/mailman/listinfo/qgis-user" target="_blank">http://lists.osgeo.org/mailman/listinfo/qgis-user</a><br clear="none">>><br clear="none">>><br clear="none">>> _______________________________________________ Qgis-user mailing list<br clear="none">>> <a shape="rect" ymailto="mailto:Qgis-user@lists.osgeo.org" href="mailto:Qgis-user@lists.osgeo.org">Qgis-user@lists.osgeo.org</a> List info:<br clear="none">>> <a shape="rect" href="http://lists.osgeo.org/mailman/listinfo/qgis-user" target="_blank">http://lists.osgeo.org/mailman/listinfo/qgis-user </a>Unsubscribe:<br clear="none">>> <a shape="rect" href="http://lists.osgeo.org/mailman/listinfo/qgis-user" target="_blank">http://lists.osgeo.org/mailman/listinfo/qgis-user</a><br clear="none">><br clear="none">><br clear="none">> __________ Information from ESET Mail Security, version of virus<br clear="none">> signature database 12996 (20160208) __________<br clear="none">><br clear="none">> The message was checked by ESET Mail Security.<br clear="none">> <a shape="rect" href="http://www.eset.com/" target="_blank">http://www.eset.com</a><br clear="none">><br clear="none">><br clear="none">> _______________________________________________<br clear="none">> Qgis-user mailing list<br clear="none">> <a shape="rect" ymailto="mailto:Qgis-user@lists.osgeo.org" href="mailto:Qgis-user@lists.osgeo.org">Qgis-user@lists.osgeo.org</a><br clear="none">> List info: <a shape="rect" href="http://lists.osgeo.org/mailman/listinfo/qgis-user" target="_blank">http://lists.osgeo.org/mailman/listinfo/qgis-user</a><br clear="none">> Unsubscribe: <a shape="rect" href="http://lists.osgeo.org/mailman/listinfo/qgis-user" target="_blank">http://lists.osgeo.org/mailman/listinfo/qgis-user</a><br clear="none">><br clear="none"><br clear="none"><br clear="none"><br clear="none">__________ Information from ESET Mail Security, version of virus signature database 12999 (20160209) __________<br clear="none"><br clear="none">The message was checked by ESET Mail Security.<br clear="none"><a shape="rect" href="http://www.eset.com/" target="_blank">http://www.eset.com</a><div class="yqt4417035118" id="yqtfd44986"><br clear="none"><br clear="none"><br clear="none">_______________________________________________<br clear="none">Qgis-user mailing list<br clear="none"><a shape="rect" ymailto="mailto:Qgis-user@lists.osgeo.org" href="mailto:Qgis-user@lists.osgeo.org">Qgis-user@lists.osgeo.org</a><br clear="none">List info: <a shape="rect" href="http://lists.osgeo.org/mailman/listinfo/qgis-user" target="_blank">http://lists.osgeo.org/mailman/listinfo/qgis-user</a><br clear="none">Unsubscribe: <a shape="rect" href="http://lists.osgeo.org/mailman/listinfo/qgis-user" target="_blank">http://lists.osgeo.org/mailman/listinfo/qgis-user</a></div><br><br></div> </div> </div> </div></div></body></html>