On Thu, 29 Aug 2013 05:50:11 -0700, Timothy W. Cook <[email protected]> wrote:
> Hi All, > > I am new to ML and I'm designing a project using a developer license > for ML6 on CentOS. > > My XML data is schema validated, Each schema 'includes' a parent > schema from which it constrains the complexTypes from the parent. > > The design of the eco-system is such that the namespace; mlhim2 is > consistent across all versions of the parent schema. Only the namespce > of the parent (and physical location) changes. > > For example: > xmlns:mlhim2="http://www.mlhim.org/xmlns/mlhim2/2_4_0"; > xmlns:mlhim2="http://www.mlhim.org/xmlns/mlhim2/2_4_1"; > xmlns:mlhim2="http://www.mlhim.org/xmlns/mlhim2/2_4_2"; > > all point to a schema named mlhim2.xsd > > I use a catalog.xml in other tools to resolve which local mlhim2.xsd > is referenced by the child schemas. > > I see that using the Flow editor I can set the URI to something like: > /content/mlhim2/2_4_x" for each schema and import them. But how do I > tell ML that the parent schema locally resides here when the child > does the include? Just to make sure I am understanding your setup properly: schema A1: namespace A1 imports schema B namespace http://www.mlhim.org/xmlns/mlhim2/2_4_1 schema A2: namespace A2 imports schema B namespace http://www.mlhim.org/xmlns/mlhim2/2_4_2 etc. Right? The easiest answer, if there is only one schema document for each of the http://www.mlhim.org/xmlns/mlhim2/... namespaces, is to just not put a schema location in the import at all. MarkLogic will look up the schema via its namespace, and will find it as long as it in is in the schemas database. If you do have a location on the import, then it matters whether it is abolute (http://whatever) or relative ("mlhim2.xsd"). An absolute URI will find that schema at the given URI in the schemas database. So if you said the location was "http://www.example.org/ns1/mlhim2.xsd"; then doc("http://www.example.org/ns1/mlhim2.xsd";) (in the schemas database) is where we go looking for the schema. If the URI on the import is relative, then we will use the base uri of the importing schema to resolve that URI. Suppose schema A1 were at "/ns1/A1.xsd" and schema A2 were at "/ns2/A2.xsd" and the import said simply "mlhim2.xsd", then A1 would look for the schema at "/ns1/mlhim2.xsd" and A2 would look for the schema at "/ns2/mlhim2.xsd". Where this gets tricky is if you have the base schemas at some relative location, such as "A1.xsd". Resolving a relative URI against a relative path ends up not working as you'd expect (in this case I believe we'd end up looking at "/mlhim2.xsd"), and we don't recommend that practice. //Mary _______________________________________________ General mailing list [email protected] http://developer.marklogic.com/mailman/listinfo/general
