Hallo, We did some test to use RDF (also Miklos his advice) to store some information about images in a Writerdoc, before we stored the information in the ImageObject , what was not a good idea. Greetz Fernand
below the code who works well const BASE_TYPE = "http://PMGopmaak.writer/Allgraphics";<http://PMGopmaak.writer/Allgraphics> const TYPE_NAME = "GraphicURL" const FILE_NAME = "PMGopmaak.writer/Allgraphics.rdf" Sub StoreRDF RemoveRDF() ' "graph" staat hier niet voor een graphic maar voor een ellement die data kan omschrijven en stockeren ' een "litera" is een string die we stockeren en kunnen opvragen doc = ThisComponent repo = doc.getRDFRepository() type_uri = GetTypeURI() graph_name = doc.addMetadataFile(FILE_NAME, Array(type_uri)) named_graph = repo.getGraph(graph_name) value_uri = CreateURI(com.sun.star.rdf.URIs.RDF_VALUE) if doc.supportsService("com.sun.star.text.TextDocument") then oDocGraphics = doc.GraphicObjects vAlleFotos = oDocGraphics.getElementNames() aantal = oDocGraphics.count For iG = 0 to aantal - 1 iFotNr = ig + 1 ' sFotMetNr = vAlleFotos(iG) oGraphic1 = oDocGraphics.GetbyIndex(ig) named_graph.addStatement(CreateURI(oGraphic1.Name &":", "URL"),value_uri,CreateLiteral(Ographic1.Title)) named_graph.addStatement(CreateURI(oGraphic1.Name &":", "ONDERSCHRIFT"),value_uri,CreateLiteral(Ographic1.Hyperlinkname)) next IG End If doc.setModified(True) End Sub Sub ReadRDF() zoek = "Image1:URL" doc = ThisComponent repo = doc.getRDFRepository() type_uri = GetTypeURI() graph_names = doc.getMetadataGraphsWithType(type_uri) If ubound(graph_names) = -1 then exit sub graph_name = graph_names(0) named_graph = repo.getGraph(graph_name) value_uri = CreateURI(com.sun.star.rdf.URIs.RDF_VALUE) ' named_graph.clear enume = named_graph.getStatements(Null, value_uri, Null) ' xray named_graph.getStatements(null, value_uri, Null) s = "" Do while Enume.hasMoreElements st = enume.nextElement() name = st.Subject.StringValue sts = st.Object.StringValue s = s & name & ": " & sts & chr(10) print name ' if name = zoek then ' print sts ' named_graph.removeStatements(st.Subject, value_uri,st.Object) ' exit do ' endif loop msgbox s End Sub Sub RemoveRDF doc = ThisComponent repo = doc.getRDFRepository() type_uri = GetTypeURI() graph_names = doc.getMetadataGraphsWithType(type_uri) if ubound(graph_names) = -1 then exit sub graph_name = graph_names(0) doc.removeMetadataFile(graph_name) doc.setModified(True) End Sub Function GetTypeURI() As variant GetTypeURI = CreateURI(BASE_TYPE & "/" & TYPE_NAME) End Function Function CreateURI(n, Optional l) As Variant If IsMissing(l) Then a = Array(n) Else a = Array(n, l) End If uri = CreateUnoServiceWithArguments("com.sun.star.rdf.URI", a) CreateURI = uri End Function Function CreateLiteral(v) As Variant a = Array(v) literal = CreateUnoServiceWithArguments("com.sun.star.rdf.Literal", a) CreateLiteral = literal End Function On 1/11/2019 9:27 AM, Miklos Vajna wrote: Hi Vincent, On Thu, Jan 10, 2019 at 02:13:58PM +0000, LORENZO Vincent <vincent.lore...@cea.fr><mailto:vincent.lore...@cea.fr> wrote: Question 1: Currently we are studying the possibility to use the Bookmark field's name to store these datas. We want to use the Bookmark with start and end tag, to get something like that. <text:bookmark-start text:name="UML_element_qualified_name_and_others_useful_informations" /> Some text <text:bookmark-end text:name=" UML_element_qualified_name_and_others_useful_informations " /> In reality, I would prefer to be able to use a more structured datas, extending the Bookmark object (or using another solution ? ) to get that, something like that: <text:bookmark-start text:name="Bookmark1" umlElementQualifiedName="Package1::Class1" umlElementKind="Class" umlElementfeature="name" id="xxx"/> Some text <text:bookmark-end text:name="Bookmark1" /> but I have no idea if it is possible and how to do that. Please, does someone have an idea to structure my datas ? Of course, if you have a better solution than Bookmark, fell free to propose it. Bookmark is just a string; one hack you can do is to give it a unique name, which refers to a per-document custom metadata, where the value disturbs the user less. But that's still just a string, so you need to serialize/load to/from string if you want to store key-value pairs. An alternative is to use RDF annotations, which allows more complex structures: https://wiki.openoffice.org/wiki/Documentation/DevGuide/OfficeDev/RDF_metadata (Though you need to use archive.org or something like that to access it, seems the Apache site is currently not available.) One shortcoming of RDF is that it's ODF-only; while the per-doc custom metadata + bookmarks works with most other formats as well. So, in this example, I write text, I look for it, then I create a XTextRange and at the end, I can create the Bookmark. I'm not very have happy with this solution. I would prefer a solution without the search step, but I don't know how to get a valid XTextRange (I admit I don't yet well understand the behavior of the API). Please, do you have some pointer or code example ? Some kind of iteration or search is necessary: if you take the user's current selection or you iterate to the nth paragraph or you search for text is up to you. An XTextRange is just a selection (where start and end may point to the same document model position). I thank you to read me until here. Hope this helps. :-) Regards, Miklos _______________________________________________ LibreOffice mailing list LibreOffice@lists.freedesktop.org<mailto:LibreOffice@lists.freedesktop.org> https://lists.freedesktop.org/mailman/listinfo/libreoffice
_______________________________________________ LibreOffice mailing list LibreOffice@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/libreoffice
