Where are new metadata entry XMLs stored?

[MapZombie]

As the new Geoportal does not use a database, I was wondering where the user created metadata (through metadata editor) files stored?

When a user creates metadata is it stored as an XML file on the server? If not, can this be configured to do so?

Thank you.

[umacgillivray]

The xml files are stored within ElasticSearch. Within the index there is an es type "item", and another type "clob". The xmls are within the "clob" type.

[MapZombie]

Thanks @umacgillivray . Would you know of an effective way of exporting these from ElasticSearch?

[mhogeweg]

to where?

[umacgillivray]

Java from within geoportal would look something like this `

import com.esri.geoportal.lib.elastic.util.Scroller; import java.nio.file.Files; import java.nio.file.Path; import java.nio.file.Paths; import java.util.concurrent.atomic.AtomicLong; import java.util.function.Consumer; import org.elasticsearch.search.SearchHit;

public static void testExport() throws Exception { AtomicLong count = new AtomicLong(0); String dir = "C:/Projects/elastic/metadata/sampledata2/"; Scroller scroller = new Scroller(); scroller.setIndexName("metadata"); scroller.setIndexType("clob"); //scroller.setMaxDocs(10); scroller.scroll( new Consumer(){ @Override public void accept(SearchHit hit) { try { System.err.println(count.incrementAndGet()); String id = hit.getId(); String xml = (String)hit.getSource().get("sys_clob"); //System.err.println(id); Path p = Paths.get(dir+id.replace("_xml","")+".xml"); Files.write(p,xml.getBytes("UTF-8")); } catch (Exception e) { e.printStackTrace(); } } } ); } `

[MapZombie]

Basically I'm just looking to export these to the local geoportal server as individual XMLs.

[MapZombie]

I appreciate the help @umacgillivray . Would you be able to guide me as to where I would enter this code? Which file(s) would need to be edited? I am not very proficient in Java unfortunately.

Thanks.

[MapZombie]

Can XMLs be exported server side via editing the JavaScript? I'm getting a grasp of the save function. And looking at whether we can add a function to save to a folder as ".xml" file and elasticsearch as usual.

MetadataEditor.js

_save: function(xmlString,itemId) {

      var client = new AppClient();
      var url = client.getRestUri()+"/metadata/item";
      if (typeof itemId === "string" && itemId.length > 0) {
        url += "/"+encodeURIComponent(itemId);  /* item name id */
      }
      url = client.appendAccessToken(url);     /* Elasticsearch URL? */
      var data = {                            /* Defines Data */
        app_editor_s: "gxe",                  /* Data is from GXE */
        xml: xmlString                        /* XML String */
      };
      var headers = {"Content-Type": "application/json"}; /* Header */
      var info = {handleAs:"json",headers:headers,data:JSON.stringify(data)}; /* Handles as JSON format */
      return dojoRequest.put(url,info);    /* push to server */
    },

[valentinedwv]

Would this be better done as an enhancement to the geoportal-harvester?

Connect to a geoportal 2.0, export to folder.

[mhogeweg]

yes! especially since harvesting to a folder is already supported. you would create an input broker for Geoportal Server Catalog 2.x.

adding @pandzel in case he has already worked on such a broker for harvester