repare first steps

doc/first_steps/fuseki.config.ttl

@@ -29,7 +29,7 @@ PREFIX ja:      <http://jena.hpl.hp.com/2005/11/Assembler#>
     ] ;
     fuseki:endpoint [ 
         fuseki:operation fuseki:gsp-rw ; 
-        fuseki:name "data"
+        fuseki:name "store"
     ] ; 
     fuseki:endpoint [ 
         fuseki:operation fuseki:patch ;

doc/first_steps/fuseki.yaml

@@ -1,10 +1,14 @@
 version: kgsteward_yaml_2
 server:
-  brand            : fuseki
+  type             : fuseki
   location         : http://localhost:3030
   server_config    : fuseki.config.ttl # actually unused
   repository       : first_steps
-  file_server_port : 8003
+context_base_IRI: http://example.org/context/
+file_loader:
+  type: store_chunks
+url_loader:
+  type: sparql_load
 dataset: !include dataset.yaml
 validations: 
   - sparql/validate/*.rq

doc/first_steps/graphdb.yaml

@@ -1,11 +1,16 @@
 version: kgsteward_yaml_2
 server:
-  brand         : graphdb
+  type         : graphdb
   location      : http://localhost:7200
   server_config : graphdb.config.ttl
   username      : ${GRAPHDB_USERNAME}
   password      : ${GRAPHDB_PASSWORD}
   repository    : first_steps
+context_base_IRI: http://example.org/context/
+file_loader:
+  type: sparql_load
+url_loader:
+  type: sparql_load
 dataset: !include dataset.yaml
 validations: 
   - sparql/validate/*.rq

doc/first_steps/rdf4j.yaml

@@ -1,10 +1,14 @@
 version: kgsteward_yaml_2
 server:
-  brand            : rdf4j
+  type             : rdf4j
   location         : http://localhost:8080/rdf4j-server
   server_config    : rdf4j.config.ttl 
   repository       : kgsteward_demo_rdf4j
-  file_server_port : 8002
+context_base_IRI: http://example.org/context/
+file_loader:
+  type: store_chunks
+url_loader:
+  type: sparql_load
 dataset: !include dataset.yaml
 validations:
   - sparql/validate/*.rq

src/kgsteward/generic.py

@@ -2,6 +2,7 @@
 # that might work on a real triplestore (maybe?)
 
 import subprocess
+import urllib
 
 # How to dump a context from the command line:
 # curl -G --data-urlencode "graph=http://example.org/context/ReconX_schema" http://localhost:7200/repositories/ReconXKG2/rdf-graphs/service
@@ -121,7 +122,7 @@ def _flush_buf( self, context, data, headers = {}, echo = True ):
         http_call(
             {
                 'method'  : 'POST',
-                'url'     : self.endpoint_store + "?graph=" + context,
+                'url'     : self.endpoint_store + "?graph=" + urllib.parse.quote_plus( context ),
                 'headers' : {
                     **headers,
                     'Content-Type' : 'text/plain',
@@ -133,7 +134,6 @@ def _flush_buf( self, context, data, headers = {}, echo = True ):
             echo
     )
 
-
     def load_from_file_using_riot( self, file, context, headers = {}, echo = True ):
         """ use graph store protocol and riot """
         if echo:

src/kgsteward/kgsteward.py

@@ -421,16 +421,16 @@ def main():
     }}
 }}""" )
         if "file" in target :
-            if config["server"]["file_loader"]["type"] == "http_server":
-                fs = LocalFileServer( port = config["server"]["file_loader"][ "port" ] )
+            if config["file_loader"]["type"] == "http_server":
+                fs = LocalFileServer( port = config["file_loader"][ "port" ] )
                 for path in target["file"] :
                     for dir, fn in expand_path( path, config["kgsteward_yaml_directory"] ):
                         if not os.path.isfile( dir + "/" + fn ):
                             stop_error( "File not found: " + dir + "/" + fn )
                         fs.expose( dir ) # cache already exposed directory
                         if not is_running_in_a_container: 
                             try:
-                                path = "http://localhost:" + str( config["server"]["file_loader"][ "port" ] ) + "/" + fn
+                                path = "http://localhost:" + str( config["file_loader"][ "port" ] ) + "/" + fn
                                 server.sparql_update( f"LOAD <{path}> INTO GRAPH <{context}>" )
                             except Exception as e:
                                 msg = str( e )
@@ -451,16 +451,16 @@ def main():
     }}
 }}""" )
                 fs.terminate()
-            else: # config["server"]["file_loader"]["type"] != "http_server"
+            else: # config["file_loader"]["type"] != "http_server"
                 for path in target["file"] :
                     for dir, fn in expand_path( path, config["kgsteward_yaml_directory"] ):
                         filename = dir + "/" + fn
-                        if config["server"]["file_loader"]["type"] == "sparql_load":
+                        if config["file_loader"]["type"] == "sparql_load":
                             server.sparql_update( f"LOAD <file://{filename}> INTO GRAPH <{context}>" )
-                        elif config["server"]["file_loader"]["type"] == "chunked_store_file":
+                        elif config["file_loader"]["type"] == "store_chunks":
                             server.load_from_file_using_riot( filename, context )
                         else:
-                            raise SystemError( "Unexpected file loader type: " + config["server"]["file_loader"]["type"] )
+                            raise SystemError( "Unexpected file loader type: " + config["file_loader"]["type"] )
                         server.sparql_update( f"""PREFIX void: <http://rdfs.org/ns/void#>
 INSERT DATA {{
     GRAPH <{context}> {{

src/kgsteward/yamlconfig.py

@@ -154,15 +154,6 @@ class DatasetConf( BaseModel ):
     update   : Optional[ list[ str ]] = Field( None,  title = "SPARQL update file(s)", description = describe(  "update" ))
     zenodo   : Optional[ list[ int ]] = Field( None,  title = "Ignore me", description = describe(  "zenodo" ))
 
-class KGStewardConf( BaseModel ):
-    model_config = ConfigDict( extra='allow' )
-    version           : Literal[ "kgsteward_yaml_2" ] = Field( title = "YAML syntax version", description = "This mandatory fixed value determines the admissible YAML syntax" )
-    server            : Union[ GraphDBConf, RDF4JConf, FusekiConf ]
-    dataset           : list[ DatasetConf ] = Field( title = "Knowledge Graph content", description = describe( "dataset" ))
-    context_base_IRI  : str = Field( title = "context base IRI", description = describe( "context_base_IRI" ) )
-    queries           : Optional[ list[ str ]]  = Field( None, title = "GraphDB queries", description = describe( "queries" ))
-    validations       : Optional[ list[ str ]]  = Field( None, title = "Validation queries", description = describe( "validations" ))
-
 class DirectFileLoader( BaseModel ):
     type : Literal[ "sparql_load" ] = Field( title = "direct file_loader", description = describe( "direct_file_loader" ) )
 
@@ -171,17 +162,28 @@ class HttpServerFileLoader( BaseModel ):
     port : Optional[ int ] = Field( 8000, title = "file_server_port", description = describe( "file_server_port" ))
 
 class ChunkedStoreFileLoader( BaseModel ):
-    type : Literal[ "chunked_store" ] = Field( title = "HTTP file server", description = "http_file_server" )
-    port : Optional[ int ] = Field( 1e8, title = "chunked_store", description = "chunked_store_file" )
+    type : Literal[ "store_chunks" ] = Field( title = "HTTP file server", description = "http_file_server" )
+    size : Optional[ int ] = Field( 100_000_000, title = "chunked_store", description = "chunked_store_file" )
 
 class DirectUrlLoader( BaseModel ):
-    type : Literal[ "load_url" ] = Field( title = "direct url loader", description = describe( "direct_url_loader" ))
+    type : Literal[ "sparql_load" ] = Field( title = "direct url loader", description = describe( "direct_url_loader" ))
 
 class CurlChunkedStoreUrlLoader( BaseModel ):
     type : Literal[ "curl_chunked_store_url_loader" ] = Field( title = "direct url loader", description = "direct_url_loader" )
     tmp_dir : Optional[ str ] = Field( "/tmp", title = "temporary directory", description = "temporary directory" )
     port : Optional[ int ] = Field( 1e8, title = "chunked_store", description = "chunked_store_file")
 
+class KGStewardConf( BaseModel ):
+    model_config = ConfigDict( extra='allow' )
+    version           : Literal[ "kgsteward_yaml_2" ] = Field( title = "YAML syntax version", description = "This mandatory fixed value determines the admissible YAML syntax" )
+    server            : Union[ GraphDBConf, RDF4JConf, FusekiConf ]
+    file_loader       : Union[ DirectFileLoader, HttpServerFileLoader, ChunkedStoreFileLoader ]
+    url_loader        : Union[ DirectUrlLoader ]
+    dataset           : list[ DatasetConf ] = Field( title = "Knowledge Graph content", description = describe( "dataset" ))
+    context_base_IRI  : str = Field( title = "context base IRI", description = describe( "context_base_IRI" ) )
+    queries           : Optional[ list[ str ]]  = Field( None, title = "GraphDB queries", description = describe( "queries" ))
+    validations       : Optional[ list[ str ]]  = Field( None, title = "Validation queries", description = describe( "validations" ))
+
 def parse_yaml_conf( path : str ):
     """Parsing kgsteward YAML config file(s) is a three step process:
     1. Parse YAML file(s) and execute !include directive