allow to override the standard ntriples fetch command (closes issue #8)

also some refactoring

CHANGELOG.md

@@ -6,6 +6,10 @@
 * document password retrieval for webid keystore
 * add a distinct option instead of new commands
 
+## 0.8.0 (unpublished)
+
+* allow to override the standard ntriples fetch command with alternatives such as any23's rover CLI
+
 ## 0.7.0 (May 2016)
 
 * new command: turtleize - outputs an RDF file in turtle, using as much as possible prefix declarations 

README.md

@@ -398,5 +398,7 @@ this is the place to setup personal configuration options such as
 * WebID support
 * syntax highlighting suppression
 * setup of preferred accept headers
+* setup of alternate ntriples fetch program such as any23's rover (see [this feature request](https://github.com/seebi/rdf.sh/issues/8) for background infos)
 
 Please have a look at the [example rc file](https://github.com/seebi/rdf.sh/blob/master/example.rc).
+

example.rc

@@ -1,12 +1,18 @@
-# this file can be copied to you rdf.sh config directory
+# this file can be copied to your rdf.sh configuration directory
 # (e.g. to $HOME/.config/rdf.sh/rc) and is imported on each run
 
 # which accept header should be used for requests
 #RDFSH_ACCEPT_HEADER="text/turtle; q=1.0, application/x-turtle; q=0.9, text/n3; q=0.8, application/rdf+xml; q=0.5, text/plain; q=0.1"
 
-# suppress highlightning? (true or false)
+# suppress highlighting? (true or false)
 #RDFSH_HIGHLIGHTING_SUPPRESS="true"
 
 ### add some additional curl options:
-# allow webid authentification and ignores ssl server cert errors
+# allow webid authentication and ignores ssl server cert errors
 #RDFSH_CURLOPTIONS_ADDITONS="-E $HOME/.webid.pem --insecure"
+
+# allow for overriding the default ntriples curl fetch command
+# the command needs to ouput ntriples to stdout, stderr is suppressed
+# the command takes one parameter (the URL of the resource)
+RDFSH_HTTPGETNTRIPLES_COMMAND="any23 rover -t -f ntriples"
+

rdf

@@ -56,19 +56,6 @@ else
     mimetypes="$RDFSH_ACCEPT_HEADER"
 fi
 
-# check for actively suppressing the pygmentize highlighting
-if [ "$RDFSH_HIGHLIGHTING_SUPPRESS" == "true" ]
-then
-    highlight=""
-else
-    echo 'ttt'| pygmentize -l turtle 2>/dev/null >/dev/null
-    if [ "$?" == "0" ]; then
-        highlight="yes"
-    else
-        highlight=""
-    fi
-fi
-
 ### mac workarounds
 uname=`uname`
 if [ "$uname" == "Darwin" ]
@@ -85,7 +72,50 @@ fi
 # private functions
 ###
 
-# takes an input command name and checks for availability with which
+# outputs a given input turtle file
+_outputTurtle ()
+{
+    turtleFile=$1
+    if [ "$turtleFile" == "" ]
+    then
+        echo "_outputTurtle error: need an parameter"
+        exit 1
+    fi
+
+    # check for actively suppressing the pygmentize highlighting
+    if [ "$RDFSH_HIGHLIGHTING_SUPPRESS" == "true" ]
+    then
+        cat $turtleFile
+    else
+        # if hightlight is enable, try to pipe it through pygmentize, otherwise just cat it
+        echo 'ttt'| pygmentize -l turtle 2>/dev/null >/dev/null
+        if [ "$?" == "0" ]; then
+            cat $tmpfile.ttl | pygmentize -l turtle
+        else
+            cat $turtleFile
+        fi
+    fi
+}
+
+# fetches an URL as ntriples and outputs it to stdout
+_httpGetNtriples ()
+{
+    url=$1
+    if [ "$url" == "" ]
+    then
+        echo "_httpGetNtriples error: need an parameter"
+        exit 1
+    fi
+
+    if [ "$RDFSH_HTTPGETNTRIPLES_COMMAND" == "" ]
+    then
+        $curlcommand -H "Accept: $mimetypes" $url | rapper -q -i guess -o ntriples -I $url -
+    else
+        $RDFSH_HTTPGETNTRIPLES_COMMAND $url 2>/dev/null
+    fi
+}
+
+# takes an input command name and checks for availability
 _checkTool ()
 {
     tool=$1
@@ -426,7 +456,7 @@ _getRelatedResources ()
     uri=`_expandQName $resource`
 
     tmpfile=`_getTempFile`
-    $thisexec get-ntriples $uri >$tmpfile.nt
+    _httpGetNtriples $uri >$tmpfile.nt
     roqet -q -e "SELECT DISTINCT ?o {<$uri> ?property ?o. FILTER(isUri(?o))}" -D $tmpfile.nt 2>/dev/null | cut -d "<" -f 2 | cut -d ">" -f 1 | grep "^http"
     rm $tmpfile $tmpfile.nt
 }
@@ -448,7 +478,7 @@ _isPingbackEnabled ()
     if [ "$pingbackServer" == "" ]
     then
         tmpfile=`_getTempFile`
-        $thisexec get-ntriples $uri >$tmpfile.nt
+        _httpGetNtriples $uri >$tmpfile.nt
         pingbackServer=`roqet -q -e "SELECT ?o {<$uri> <http://purl.org/net/pingback/to> ?o}" -D $tmpfile.nt 2>/dev/null | head -1 | cut -d "<" -f 2 | cut -d ">" -f 1`
         rm $tmpfile $tmpfile.nt
     fi
@@ -505,7 +535,7 @@ do_edit ()
     uri=`_expandQName $uri`
     doEditTmpfile=`_getTempFile`
 
-    $thisexec get-ntriples $uri >${doEditTmpfile}
+    _httpGetNtriples $uri >${doEditTmpfile}
     features=`_getFeatures ${doEditTmpfile}`
     cat ${doEditTmpfile} | rapper -q ${features} -i ntriples -o turtle -I $uri - >"${doEditTmpfile}.ttl"
     $EDITOR $doEditTmpfile.ttl
@@ -582,22 +612,17 @@ do_desc ()
     fi
     uri=`_expandQName $uri`
     tmpfile=`_getTempFile`
-    $thisexec get-ntriples $uri >$tmpfile.nt
-    roqet -q -e "CONSTRUCT {<$uri> ?p ?o} WHERE {<$uri> ?p ?o}" -D $tmpfile.nt -r turtle | rapper -i turtle -o ntriples -q - "$uri" >$tmpfile.out
-
-    features=`_getFeatures $tmpfile.out`
+    _httpGetNtriples $uri >$tmpfile
 
-    rapper -q ${features} -i ntriples $tmpfile.out -o $output >$tmpfile.ttl
+    # fetches only triples with URI as subject (output is turtle)
+    roqet -q -e "CONSTRUCT {<$uri> ?p ?o} WHERE {<$uri> ?p ?o}" -D $tmpfile >$tmpfile.out
 
-    # if hightlight is enable, pipe it through pygmentize, otherwise just cat it
-    if [ "$highlight" == "yes" ]; then
-        cat $tmpfile.ttl | pygmentize -l turtle
-    else
-        cat $tmpfile.ttl
-    fi
+    # reformat and output turtle file
+    $thisexec turtleize $tmpfile.out >$tmpfile.ttl
+    _outputTurtle $tmpfile.ttl
 
     # clean up
-    rm $tmpfile $tmpfile.out $tmpfile.nt $tmpfile.ttl
+    rm -f $tmpfile $tmpfile.out $tmpfile.ttl
 
     # add history
     _addToHistory $uri $historyfile
@@ -616,14 +641,12 @@ do_list ()
     fi
     uri=`_expandQName $uri`
     tmpfile=`_getTempFile`
-    # turn history off for this internal call (dirty URIs)
-    export noHistory="true"
-    $thisexec get-ntriples $uri >$tmpfile.nt
+    _httpGetNtriples $uri >$tmpfile.nt
     roqet -q -e "SELECT DISTINCT ?s WHERE {?s ?p ?o. FILTER isURI(?s) } " -D $tmpfile.nt 2>/dev/null | cut -d "<" -f 2 | cut -d ">" -f 1 | grep $uri
     rm $tmpfile $tmpfile.nt
 }
 
-docu_get () { echo "curls rdf in xml or turtle to stdout (tries accept header)"; }
+docu_get () { echo "fetches an URL as RDF to stdout (tries accept header)"; }
 do_get ()
 {
     _checkTool curl
@@ -635,7 +658,10 @@ do_get ()
         exit 1
     fi
     uri=`_expandQName $uri`
-    $curlcommand -H "Accept: $mimetypes" $uri
+    tmpfile=`_getTempFile`
+    _httpGetNtriples $uri >$tmpfile.nt
+    $thisexec turtleize $tmpfile.nt
+    rm $tmpfile $tmpfile.nt
     _addToHistory $uri $historyfile
 }
 
@@ -651,7 +677,7 @@ do_get-ntriples ()
         exit 1
     fi
     uri=`_expandQName $uri`
-    $curlcommand -H "Accept: $mimetypes" $uri | rapper -q -i guess -o ntriples -I $uri -
+    _httpGetNtriples $uri
     _addToHistory $uri $historyfile
 }
 

rdf.1

@@ -1,4 +1,4 @@
-.TH rdf.sh 1  "05/2016" "version 0.7.0" "USER COMMANDS"
+.TH rdf.sh 1  "05/2016" "version 0.8.0" "USER COMMANDS"
 .SH NAME
 rdf.sh \- A multi-tool shell script for doing Semantic Web jobs on the command line.
 .SH SYNOPSIS