Fixes @prefix counting (in case of 0) & adds new cmd 'nssort' & minor things

README.md

@@ -146,6 +146,7 @@ rdf.sh currently provides these subcommands:
 * nscollect: collects prefix declarations of a list of ttl/n3 files
 * nsdist: distributes prefix declarations from one file
   to a list of other ttl/n3 files
+* nssort: sorts the prefix declarations in files
 * put: replaces an existing linked data resource via LDP
 * split: split an RDF file into pieces of max X triple and output the file names
 * turtleize: outputs an RDF file in turtle,

rdf

@@ -969,7 +969,7 @@ do_nscollect()
         countBefore=0
     fi
 
-    files=($(find . -name "*.ttl" | grep -v "$prefixfile"))
+    mapfile -t files < <(find . -name "*.ttl" | grep -v "$prefixfile")
     rm -f "$prefixfile"
     for file in "${files[@]}"
     do
@@ -1000,24 +1000,29 @@ do_nsdist ()
 
     if [ "${2:-}" == "" ]
     then
-        files=($(find . -name "*.ttl" | grep -v "$prefixfile"))
+        mapfile -t files < <(find . -name "*.ttl" | grep -v "$prefixfile")
     else
         files=("$@")
     fi
 
+    _count_prefixes() {
+        local ttl_file
+        ttl_file="$1"
+
+        grep "@prefix " -r "$ttl_file" 2> /dev/null | wc -l || true
+    }
+
     count=$(wc -l < "$prefixfile")
     tmpfile=$(_getTempFile)
     for target in "${files[@]}"
     do
         if [ -f "$target" ]
         then
-            # shellcheck disable=SC2126
-            before=$(grep "@prefix " < "$target" | wc -l || echo 0)
-            grep -v "@prefix " < "$target" >"$tmpfile" || true
-            cat $prefixfile >"$target"
-            cat "$tmpfile" >>"$target"
-            # shellcheck disable=SC2126
-            after=$(grep "@prefix " < "$target" | wc -l || echo 0)
+            before=$(_count_prefixes "$target")
+            grep -v "@prefix " < "$target" > "$tmpfile" || true
+            cat $prefixfile > "$target"
+            cat "$tmpfile" >> "$target"
+            after=$(_count_prefixes "$target")
             result=$((after - before)) || true
             if [ "$result" -ge "0" ]
             then
@@ -1030,6 +1035,76 @@ do_nsdist ()
     rm "$tmpfile"
 }
 
+docu_nssort () { echo "sorts the prefix declarations in files, re-inserting all at the line of the first found prefix";}
+do_nssort ()
+{
+    local before prefixes after
+    local -a files
+
+    _checkTool awk mktemp
+
+    if [ "${2:-}" == "" ]
+    then
+        >&2 echo "Syntax: $this $command <target-turtle-files>"
+        >&2 echo "($(docu_nssort))"
+        exit 1
+    else
+        files=("$@")
+    fi
+
+    tmpfile=$(_getTempFile)
+    for target in "${files[@]}"
+    do
+        if [ -f "$target" ]
+        then
+            awk '
+BEGIN {
+    nBef = 1
+    nPref = 1
+    nAft = 1
+}
+
+/^@prefix/ {
+    lStart = sprintf("%s %-8s", $1, $2)
+    $1=$2=""
+    aPref[nPref++] = sprintf("%s%s", lStart, $0)
+    # skips processing this line further
+    next
+}
+
+{
+    if (nPref < 2) {
+        aBef[nBef++] = $0
+    } else {
+        aAft[nAft++] = $0
+    }
+}
+
+END {
+    # sorts the prefixes
+    # NOTE This is a GNU AWK (gawk) function.
+    #      If your AWK does not suport it, install gawk
+    #      or add an asort implemention here in the code.
+    nPref = asort(aPref)
+    # Do not know why, but this is necessary
+    nBef--
+    nAft--
+    for (i = 1; i <= nBef; i++) {
+        print aBef[i]
+    }
+    for (i = 1; i <= nPref; i++) {
+        print aPref[i]
+    }
+    for (i = 1; i <= nAft; i++) {
+        print aAft[i]
+    }
+}' < "$target" > "$tmpfile"
+             mv "$tmpfile" "$target"
+            echo "$target: prefixes sorted"
+        fi
+    done
+}
+
 docu_turtleize() { echo "outputs an RDF file in turtle, using as much as possible prefix declarations"; }
 do_turtleize ()
 {
@@ -1077,7 +1152,7 @@ do_skolemize ()
     tmpOutput="/tmp/$RANDOM-$(basename "$file")"
     rapper -i guess "$file" 2>/dev/null | sort -u >"$tmpInput" || true
 
-    # in order to have stable uuids with the same file, change seed to md5sum of the file
+    # in order to have stable UUIDs with the same file, change seed to md5sum of the file
     seed="$tmpInput/"
     awk -v domain="$domain" -v seed="$seed" '
         BEGIN{FS=" "; OFS=" "; ORS=""}

rdf.1

@@ -25,6 +25,7 @@ This includes inspecting resources and namespaces, listing of schema elements an
     ns              -- curls the namespace from prefix.cc
     nscollect       -- collects prefix declarations of a list of ttl/n3 files
     nsdist          -- distributes prefix declarations from one file to a list of other ttl/n3 files
+    nssort          -- sorts the prefix declarations in files, inserting all at the line of the first found prefix
     ping            -- sends a semantic pingback request from a source to a target or to all possible targets
     pingall         -- sends a semantic pingback request to all possible targets of a given resource
     put             -- replaces an existing linked data resource via LDP