Add tasks for all the new languages

Task/Inverted-index/EchoLisp/inverted-index-1.echolisp

@@ -0,0 +1,23 @@
+;; set of input files
+(define FILES {T0.txt T1.txt T2.txt})
+;; store name for permanent inverted index
+(define INVERT "INVERTED-INDEX")
+
+;; get text for each file, and call (action filename text)
+(define (map-files action files)
+	(for ((file files))
+	(file->string action file)))
+	
+;; create store
+(local-make-store INVERT)
+
+; invert-word : word -> set of files
+(define (invert-word word file store)
+	    (local-put-value word
+	    	(make-set  (cons file (local-get-value word store))) store))
+	
+; parse file text and invert each word
+(define (invert-text file text)
+		(writeln 'Inverting file text)
+		(let ((text (text-parse text)))
+		(for ((word text))  (invert-word (string-downcase word) file INVERT))))

Task/Inverted-index/EchoLisp/inverted-index-2.echolisp

@@ -0,0 +1,11 @@
+;; usage : (inverted-search w1 w2 ..)
+(define-syntax-rule (inverted-search w ...)
+			(and-get-invert (quote w )))
+
+;; intersects all sets referenced by words
+;; returns the intersection
+(define (and-get-invert words)
+		(foldl
+			(lambda(word res)
+				 (set-intersect res (local-get-value word  INVERT)))
+			FILES words))

Task/Inverted-index/EchoLisp/inverted-index-3.echolisp

@@ -0,0 +1,9 @@
+(map-files invert-text FILES)
+(inverted-search is it)
+[0]→ { T0.txt T1.txt T2.txt }
+(inverted-search is banana)
+[1]→ { T2.txt }
+(inverted-search is what)
+[2]→ { T0.txt T1.txt }
+(inverted-search boule)
+[3]→ null

Task/Inverted-index/Phix/inverted-index.phix

@@ -0,0 +1,81 @@
+integer word_count = 0
+sequence filenames = {}
+
+function is_ascii(string txt)
+    for i=1 to length(txt) do
+        integer ch = txt[i]
+        if ch='\0' or ch>=#7F then return 0 end if
+    end for
+    return 1
+end function
+
+procedure add_words(string name, sequence words)
+    filenames = append(filenames,name)
+    integer fn = length(filenames)
+    for i=1 to length(words) do
+        string word = words[i]
+        if word[1]>='a'         -- skip numbers
+        and word[1]<='z' then
+            integer node = getd_index(word)
+            if node=0 then  -- not present
+                setd(word,{fn})
+                word_count += 1
+            else
+                sequence val = getd_by_index(node)
+                if find(fn,val)=0 then
+                    setd(word,append(val,fn))
+                end if
+            end if
+        end if
+    end for
+end procedure
+
+procedure load_directory()
+sequence d = dir(".")
+    for i=1 to length(d) do
+        if not find('d',d[i][D_ATTRIBUTES])     -- skip directories
+        and d[i][D_SIZE]<1024*1024*1024 then    -- and files > 1GB
+            string name = d[i][D_NAME]
+            integer fn = open(name,"rb")
+            string txt = lower(get_text(fn))
+            close(fn)
+            if is_ascii(txt) then               -- skip any bitmaps etc
+                sequence words = split_any(txt,"\0\r\n\t !\"#$%&\'()*+,-./:;<=>?@[\\]^`{|}~",0,1)
+                add_words(name,words)
+            end if
+        end if
+    end for
+end procedure
+
+function lookup(sequence words)
+sequence files = {}     -- indexes to filenames
+    for i=1 to length(words) do
+        string word = words[i]
+        integer node = getd_index(word)
+        if node=0 then return {} end if
+        sequence val = getd_by_index(node)
+        if i=1 then
+            files = val
+        else
+            for j=length(files) to 1 by -1 do
+                if not find(files[j],val) then
+                    files[j..j] = {}
+                end if
+            end for
+            if length(files)=0 then return {} end if
+        end if
+    end for
+    for i=1 to length(files) do
+        files[i] = filenames[files[i]]
+    end for
+    return files
+end function
+
+load_directory()
+?word_count
+?lookup({"load_directory"})     -- should only be this file
+?lookup({"dir"})                -- currently two use this
+?lookup({"lower"})              -- currently four use this
+?lookup({"lower","dir"})        -- currently two use both
+?lookup({"dir","lower"})        -- should be the same two
+?lookup({"ban"&"anafish"})      -- should be none ({})

Task/Inverted-index/jq/inverted-index-1.jq

@@ -0,0 +1,18 @@
+# Given an array of [ doc, array_of_distinct_words ]
+# construct a lookup table: { word: array_of_docs }
+def inverted_index:
+  reduce .[] as $pair
+    ({};
+     $pair[0] as $doc
+     | reduce $pair[1][] as $word
+       (.; .[$word] += [$doc]));
+
+def search(words):
+  def overlap(that): . as $this
+  | reduce that[] as $item ([]; if $this|index($item) then . + [$item] else . end);
+
+  . as $dict
+  | if (words|length) == 0 then []
+    else reduce words[1:][] as $word
+      ( $dict[words[0]]; overlap( $dict[$word] ) )
+    end ;

Task/Inverted-index/jq/inverted-index-2.jq

@@ -0,0 +1,8 @@
+def prompt_search:
+  "Enter a string or an array of strings to search for, quoting each string, or 0 to exit:",
+  ( (input | if type == "array" then . elif type == "string" then [.]
+             else empty
+             end) as $in
+    | search($in), prompt_search ) ;
+
+$in | inverted_index | prompt_search

Task/Inverted-index/jq/inverted-index-3.jq

@@ -0,0 +1,10 @@
+$ jq -r -c -n --argfile in <(jq -R 'split(" ") | select(length>0) | [input_filename, unique]' T?.txt) -f Inverted_index.jq
+Enter a string or an array of strings to search for, quoting each string, or 0 to exit:
+"is"
+["T0.txt","T1.txt","T2.txt"]
+Enter a string or an array of strings to search for, quoting each string, or 0 to exit:
+["is", "banana"]
+["T2.txt"]
+Enter a string or an array of strings to search for, quoting each string, or 0 to exit:
+0
+$