Data update

Task/Anagrams/DuckDB/anagrams.duckdb

@@ -0,0 +1,20 @@
+# The MAP giving the frequency counts of characters in the given string
+create or replace function spectrum(str) as (
+  select histogram(c)
+  from (select unnest(regexp_extract_all(str,'.')) as c)
+);
+
+# Find the anagram groups having the most members.
+# Each group is sorted, and the groups are sorted by first word in the group.
+with words as (from read_csv('unixdict.txt', header=false) _(word)),
+  histograms as (select word, spectrum(word) as h from words),
+  groups as (select h, count(h) as c from histograms group by h order by c desc),
+  mx as (select max(c) as mx from groups),
+  maximals as (select h from groups, mx where c = mx.mx),
+  results as (select (select array_agg(word).list_sort()
+                      from histograms
+                      where histograms.h = maximals.h ) as anagrams
+              from maximals)
+  select anagrams
+  from results
+  order by anagrams[1] ;

Task/Anagrams/TAV/anagrams.tav

@@ -0,0 +1,38 @@
+\(  Find anagrams (words with the same set of letters)
+    Words are the lines of an input file.
+    Sorting the letters of the word gives the key for a map
+    having a list (row) of all words so far.
+\)
+\+ stdlib
+main(parms):+
+    parms[1] =~ 'unixdict.txt'         \ default input file
+    wordmap =: get words from parms[1]
+    maxw =: wordmap.maxlen
+    ?# idx =: map wordmap give keys ascending
+        words =: wordmap{idx}
+        ? maxw = words.count
+            print words::join values by ', '
+
+\ read the file (infn) and return a map
+get words from (infn):
+    res =: new map                      \ create the result map
+    res.maxlen =: 0
+    ?# line =: file infn give lines     \ or: infn::give lines
+        add line to res
+    :> res
+
+\ add a word to the map
+add (word) to (amap):
+    idx =: sort characters of word::case to upper
+    row =: amap{idx}                    \ get indexed words
+    ? row = ()                          \ new entry if void
+        row =: new row
+        amap{idx} =: row
+    row[] =: word                       \ add new entry
+    amap.maxlen =: maximum of amap.maxlen, row.count
+
+\ Sort the characters of a string
+sort characters of (s):
+    sr =: string s as character row
+    row sr sort ascending
+    :> row sr join values as string     \ without separator