772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
|
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
|
-
+
+
-
+
+
-
+
-
+
+
+
+
+
+
+
+
+
+
-
+
|
pragma Assert (for all Count of Counter.Map => Count >= Threshold_Count);
end Filter_By_Count;
function Simple_Dictionary
(Counter : in Word_Counter;
Word_Count : in Natural)
Word_Count : in Natural;
Method : in Methods.Enum := Methods.Encoded)
return String_Lists.List
is
use type Ada.Containers.Count_Type;
Target_Count : constant Ada.Containers.Count_Type
:= Ada.Containers.Count_Type (Word_Count);
Set : Scored_Word_Sets.Set;
Result : String_Lists.List;
begin
for Cursor in Word_Maps.Iterate (Counter.Map) loop
Scored_Word_Sets.Include (Set, To_Scored_Word (Cursor));
Scored_Word_Sets.Include (Set, To_Scored_Word (Cursor, Method));
if Scored_Word_Sets.Length (Set) > Target_Count then
Scored_Word_Sets.Delete_Last (Set);
end if;
end loop;
for Cursor in Scored_Word_Sets.Iterate (Set) loop
Result.Append (Scored_Word_Sets.Element (Cursor).Word);
end loop;
return Result;
end Simple_Dictionary;
procedure Simple_Dictionary_And_Pending
(Counter : in Word_Counter;
Word_Count : in Natural;
Selected : out String_Lists.List;
Pending : out String_Lists.List;
Method : in Methods.Enum := Methods.Encoded;
Max_Pending_Count : in Ada.Containers.Count_Type
:= Ada.Containers.Count_Type'Last)
is
use type Ada.Containers.Count_Type;
Target_Count : constant Ada.Containers.Count_Type
:= Ada.Containers.Count_Type (Word_Count);
Set : Scored_Word_Sets.Set;
begin
for Cursor in Word_Maps.Iterate (Counter.Map) loop
Scored_Word_Sets.Insert (Set, To_Scored_Word (Cursor));
Scored_Word_Sets.Insert (Set, To_Scored_Word (Cursor, Method));
end loop;
Selected := String_Lists.Empty_List;
Pending := String_Lists.Empty_List;
for Cursor in Scored_Word_Sets.Iterate (Set) loop
if String_Lists.Length (Selected) < Target_Count then
Selected.Append (Scored_Word_Sets.Element (Cursor).Word);
else
Pending.Append (Scored_Word_Sets.Element (Cursor).Word);
exit when String_Lists.Length (Selected) >= Max_Pending_Count;
end if;
end loop;
end Simple_Dictionary_And_Pending;
function To_Scored_Word (Cursor : in Word_Maps.Cursor)
function To_Scored_Word
(Cursor : in Word_Maps.Cursor;
Method : in Methods.Enum)
return Scored_Word
is
Word : constant String := Word_Maps.Key (Cursor);
Factor : Score_Value;
begin
case Method is
when Methods.Encoded => Factor := Word'Length;
when Methods.Frequency => Factor := 1;
when Methods.Gain => Factor := Word'Length - 1;
end case;
return Scored_Word'
(Size => Word'Length,
Word => Word,
Score => Score_Value (Word_Maps.Element (Cursor)) * Word'Length);
Score => Score_Value (Word_Maps.Element (Cursor)) * Factor);
end To_Scored_Word;
function Worst_Index
(Dict : in Dictionary;
Counts : in Dictionary_Counts;
Method : in Methods.Enum)
|