Natools

Check-in [2995e1835c]
Login
Overview
Comment:tools/smaz: add a command-line option for maximum word size
Timelines: family | ancestors | descendants | both | trunk
Files: files | file ages | folders
SHA1: 2995e1835c55823f99d290818eb59a867df82c9f
User & Date: nat on 2016-10-05 14:49:50
Other Links: manifest | tags
Context
2016-10-06
17:26
smaz-tools: add a procedure to help evaluate and improve dictionaries check-in: 3c8fd02a5c user: nat tags: trunk
2016-10-05
14:49
tools/smaz: add a command-line option for maximum word size check-in: 2995e1835c user: nat tags: trunk
2016-10-04
15:14
tools/smaz: add command-line options for substring size interval check-in: 9a1462e3f3 user: nat tags: trunk
Changes

Modified tools/smaz.adb from [a553821384] to [abe2a95020].

56
57
58
59
60
61
62

63
64
65
66
67
68
69
70
71
72
73
74
75
76

77
78
79
80
81
82
83
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85







+














+







         Help,
         Sx_Dict_Output,
         Min_Sub_Size,
         Max_Sub_Size,
         Stat_Output,
         No_Stat_Output,
         Word_List_Input,
         Max_Word_Size,
         Sx_Output,
         No_Sx_Output);
   end Options;

   package Getopt is new Natools.Getopt_Long (Options.Id);

   type Callback is new Getopt.Handlers.Callback with record
      Display_Help : Boolean := False;
      Need_Dictionary : Boolean := False;
      Stat_Output : Boolean := False;
      Sx_Output : Boolean := False;
      Sx_Dict_Output : Boolean := False;
      Min_Sub_Size : Positive := 1;
      Max_Sub_Size : Positive := 3;
      Max_Word_Size : Positive := 10;
      Action : Actions.Enum := Actions.Nothing;
      Ada_Dictionary : Ada.Strings.Unbounded.Unbounded_String;
      Hash_Package : Ada.Strings.Unbounded.Unbounded_String;
      Dict_Source : Dict_Sources.Enum := Dict_Sources.S_Expression;
   end record;

   overriding procedure Option
172
173
174
175
176
177
178



179
180
181
182
183
184
185
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190







+
+
+







            Handler.Sx_Dict_Output := True;

         when Options.Min_Sub_Size =>
            Handler.Min_Sub_Size := Positive'Value (Argument);

         when Options.Max_Sub_Size =>
            Handler.Max_Sub_Size := Positive'Value (Argument);

         when Options.Max_Word_Size =>
            Handler.Max_Word_Size := Positive'Value (Argument);
      end case;
   end Option;


   function Getopt_Config return Getopt.Configuration is
      use Getopt;
      use Options;
193
194
195
196
197
198
199

200
201
202
203
204
205
206
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212







+







      R.Add_Option ("hash-pkg",      'H', Required_Argument, Output_Hash);
      R.Add_Option ("sx-dict",       'L', No_Argument,       Sx_Dict_Output);
      R.Add_Option ("min-substring", 'm', Required_Argument, Min_Sub_Size);
      R.Add_Option ("max-substring", 'M', Required_Argument, Max_Sub_Size);
      R.Add_Option ("stats",         's', No_Argument,       Stat_Output);
      R.Add_Option ("no-stats",      'S', No_Argument,       No_Stat_Output);
      R.Add_Option ("word-list",     'w', No_Argument,       Word_List_Input);
      R.Add_Option ("max-word-len",  'W', Required_Argument, Max_Word_Size);
      R.Add_Option ("s-expr",        'x', No_Argument,       Sx_Output);
      R.Add_Option ("no-s-expr",     'X', No_Argument,       No_Sx_Output);

      return R;
   end Getopt_Config;


331
332
333
334
335
336
337





338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357


358
359




360
361
362
363
364
365
366
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370


371
372
373
374
375
376
377
378
379
380
381







+
+
+
+
+




















+
+
-
-
+
+
+
+







               Put_Line (Output, Indent & Indent
                 & "Minimum substring size when building a dictionary");

            when Options.Max_Sub_Size =>
               New_Line (Output);
               Put_Line (Output, Indent & Indent
                 & "Maximum substring size when building a dictionary");

            when Options.Max_Word_Size =>
               New_Line (Output);
               Put_Line (Output, Indent & Indent
                 & "Maximum word size when building a dictionary");
         end case;
      end loop;
   end Print_Help;

   function To_Dictionary
     (Handler : in Callback'Class;
      Input : in Natools.Smaz.Tools.String_Lists.List)
     return Natools.Smaz.Dictionary is
   begin
      case Handler.Dict_Source is
         when Dict_Sources.S_Expression =>
            return Natools.Smaz.Tools.To_Dictionary (Input, True);

         when Dict_Sources.Word_List =>
            declare
               Counter : Natools.Smaz.Tools.Word_Counter;
            begin
               for S of Input loop
                  Natools.Smaz.Tools.Add_Substrings
                    (Counter, S, Handler.Min_Sub_Size, Handler.Max_Sub_Size);

                  if Handler.Max_Word_Size > Handler.Max_Sub_Size then
                  Natools.Smaz.Tools.Add_Words
                    (Counter, S, Handler.Max_Sub_Size + 1, 10);
                     Natools.Smaz.Tools.Add_Words
                       (Counter, S,
                        Handler.Max_Sub_Size + 1, Handler.Max_Word_Size);
                  end if;
               end loop;

               return Natools.Smaz.Tools.To_Dictionary
                 (Natools.Smaz.Tools.Simple_Dictionary (Counter, 254),
                  True);
            end;
      end case;