Natools

Check-in [9a1462e3f3]
Login
Overview
Comment:tools/smaz: add command-line options for substring size interval
Timelines: family | ancestors | descendants | both | trunk
Files: files | file ages | folders
SHA1: 9a1462e3f39c38a3ace090f6db8cab6be1030fca
User & Date: nat on 2016-10-04 15:14:26
Other Links: manifest | tags
Context
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
2016-10-03
19:27
tools/smaz: also add words from input phrases, on top of substrings check-in: 15ea367b55 user: nat tags: trunk
Changes

Modified tools/smaz.adb from [edede487cd] to [a553821384].

51
52
53
54
55
56
57


58
59
60
61
62
63
64
65
66
67
68
69
70
71
72


73
74
75
76
77
78
79
        (Output_Ada_Dict,
         Dictionary_Input,
         Decode,
         Encode,
         Output_Hash,
         Help,
         Sx_Dict_Output,


         Stat_Output,
         No_Stat_Output,
         Word_List_Input,
         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;


      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







>
>















>
>







51
52
53
54
55
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
        (Output_Ada_Dict,
         Dictionary_Input,
         Decode,
         Encode,
         Output_Hash,
         Help,
         Sx_Dict_Output,
         Min_Sub_Size,
         Max_Sub_Size,
         Stat_Output,
         No_Stat_Output,
         Word_List_Input,
         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;
      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
162
163
164
165
166
167
168






169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184


185
186
187
188
189
190
191
192
193
194
195
196

         when Options.Word_List_Input =>
            Handler.Dict_Source := Dict_Sources.Word_List;

         when Options.Sx_Dict_Output =>
            Handler.Need_Dictionary := True;
            Handler.Sx_Dict_Output := True;






      end case;
   end Option;


   function Getopt_Config return Getopt.Configuration is
      use Getopt;
      use Options;
      R : Getopt.Configuration;
   begin
      R.Add_Option ("ada-dict",  'A', Optional_Argument, Output_Ada_Dict);
      R.Add_Option ("decode",    'd', No_Argument,       Decode);
      R.Add_Option ("dict",      'D', No_Argument,       Dictionary_Input);
      R.Add_Option ("encode",    'e', No_Argument,       Encode);
      R.Add_Option ("help",      'h', No_Argument,       Help);
      R.Add_Option ("hash-pkg",  'H', Required_Argument, Output_Hash);
      R.Add_Option ("sx-dict",   'L', No_Argument,       Sx_Dict_Output);


      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 ("s-expr",    'x', No_Argument,       Sx_Output);
      R.Add_Option ("no-s-expr", 'X', No_Argument,       No_Sx_Output);

      return R;
   end Getopt_Config;


   procedure Print_Dictionary
     (Filename : in String;







>
>
>
>
>
>









|
|
|
|
|
|
|
>
>
|
|
|
|
|







166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208

         when Options.Word_List_Input =>
            Handler.Dict_Source := Dict_Sources.Word_List;

         when Options.Sx_Dict_Output =>
            Handler.Need_Dictionary := True;
            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);
      end case;
   end Option;


   function Getopt_Config return Getopt.Configuration is
      use Getopt;
      use Options;
      R : Getopt.Configuration;
   begin
      R.Add_Option ("ada-dict",      'A', Optional_Argument, Output_Ada_Dict);
      R.Add_Option ("decode",        'd', No_Argument,       Decode);
      R.Add_Option ("dict",          'D', No_Argument,       Dictionary_Input);
      R.Add_Option ("encode",        'e', No_Argument,       Encode);
      R.Add_Option ("help",          'h', No_Argument,       Help);
      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 ("s-expr",        'x', No_Argument,       Sx_Output);
      R.Add_Option ("no-s-expr",     'X', No_Argument,       No_Sx_Output);

      return R;
   end Getopt_Config;


   procedure Print_Dictionary
     (Filename : in String;
309
310
311
312
313
314
315










316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334

335

336
337
338
339
340
341
342
               Put_Line (Output, Indent & Indent
                 & "Compute dictionary from word list in input S-expression");

            when Options.Sx_Dict_Output =>
               New_Line (Output);
               Put_Line (Output, Indent & Indent
                 & "Output the dictionary as a S-expression");










         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, 1, 3);

                  Natools.Smaz.Tools.Add_Words (Counter, S, 4, 10);

               end loop;

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







>
>
>
>
>
>
>
>
>
>


















|
>
|
>







321
322
323
324
325
326
327
328
329
330
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
               Put_Line (Output, Indent & Indent
                 & "Compute dictionary from word list in input S-expression");

            when Options.Sx_Dict_Output =>
               New_Line (Output);
               Put_Line (Output, Indent & Indent
                 & "Output the dictionary as a S-expression");

            when Options.Min_Sub_Size =>
               New_Line (Output);
               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");
         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);
                  Natools.Smaz.Tools.Add_Words
                    (Counter, S, Handler.Max_Sub_Size + 1, 10);
               end loop;

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