comp.lang.ada
 help / color / mirror / Atom feed
From: "Dmitry A. Kazakov" <mailbox@dmitry-kazakov.de>
Subject: Re: Ada bench : count words
Date: Tue, 22 Mar 2005 13:17:43 +0100
Date: 2005-03-22T13:17:41+01:00	[thread overview]
Message-ID: <1q9cx4jt7802s.k45m6mcntl87$.dlg@40tude.net> (raw)
In-Reply-To: mailman.48.1111492666.23655.comp.lang.ada@ada-france.org

On Tue, 22 Mar 2005 11:57:22 +0000, Marius Amado Alves wrote:

>>> ... To implement buffering, I have resorted to
>>> Ada.Direct_IO, which I think cannot apply to standard input.
>>
>> Is Text_IO that bad?
> 
> No, if you can solve The Get_Line puzzle :-)

What about Get (Item : out Character)?

I wonder if calling C-lib's read would qualify! (:-))

>>>     procedure Process (S : in String) is
>>>     begin
>>>        Lines := Lines + Ada.Strings.Fixed.Count (S, EOL);
>>
>> Isn't it an extra pass? I think you should do parsing using FSM.
>> Character
>> classes are: EOL, delimiter, letter. It is either two character map 
>> tests
>> or one case statement. I don't know what is faster. Probably you should
>> test both.
>>
>>>        for I in S'Range loop
>>>           if Is_Separator (S (I)) then
>>>              if In_Word then Finish_Word; end if;
>>>           else
>>>              if not In_Word then Start_Word; end if;
>>>           end if;
>>>        end loop;
>>>     end;
> 
> Note EOL is not a character, but a string, because in some environments 
> the thing is a combination of two characters.

I think Text_IO should translate it into LF. But anyway you can always
assume LF = new line, CR = space in the FSM.

> This one-character 
> version improves speed (but still only to 1/2 of C):
> 
>           for I in S'Range loop
>              if S (I) = EOL then Lines := Lines + 1; end if;
>              if Is_Separator (S (I)) then
>                 if In_Word then Finish_Word; end if;
>              else
>                 if not In_Word then Start_Word; end if;
>              end if;
>           end loop;
> 
> I have not tried with string matching (if that's what you mean with 
> "FSM") because the iteration was already there, and I doubt the 
> standard library implements it more efficiently than that.

No, I meant a finite state machine:

<<Word>>
   read character;
   case Char is
      when EOL    => Inc line count; goto Blank;
      when Space => goto Blank;
      when Letter => goto Word;
   end case;

<<Blank>>
   read character;
   case Char is
      when EOL    => Inc line count; goto Blank;
      when Space => goto Blank;
      when Letter => Inc word count; goto Word;
   end case;

<<Word>>
   read character;
   case Char is
      when EOL    => Inc line count; goto Blank;
      when Space => goto Blank;
      when Letter => goto Word;
   end case;

FSM is one of that rare cases where gotos are natural. What a pity that Ada
does not have arrays of labels! (:-))

-- 
Regards,
Dmitry A. Kazakov
http://www.dmitry-kazakov.de



  reply	other threads:[~2005-03-22 12:17 UTC|newest]

Thread overview: 122+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-03-19 16:22 Ada bench Pascal Obry
2005-03-19 16:55 ` Dr. Adrian Wrigley
2005-03-19 21:32   ` Michael Bode
2005-03-20  9:20     ` Pascal Obry
2005-03-20  9:39       ` Michael Bode
2005-03-20 11:16         ` Pascal Obry
2005-03-20 12:20           ` Dmitry A. Kazakov
2005-03-20 12:32             ` Pascal Obry
2005-03-21  1:42             ` (see below)
2005-03-21  2:24               ` (see below)
2005-03-21 15:00                 ` (see below)
2005-03-21  3:54               ` Ed Falis
2005-03-21 12:20                 ` Jeff C
2005-03-21 15:18                 ` (see below)
2005-03-21 15:24                   ` (see below)
2005-03-21 18:56                   ` Isaac Gouy
2005-03-21 21:31                 ` Randy Brukardt
2005-03-21 22:14                   ` Ed Falis
2005-03-21 18:07               ` Pascal Obry
2005-03-20 10:11       ` Adrian Knoth
2005-03-20 10:30         ` Michael Bode
2005-03-21 23:27       ` Georg Bauhaus
2005-03-22  1:16         ` Ada bench : count words Marius Amado Alves
2005-03-22 10:59           ` Dmitry A. Kazakov
2005-03-22 11:57             ` Marius Amado Alves
2005-03-22 12:17               ` Dmitry A. Kazakov [this message]
2005-03-22 12:47                 ` Marius Amado Alves
2005-03-22 13:08                   ` Dmitry A. Kazakov
2005-03-22 13:28                     ` Marius Amado Alves
2005-03-22 16:48                     ` Marius Amado Alves
2005-03-22 17:34                       ` Dmitry A. Kazakov
2005-03-27 20:14                         ` jtg
2005-03-27 21:22                           ` Dmitry A. Kazakov
2005-03-28 19:54                             ` jtg
2005-03-28 20:56                               ` Dmitry A. Kazakov
2005-03-29 12:40                                 ` jtg
2005-03-29 13:00                                   ` [OT] " Tapio Kelloniemi
2005-03-29 13:47                                     ` Dmitry A. Kazakov
2005-03-29 15:53                                       ` Tapio Kelloniemi
2005-03-29 16:17                                         ` Dmitry A. Kazakov
     [not found]                                         ` <k33j419lgei1ui89s26o1dlr9ccf1qe1hd@4ax.com>
2005-04-11 23:04                                           ` Marius Amado Alves
2005-03-29 13:47                                   ` Dmitry A. Kazakov
2005-04-01 20:58                                   ` Georg Bauhaus
2005-04-01 20:18                                     ` Pascal Obry
2005-03-22 12:53                 ` Marius Amado Alves
     [not found]                 ` <f205219321dd18dba878fab16b7cb50d@netcabo.pt>
2005-03-22 13:12                   ` Marius Amado Alves
2005-03-23 16:58                     ` Isaac Gouy
2005-03-22 13:58                 ` Robert A Duff
2005-03-22 16:30                   ` Marius Amado Alves
2005-03-22 16:41                     ` Tapio Kelloniemi
2005-03-22 17:39                       ` Marius Amado Alves
2005-03-22 18:59                         ` Dmitry A. Kazakov
2005-03-22 19:08                         ` Tapio Kelloniemi
2005-03-22 18:34                     ` Georg Bauhaus
2005-03-22 19:32                     ` Robert A Duff
2005-03-22 20:04                       ` tmoran
2005-03-23 16:55                       ` Isaac Gouy
     [not found]                   ` <1820eab50b57f2fe1c4e8e50bb0f4fe5@netcabo.pt>
2005-03-22 22:49                     ` Stephen Leake
2005-03-22 22:58                       ` Robert A Duff
2005-03-22 23:27                   ` Larry Kilgallen
2005-03-23 22:33                     ` Robert A Duff
2005-03-24  5:02                     ` Larry Kilgallen
     [not found]                     ` <wccpsxqro0c.fsfOrganization: LJK Software <bM40pHW6P2KW@eisner.encompasserve.org>
2005-03-25  1:34                       ` Robert A Duff
2005-03-22 12:22             ` Jeff C
2005-03-23 16:48               ` Isaac Gouy
2005-03-23 17:06             ` Isaac Gouy
2005-03-22 19:49           ` tmoran
2005-03-22 21:51             ` Dmitry A. Kazakov
2005-03-23  0:16               ` tmoran
2005-03-23  7:25                 ` Dmitry A. Kazakov
2005-03-22 22:33             ` Marius Amado Alves
     [not found]             ` <00b362390273e6c04844dd4ff1885ee0@netcabo.pt>
2005-03-23 15:09               ` Marius Amado Alves
2005-03-23 19:00                 ` tmoran
2005-03-23 19:30                   ` tmoran
2005-03-23 21:38                     ` tmoran
2005-03-25  7:30                       ` Simon Wright
2005-03-25  9:38                         ` tmoran
2005-03-25 16:20                           ` Tapio Kelloniemi
2005-03-25 22:18                             ` Ada:The High Performance Language for Hyperthreaded and Multicore CPUs; was " tmoran
2005-03-23 19:54                   ` Tapio Kelloniemi
2005-03-23 20:39                     ` Ada bench : word frequency Marius Amado Alves
2005-03-23 21:26                       ` Isaac Gouy
2005-03-24  1:24                         ` Marius Amado Alves
2005-03-24 17:23                           ` Isaac Gouy
2005-03-24 19:52                           ` Martin Dowie
2005-04-11 15:11                             ` Marius Amado Alves
2005-03-24 20:16                           ` Pascal Obry
2005-03-24 22:54                             ` tmoran
2005-04-11 15:38                               ` Marius Amado Alves
2005-03-24 21:18                           ` Gautier Write-only
2005-04-11 15:32                             ` Marius Amado Alves
2005-04-11 23:56                               ` Robert A Duff
2005-04-28  4:04                       ` Matthew Heaney
2005-04-28 20:40                         ` Matthew Heaney
2005-03-23 21:38                     ` Ada bench : count words tmoran
2005-03-24 20:19                   ` tmoran
2005-03-24 21:00                     ` Pascal Obry
2005-03-24 22:54                       ` tmoran
2005-03-30 16:08                 ` Andre
2005-03-30 16:36                   ` Pascal Obry
2005-03-22 22:27           ` Dmitry A. Kazakov
2005-03-23  7:46             ` Pascal Obry
2005-03-23  7:56               ` Dmitry A. Kazakov
2005-03-23 13:38                 ` Robert A Duff
2005-03-22  7:05         ` Ada bench Pascal Obry
2005-04-07 20:59   ` David Sauvage
2005-04-07 23:40     ` David Sauvage
2005-04-08 17:11     ` Pascal Obry
2005-04-08 17:39       ` tmoran
2005-04-08 18:49       ` David Sauvage
2005-04-18 19:14       ` David Sauvage
2005-04-19 16:43         ` Matthew Heaney
2005-04-19 23:22           ` David Sauvage
2005-04-20  0:49             ` Matthew Heaney
2005-04-20  4:22               ` Georg Bauhaus
2005-04-20 21:24               ` David Sauvage
2005-04-20 23:06                 ` Georg Bauhaus
2005-04-20  9:41             ` Pascal Obry
2005-04-20 11:44               ` Matthew Heaney
2005-04-20 14:47                 ` Pascal Obry
2005-04-20 19:26                   ` Georg Bauhaus
2005-04-20 19:34                     ` Pascal Obry
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox