Hi,

I’m Andrew and developing a soft-bot which can speak Korean, and there is one question.

1. How can I make the Korean words work in a topic keyword list?

  (1) I registerd some Korean Noun words in the canonical.txt file.

  (2) I checked if the _Match variables catch the Korean words in the introductions.top.
      Success!

  (3) And I made a new topic named ~KoreanTest and put the Korean Noun words in that topic
      keyword list, and write some rules.

  (4) :verify keyword—>  “Missing keyword ~KoreanTest.x.x”

      Why are not the Korean words in the topic keyword list recognized?

  (5) by the way, :verify ~KoreanTest—>

      Missing keyword ~KoreanTest.x.x
      Pattern failed to match 6 ~KoreanTest.x.x
        Adjusted Input :
        Canonical Input :  <——result are O.K.

  (6) Alos, I tried to make a concept like this:

        concept: ~KoreanNoun NOUN ( korean-noun-words-list )

      and :concepts korean-word

      then, there are no concepts list. (expected result was ~KoreanNoun)

  (7) :prepare Above_One_Korean_Word

      then results are;

        Original User Input : Above_One_Korean_Word
        Tokenized into : Above_One_Korean_Word
        Actual used input : Above_One_Korean_Word

        Concepts : (blank)

        Above_One_Korean_Word (raw) : +~noun_infinitive etc //
        Above_One_Korean_Word (canonical) :  // +~utf8

        Seauences : (blank)
        After parse TokenFlags : USERINPUT

I don’t know how can I activate the Korean Word in the topic keyword list.

Please help me.

Thanks in advance.

insufficient data.  Maybe if you emailed me your topic file and a sample input you wanted it to react to, I could diagnose it.

Your files are not saved as UTF8 files, which is what chatscript needs to handle foreign words.
Use notepad, for example, to open them, then do “save as” and select file type utf8

Thank you very much. my mistake. I’m sorry.
(I’m trying to use the command :source because windows console cann’t be input Korean in UTF-8 mode.)

And there are still one problem.
Once Korean word matched, then CS engine couldn’t find that word any more.
It is strange. I’ll check more deeply.

Thank you very much.

if something succeeds the first time and fails thereafter, it usually means you used a responder and it got erased because the topic was not marked KEEP or because the system avoids repeating itself within 20 volleys, and you can override that by declaring the topic REPEAT

Yes. Thanks a lot. (I already declared the word “keep” and “repeat” on the topic.)
Probably the character set seems to cause the problem. (I’m not sure.)

Now I am trying to install the Ubuntu in my PC and I’ll test Linux environment
to solve the input and output of the Korean word in UTF 8 mode.
(In windows environment it’s difficult to test my scripts because the cmd window cann’t be input in UTF8 characters via a keyboard.)

I’ll be back.
Thank you again.

Hi, there are some progress but not :verify command.

1. Korean words in a topic list work well.
  (1) saved files as UTF8 format.
  (2) a test file was provided by the command :source, which was saved as UTF8 format.
  (3) I checked the result via USERS/*.txt not via Windows cmd screen.

2. Also I can make a name of a concept by a Korean word.

3. And the following commands seems to work well on a Korean word.
  (1) :concepts
  (2) :prepare
  (3) :topicinfo

4. But the :verify command doesn’t work.
  (1) :verify ~KoreanTopicTest

      Command: :verify ~KoreanTopicTest
      VERIFYING ~koreantopictest ......
      0 verify findings of 6 trials.

    (2) :verify ~KoreanTopicTest pattern

      Command: :verify ~KoreanTopicTest pattern
      VERIFYING ~koreantopictest ......
      0 verify findings of 6 trials.

    (3) :verify ~KoreanTopicTest keyword

      Command: :verify ~KoreanTopicTest keyword
      VERIFYING ~koreantopictest ......
      0 verify findings of 0 trials.

If I don’t understand the :verify command properly, please tell me.

I attached 3 files.

  - introductions.top
  - main1.txt —> test file
  - log-alphago.txt—> log file

Hi,

There is one question.

CS has stopped working when it read the “canon1.txt” file in the Topic folder.

(1) I wrote a script file to make some Korean verb canonical form. (please see the attached file “KoreanVerb.top”)

(2) I saved that file in the harry folder as UTF-8.

(3) And :build harry

(4) Then CS compiled my script successfully and CS tried to read the canon1.txt which was just made, but CS suddenly has stopped working. (please see the attached file “screen_capture.PNG)

(5) I looked into the canon1.txt file but I couldn’t find any strange. (please see the attached file “canon1.txt”)

Where am I wrong?
Please advise me.

(And I replaced my console, from MS-Windows cmd to ConEmu by Maximus5 which is more stable to process UTF-8)

Thank you very much.

Bug in chatscript relating to loading canon1.txt
I have fixed this.  New CS release later this week.

Thank you very much, again!