cutano / kawazu Goto Github PK

Hi,
First thanks for the great work with Kawazu, very heplful!
But sometimes it generates an exception.
Step to reproduce :

var converter = new KawazuConverter();
var inputH = converter.Convert("だれでうどうんづしますか", To.Hiragana, Mode.Normal).Result;

Stacktrace:

   at Kawazu.Division..ctor(MeCabIpaDicNode node, TextType type, RomajiSystem system)
   at Kawazu.KawazuConverter.<>c__DisplayClass6_0.<Convert>b__1(MeCabIpaDicNode node)
   at System.Linq.Enumerable.SelectArrayIterator`2.ToList()
   at Kawazu.KawazuConverter.<>c__DisplayClass6_0.<Convert>b__0()
   at System.Threading.Tasks.Task`1.InnerInvoke()
   at System.Threading.ExecutionContext.RunFromThreadPoolDispatchLoop(Thread threadPoolThread, ExecutionContext executionContext, ContextCallback callback, Object state)
--- End of stack trace from previous location ---
   at System.Runtime.ExceptionServices.ExceptionDispatchInfo.Throw()
   at System.Threading.Tasks.Task.ExecuteWithThreadLocal(Task& currentTaskSlot, Thread threadPoolThread)
--- End of stack trace from previous location ---
   at System.Runtime.ExceptionServices.ExceptionDispatchInfo.Throw()
   at System.Runtime.CompilerServices.TaskAwaiter.ThrowForNonSuccess(Task task)
   at System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task)
   at Kawazu.KawazuConverter.<Convert>d__6.MoveNext()

When I test your lastest nuget package following the approach in readme.md, it still report char.bin not found error.

Then I test the elder verison(Kawazu1.0.0) with .net core 3.1, the resources was not been copied, either.

the current version Kawazu works only with .Net core1-3, I tested .Net5 and it works good only if I copy the matrix.bin and blabal files to the build output floder.

So it would be easy to support .Net5 by modify the csproj file.

BTW, it would helpful if you can support the old .net frameworks4+, there are a lot apps running on .net frameworks, but I'm not sure if Kawzu can run on .NetFramework4 without code changing, no test yet.

I don't know much about Japanese Romaji
But why it gets differet Romaji of こんにちは from google and Kawazu? Can I get the google's Romaji from Kawazu?

Google

return Kon'nichiwa

Kawazu

I tried different parameters of Converter.GetDivisions all return: konnichiha

I tried to use Kawazu-Cli to test input ノ森中ノ丁, but exception occurred.

The exception occurred in Division.cs file line 144:

Can you check this problem? I tried to understand the code but didn't succeed.

Thank you!

I could be doing things incorrectly but i am trying to basically do 2 things given almost any imput in a japanese text field.

1. Convert to Kana
2. Convert to Romaji

The Romaji conversion is thowing a heap of erorrs most of the time like the one below from a unit test.

        [Theory]
        [InlineData("袖ケ浦港運", "Sodegaura-kō un")]
        public async Task RomajiTransliterationTest(string input, string expectedOutput)
        {
            KawazuConverter converter = new();
            var output = await converter.Convert(input, To.Romaji, Mode.Okurigana, RomajiSystem.Hepburn);
            Assert.Equal(expectedOutput, output);
        }

System.ArgumentOutOfRangeException: Length cannot be less than zero. (Parameter 'length')
   at System.Text.StringBuilder.ToString(Int32 startIndex, Int32 length)
   at Kawazu.Division..ctor(MeCabIpaDicNode node, TextType type, RomajiSystem system)
   at Kawazu.KawazuConverter.<>c__DisplayClass6_0.<Convert>b__1(MeCabIpaDicNode node)
   at System.Linq.Enumerable.SelectArrayIterator`2.ToList()
   at System.Linq.Enumerable.ToList[TSource](IEnumerable`1 source)
   at Kawazu.KawazuConverter.<>c__DisplayClass6_0.<Convert>b__0()
   at System.Threading.Tasks.Task`1.InnerInvoke()
   at System.Threading.Tasks.Task.<>c.<.cctor>b__277_0(Object obj)
   at System.Threading.ExecutionContext.RunFromThreadPoolDispatchLoop(Thread threadPoolThread, ExecutionContext executionContext, ContextCallback callback, Object state)

Hi,
Not sure if this is a Kawazu or LibNMeCab issue, but when converting kanjis ignores exceptions.
For example, if one wants to convert 300 with 三百, it will output さんひゃく (sanhyaku) but the correct answer is さんびゃく (sanbyaku).
Same for 600 and 900.

Currently working on a workaround on my fork: lasyan3@156bf7e

Hi Cutano. This is more of a request than an issue. Would you include a PartsOfSpeech property to Division.cs? I believe you can pull it from MeCabIpaDicNode.

https://github.com/Cutano/Kawazu/blob/master/Kawazu/Division.cs#L131-L154

This fails to correctly segment the readings for words without kanji in the middle. Eg 言い方 and 当たり前 both fail.

Hi, thanks for your nice work, but I'd like to get pronunciations one by one.

as"今晩"

var result = await converter.Convert("今晩は", To.Romaji, Mode.Normal, RomajiSystem.Hepburn, "(", ")");

Kawazu return "komban" now

I wanna get a list {"kom", "ban"} for searching match purpose

is there any way to get this list?

Hi Cutano. Sorry to bother you again. I was just wondering. Is it possible to covert Romaji to Hiragana?

I think you should add a dispose method in the KawazuConverter class to dispose the MeCab tagger and release unmanaged memory because the garbage collector can't clean unmanaged memory.

Hi Cutano. I was playing around with the KawazuConverter and found that when converting to To.Hiragana in Mode.Furigana it sometimes returns inaccurate results. For example, with the word:

あの方
the result was: あのほう
and it should have been: あのかた

Hi. This is more of a question than an issue. If I include this library in a commercial project. What do I need to do to satisfy this condition in your license?

"The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software."