BIルールで形態素分析:Spotfire + NMecab + Iron Python + DataTable





import clr    

clr.AddReferenceToFileAndPath(r"D:/Work/NMecab/bin/LibNMeCab.dll")

import System

from System import *

import NMeCab

from NMeCab import *


import Spotfire.Dxp.Application

from Spotfire.Dxp.Data import *

from Spotfire.Dxp.Application.Visuals import VisualContent

from System.Collections.Generic import HashSet

from System.IO import FileStream, FileMode, File, MemoryStream, SeekOrigin, StreamWriter


import System.String

from Spotfire.Dxp.Data.Import import TextDataReaderSettings

from Spotfire.Dxp.Data.Import import TextFileDataSource


def LoadCSV(dataTableName, stream):

settings = TextDataReaderSettings()

settings.Separator = ","

settings.AddColumnNameRow(0)

settings.ClearDataTypes(True)

settings.SetDataType(0, DataType.String)

settings.SetDataType(1, DataType.String)

settings.SetDataType(2, DataType.Real)

settings.SetDataType(3, DataType.Real)

stream.Seek(0, SeekOrigin.Begin)

fs = TextFileDataSource(stream, settings)

if Document.Data.Tables.Contains(dataTableName):

Document.Data.Tables[dataTableName].ReplaceData(fs)

else:

Document.Data.Tables.Add(dataTableName, fs)



stream = MemoryStream()


DicDir = r"D:\Work\NMecab\dic\ipadic"


t=MeCabTagger.Create(DicDir)


nodes = t.Parse( sentence )


result = ""

csvWriter = StreamWriter(stream) #, Encoding.UTF8)

csvWriter.WriteLine("COL01,COL02,COL03,COL04,COL05,COL06,COL07,COL08,COL09,COL10\r\n")


for node in nodes :

print node.Surface + "\t" + node.Feature

result = result +  node.Surface + "\t" + node.Feature + "\n"


csvWriter.WriteLine(node.Surface + "," + node.Feature + "\n")


Document.Properties['NMeCabPaserResult'] = result



csvWriter.Flush()

LoadCSV("test", stream)


exit


コメント

人気の投稿