Added KradfileToSQLite script

Author: FalsinSoft

KradfileToSQLite/KradfileToSQLite.py

@@ -0,0 +1,105 @@
+
+from argparse import ArgumentParser, FileType, Action
+import sqlite3
+import sys
+import codecs
+import os
+
+def parse_cmdline():
+    parser = ArgumentParser()
+    parser.add_argument("--kradfile", help="path to the kradfile", default="kradfile")
+    parser.add_argument("--kradfile2", help="path to the kradfile2", default="kradfile2")
+    parser.add_argument("--sqlitefile", help="path to the sqlite database to create", required=True)
+    return parser.parse_args()
+
+
+def create_database(name):
+    sqlitefile = name
+
+    if len(sqlitefile) < 3 or sqlitefile[-3:] != ".db":
+        sqlitefile = sqlitefile + ".db"
+
+    if os.path.exists(sqlitefile):
+        os.remove(sqlitefile)
+
+    database = sqlite3.connect(sqlitefile)
+
+    c = database.cursor()
+    c.execute("CREATE TABLE radicals (data TEXT NOT NULL)")
+    c.execute("CREATE TABLE kanji (data TEXT NOT NULL)")
+    c.execute("CREATE TABLE kanji_radical (kanji_id INTEGER, radical_id INTEGER)")
+
+    return database
+
+
+def get_radical_id(radical, database):
+    c = database.cursor()
+    radical_id = 0
+
+    c.execute("SELECT rowid FROM radicals WHERE data = ?", [radical])
+    row = c.fetchone()
+
+    if row is None:
+        c.execute("INSERT INTO radicals (data) VALUES (?)", [radical])
+        radical_id = c.lastrowid
+    else:
+        radical_id = row[0]
+
+    return radical_id
+
+
+def import_data_file(database, kradfile):
+    c = database.cursor()
+    counter = 0
+
+    for line in kradfile:
+        line = line.strip()
+
+        if line[0] == "#" or line[0] == " ":
+            continue
+
+        try:
+            kanji,radicals = line.split(":")
+        except:
+            continue
+
+        c.execute("INSERT INTO kanji (data) VALUES (?)", [line[0]])
+        kanji_id = c.lastrowid
+        
+        for radical in radicals.split():
+            radical_id = get_radical_id(radical.strip(), database)
+            c.execute("INSERT INTO kanji_radical (kanji_id, radical_id) VALUES (?, ?)", (kanji_id, radical_id))
+
+        counter += 1
+        if not counter % 100:
+            print("#", end = "", flush = True)
+
+    database.commit()
+
+
+def main():
+    args = parse_cmdline()
+
+	# Example query for get the list of radicas of specific kanji
+    # select data from radicals, kanji_radical where radicals.rowid = kanji_radical.radical_id and kanji_radical.kanji_id = (select rowid from kanji where data = '鰯')
+	# Example query for get the list kanji having specific radical
+    # select data from kanji, kanji_radical where kanji.rowid = kanji_radical.kanji_id and kanji_radical.radical_id = (select rowid from radicals where data = '田')
+
+    kradfile = codecs.open(args.kradfile, 'r', 'euc-jp')
+    kradfile2 = codecs.open(args.kradfile2, 'r', 'euc-jp')
+    database = create_database(args.sqlitefile.strip())
+
+    print("Start conversion: ", end = "", flush = True)
+
+    import_data_file(database, kradfile)
+    import_data_file(database, kradfile2)
+
+    print("\nConversion finished")
+
+    kradfile.close()
+    kradfile2.close()
+    database.close()
+
+    
+if __name__ == '__main__':
+    main()

KradfileToSQLite/KradfileToSQLite.pyproj

@@ -0,0 +1,38 @@
+<Project DefaultTargets="Build" xmlns="http://schemas.microsoft.com/developer/msbuild/2003" ToolsVersion="4.0">
+  <PropertyGroup>
+    <Configuration Condition=" '$(Configuration)' == '' ">Debug</Configuration>
+    <SchemaVersion>2.0</SchemaVersion>
+    <ProjectGuid>3164fef1-678a-46dc-81d1-b4a44144f4ad</ProjectGuid>
+    <ProjectHome>.</ProjectHome>
+    <StartupFile>KradfileToSQLite.py</StartupFile>
+    <SearchPath>
+    </SearchPath>
+    <WorkingDirectory>.</WorkingDirectory>
+    <OutputPath>.</OutputPath>
+    <Name>KradfileToSQLite</Name>
+    <RootNamespace>KradfileToSQLite</RootNamespace>
+    <LaunchProvider>Standard Python launcher</LaunchProvider>
+    <CommandLineArguments>--sqlitefile=kradfile</CommandLineArguments>
+    <EnableNativeCodeDebugging>False</EnableNativeCodeDebugging>
+  </PropertyGroup>
+  <PropertyGroup Condition=" '$(Configuration)' == 'Debug' ">
+    <DebugSymbols>true</DebugSymbols>
+    <EnableUnmanagedDebugging>false</EnableUnmanagedDebugging>
+  </PropertyGroup>
+  <PropertyGroup Condition=" '$(Configuration)' == 'Release' ">
+    <DebugSymbols>true</DebugSymbols>
+    <EnableUnmanagedDebugging>false</EnableUnmanagedDebugging>
+  </PropertyGroup>
+  <ItemGroup>
+    <Compile Include="KradfileToSQLite.py" />
+  </ItemGroup>
+  <Import Project="$(MSBuildExtensionsPath32)\Microsoft\VisualStudio\v$(VisualStudioVersion)\Python Tools\Microsoft.PythonTools.targets" />
+  <!-- Uncomment the CoreCompile target to enable the Build command in
+       Visual Studio and specify your pre- and post-build commands in
+       the BeforeBuild and AfterBuild targets below. -->
+  <!--<Target Name="CoreCompile" />-->
+  <Target Name="BeforeBuild">
+  </Target>
+  <Target Name="AfterBuild">
+  </Target>
+</Project>

KradfileToSQLite/KradfileToSQLite.sln

@@ -0,0 +1,23 @@
+
+Microsoft Visual Studio Solution File, Format Version 12.00
+# Visual Studio 15
+VisualStudioVersion = 15.0.27428.2037
+MinimumVisualStudioVersion = 10.0.40219.1
+Project("{888888A0-9F3D-457C-B088-3A5042F75D52}") = "KradfileToSQLite", "KradfileToSQLite.pyproj", "{3164FEF1-678A-46DC-81D1-B4A44144F4AD}"
+EndProject
+Global
+	GlobalSection(SolutionConfigurationPlatforms) = preSolution
+		Debug|Any CPU = Debug|Any CPU
+		Release|Any CPU = Release|Any CPU
+	EndGlobalSection
+	GlobalSection(ProjectConfigurationPlatforms) = postSolution
+		{3164FEF1-678A-46DC-81D1-B4A44144F4AD}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
+		{3164FEF1-678A-46DC-81D1-B4A44144F4AD}.Release|Any CPU.ActiveCfg = Release|Any CPU
+	EndGlobalSection
+	GlobalSection(SolutionProperties) = preSolution
+		HideSolutionNode = FALSE
+	EndGlobalSection
+	GlobalSection(ExtensibilityGlobals) = postSolution
+		SolutionGuid = {2E885054-2E6B-46E6-B175-1C05CF394570}
+	EndGlobalSection
+EndGlobal