embulk-filter-gsub 0.1.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (32) hide show
  1. checksums.yaml +7 -0
  2. data/.circleci/config.yml +42 -0
  3. data/.gitignore +14 -0
  4. data/LICENSE.txt +21 -0
  5. data/README.md +141 -0
  6. data/build.gradle +49 -0
  7. data/gradle.properties +1 -0
  8. data/gradle/wrapper/gradle-wrapper.jar +0 -0
  9. data/gradle/wrapper/gradle-wrapper.properties +6 -0
  10. data/gradlew +172 -0
  11. data/gradlew.bat +84 -0
  12. data/lib/embulk/filter/gsub.rb +3 -0
  13. data/src/main/kotlin/org/embulk/filter/gsub/ColumnReplacerFactory.kt +20 -0
  14. data/src/main/kotlin/org/embulk/filter/gsub/ColumnVisitorImpl.kt +72 -0
  15. data/src/main/kotlin/org/embulk/filter/gsub/GsubFilterPlugin.kt +51 -0
  16. data/src/main/kotlin/org/embulk/filter/gsub/LowerCaseReplacerFactory.kt +23 -0
  17. data/src/main/kotlin/org/embulk/filter/gsub/RegexReplacerFactory.kt +20 -0
  18. data/src/main/kotlin/org/embulk/filter/gsub/SubstitutionRule.kt +26 -0
  19. data/src/main/kotlin/org/embulk/filter/gsub/TextReplacerFactory.kt +33 -0
  20. data/src/main/kotlin/org/embulk/filter/gsub/UpperCaseReplacerFactory.kt +23 -0
  21. data/src/main/kotlin/org/embulk/filter/gsub/replacer/CombinedReplacer.kt +7 -0
  22. data/src/main/kotlin/org/embulk/filter/gsub/replacer/LowerCaseReplacer.kt +17 -0
  23. data/src/main/kotlin/org/embulk/filter/gsub/replacer/RegexFactory.kt +22 -0
  24. data/src/main/kotlin/org/embulk/filter/gsub/replacer/RegexOptionConfig.kt +6 -0
  25. data/src/main/kotlin/org/embulk/filter/gsub/replacer/RegexReplacer.kt +7 -0
  26. data/src/main/kotlin/org/embulk/filter/gsub/replacer/TextReplacer.kt +5 -0
  27. data/src/main/kotlin/org/embulk/filter/gsub/replacer/UpperCaseReplacer.kt +18 -0
  28. data/src/test/kotlin/org/embulk/filter/gsub/TestGsubFilterPlugin.kt +221 -0
  29. data/src/test/kotlin/org/embulk/filter/gsub/replacer/CombinedReplacerTest.kt +37 -0
  30. data/src/test/kotlin/org/embulk/filter/gsub/replacer/LowerCaseReplacerTest.kt +20 -0
  31. data/src/test/kotlin/org/embulk/filter/gsub/replacer/RegexReplacerTest.kt +19 -0
  32. metadata +105 -0
data/gradlew.bat ADDED
@@ -0,0 +1,84 @@
1
+ @if "%DEBUG%" == "" @echo off
2
+ @rem ##########################################################################
3
+ @rem
4
+ @rem Gradle startup script for Windows
5
+ @rem
6
+ @rem ##########################################################################
7
+
8
+ @rem Set local scope for the variables with windows NT shell
9
+ if "%OS%"=="Windows_NT" setlocal
10
+
11
+ set DIRNAME=%~dp0
12
+ if "%DIRNAME%" == "" set DIRNAME=.
13
+ set APP_BASE_NAME=%~n0
14
+ set APP_HOME=%DIRNAME%
15
+
16
+ @rem Add default JVM options here. You can also use JAVA_OPTS and GRADLE_OPTS to pass JVM options to this script.
17
+ set DEFAULT_JVM_OPTS=
18
+
19
+ @rem Find java.exe
20
+ if defined JAVA_HOME goto findJavaFromJavaHome
21
+
22
+ set JAVA_EXE=java.exe
23
+ %JAVA_EXE% -version >NUL 2>&1
24
+ if "%ERRORLEVEL%" == "0" goto init
25
+
26
+ echo.
27
+ echo ERROR: JAVA_HOME is not set and no 'java' command could be found in your PATH.
28
+ echo.
29
+ echo Please set the JAVA_HOME variable in your environment to match the
30
+ echo location of your Java installation.
31
+
32
+ goto fail
33
+
34
+ :findJavaFromJavaHome
35
+ set JAVA_HOME=%JAVA_HOME:"=%
36
+ set JAVA_EXE=%JAVA_HOME%/bin/java.exe
37
+
38
+ if exist "%JAVA_EXE%" goto init
39
+
40
+ echo.
41
+ echo ERROR: JAVA_HOME is set to an invalid directory: %JAVA_HOME%
42
+ echo.
43
+ echo Please set the JAVA_HOME variable in your environment to match the
44
+ echo location of your Java installation.
45
+
46
+ goto fail
47
+
48
+ :init
49
+ @rem Get command-line arguments, handling Windows variants
50
+
51
+ if not "%OS%" == "Windows_NT" goto win9xME_args
52
+
53
+ :win9xME_args
54
+ @rem Slurp the command line arguments.
55
+ set CMD_LINE_ARGS=
56
+ set _SKIP=2
57
+
58
+ :win9xME_args_slurp
59
+ if "x%~1" == "x" goto execute
60
+
61
+ set CMD_LINE_ARGS=%*
62
+
63
+ :execute
64
+ @rem Setup the command line
65
+
66
+ set CLASSPATH=%APP_HOME%\gradle\wrapper\gradle-wrapper.jar
67
+
68
+ @rem Execute Gradle
69
+ "%JAVA_EXE%" %DEFAULT_JVM_OPTS% %JAVA_OPTS% %GRADLE_OPTS% "-Dorg.gradle.appname=%APP_BASE_NAME%" -classpath "%CLASSPATH%" org.gradle.wrapper.GradleWrapperMain %CMD_LINE_ARGS%
70
+
71
+ :end
72
+ @rem End local scope for the variables with windows NT shell
73
+ if "%ERRORLEVEL%"=="0" goto mainEnd
74
+
75
+ :fail
76
+ rem Set variable GRADLE_EXIT_CONSOLE if you need the _script_ return code instead of
77
+ rem the _cmd.exe /c_ return code!
78
+ if not "" == "%GRADLE_EXIT_CONSOLE%" exit 1
79
+ exit /b 1
80
+
81
+ :mainEnd
82
+ if "%OS%"=="Windows_NT" endlocal
83
+
84
+ :omega
@@ -0,0 +1,3 @@
1
+ Embulk::JavaPlugin.register_filter(
2
+ "gsub", "org.embulk.filter.gsub.GsubFilterPlugin",
3
+ File.expand_path('../../../../classpath', __FILE__))
@@ -0,0 +1,20 @@
1
+ package org.embulk.filter.gsub
2
+
3
+ import org.embulk.filter.gsub.replacer.*
4
+
5
+ class ColumnReplacerFactory {
6
+ fun create(task: GsubFilterPlugin.PluginTask): Map<String, TextReplacer> {
7
+ return task.targetColumns.mapValues { column ->
8
+ createReplacerForRules(column.value)
9
+ }
10
+ }
11
+
12
+ private fun createReplacerForRules(rules: List<SubstitutionRule>): TextReplacer {
13
+ val replacers = rules.map { createReplacer(it) }
14
+ return CombinedReplacer(replacers)
15
+ }
16
+
17
+ private fun createReplacer(rule: SubstitutionRule): TextReplacer {
18
+ return TextReplacerFactory.create(rule)
19
+ }
20
+ }
@@ -0,0 +1,72 @@
1
+ package org.embulk.filter.gsub
2
+
3
+ import org.embulk.filter.gsub.GsubFilterPlugin.PluginTask
4
+ import org.embulk.filter.gsub.replacer.TextReplacer
5
+ import org.embulk.spi.*
6
+
7
+ class ColumnVisitorImpl
8
+ constructor(
9
+ task: PluginTask,
10
+ private val pageReader: PageReader,
11
+ private val pageBuilder: PageBuilder
12
+ ) : ColumnVisitor {
13
+
14
+ private val columnReplacers: Map<String, TextReplacer> = ColumnReplacerFactory().create(task)
15
+
16
+ override fun booleanColumn(column: Column) {
17
+ if (pageReader.isNull(column)) {
18
+ pageBuilder.setNull(column)
19
+ }
20
+ else {
21
+ pageBuilder.setBoolean(column, pageReader.getBoolean(column))
22
+ }
23
+ }
24
+
25
+ override fun longColumn(column: Column) {
26
+ if (pageReader.isNull(column)) {
27
+ pageBuilder.setNull(column)
28
+ }
29
+ else {
30
+ pageBuilder.setLong(column, pageReader.getLong(column))
31
+ }
32
+ }
33
+
34
+ override fun doubleColumn(column: Column) {
35
+ if (pageReader.isNull(column)) {
36
+ pageBuilder.setNull(column)
37
+ }
38
+ else {
39
+ pageBuilder.setDouble(column, pageReader.getDouble(column))
40
+ }
41
+ }
42
+
43
+ override fun stringColumn(column: Column) {
44
+ if (pageReader.isNull(column)) {
45
+ pageBuilder.setNull(column)
46
+ }
47
+ else {
48
+ val text = pageReader.getString(column)
49
+ val replacer = columnReplacers[column.name]
50
+ val replacedText = replacer?.let { it.execute(text) } ?: text
51
+ pageBuilder.setString(column, replacedText)
52
+ }
53
+ }
54
+
55
+ override fun timestampColumn(column: Column) {
56
+ if (pageReader.isNull(column)) {
57
+ pageBuilder.setNull(column)
58
+ }
59
+ else {
60
+ pageBuilder.setTimestamp(column, pageReader.getTimestamp(column))
61
+ }
62
+ }
63
+
64
+ override fun jsonColumn(column: Column) {
65
+ if (pageReader.isNull(column)) {
66
+ pageBuilder.setNull(column)
67
+ }
68
+ else {
69
+ pageBuilder.setJson(column, pageReader.getJson(column))
70
+ }
71
+ }
72
+ }
@@ -0,0 +1,51 @@
1
+ package org.embulk.filter.gsub
2
+
3
+ import org.embulk.config.Config
4
+ import org.embulk.config.ConfigDefault
5
+ import org.embulk.config.ConfigSource
6
+ import org.embulk.config.Task
7
+ import org.embulk.config.TaskSource
8
+ import org.embulk.spi.*
9
+
10
+ class GsubFilterPlugin : FilterPlugin {
11
+ interface PluginTask : Task {
12
+ @get:Config("target_columns")
13
+ @get:ConfigDefault("{}")
14
+ val targetColumns: Map<String, List<SubstitutionRule>>
15
+ }
16
+
17
+ override fun transaction(config: ConfigSource, inputSchema: Schema,
18
+ control: FilterPlugin.Control) {
19
+ val task = config.loadConfig<PluginTask>(PluginTask::class.java)
20
+
21
+ control.run(task.dump(), inputSchema)
22
+ }
23
+
24
+ override fun open(taskSource: TaskSource, inputSchema: Schema,
25
+ outputSchema: Schema, output: PageOutput): PageOutput {
26
+ val task = taskSource.loadTask<PluginTask>(PluginTask::class.java)
27
+
28
+ return object: PageOutput {
29
+ val pageReader = PageReader(inputSchema)
30
+ val pageBuilder = PageBuilder(Exec.getBufferAllocator(), outputSchema, output)
31
+ val columnVisitor = ColumnVisitorImpl(task, pageReader, pageBuilder)
32
+
33
+ override fun add(page: Page) {
34
+ pageReader.setPage(page)
35
+
36
+ while (pageReader.nextRecord()) {
37
+ inputSchema.visitColumns(columnVisitor)
38
+ pageBuilder.addRecord()
39
+ }
40
+ }
41
+
42
+ override fun finish() {
43
+ pageBuilder.finish()
44
+ }
45
+
46
+ override fun close() {
47
+ pageBuilder.close()
48
+ }
49
+ }
50
+ }
51
+ }
@@ -0,0 +1,23 @@
1
+ package org.embulk.filter.gsub
2
+
3
+ import org.embulk.filter.gsub.replacer.LowerCaseReplacer
4
+ import org.embulk.filter.gsub.replacer.RegexFactory
5
+ import org.embulk.filter.gsub.replacer.RegexOptionConfig
6
+ import org.embulk.filter.gsub.replacer.TextReplacer
7
+
8
+ class LowerCaseReplacerFactory : TextReplacerFactory() {
9
+ override fun create(rule: SubstitutionRule): TextReplacer {
10
+ val pattern = rule.pattern.orNull()
11
+ if (pattern != null) {
12
+ // TODO set regex options
13
+ val regexOptionConfig = RegexOptionConfig()
14
+ val factory = RegexFactory()
15
+ val regex = factory.create(pattern, regexOptionConfig)
16
+
17
+ return LowerCaseReplacer(regex)
18
+ }
19
+ else {
20
+ return LowerCaseReplacer()
21
+ }
22
+ }
23
+ }
@@ -0,0 +1,20 @@
1
+ package org.embulk.filter.gsub
2
+
3
+ import org.embulk.filter.gsub.replacer.RegexFactory
4
+ import org.embulk.filter.gsub.replacer.RegexOptionConfig
5
+ import org.embulk.filter.gsub.replacer.RegexReplacer
6
+ import org.embulk.filter.gsub.replacer.TextReplacer
7
+
8
+ class RegexReplacerFactory : TextReplacerFactory() {
9
+ override fun create(rule: SubstitutionRule): TextReplacer {
10
+ val pattern = rule.pattern.get()
11
+ val to = rule.to.get()
12
+
13
+ val regexOptionConfig = RegexOptionConfig()
14
+
15
+ val factory = RegexFactory()
16
+ val regex = factory.create(pattern, regexOptionConfig)
17
+
18
+ return RegexReplacer(regex, to)
19
+ }
20
+ }
@@ -0,0 +1,26 @@
1
+ package org.embulk.filter.gsub
2
+
3
+ import com.google.common.base.Optional
4
+ import org.embulk.config.Config
5
+ import org.embulk.config.ConfigDefault
6
+ import org.embulk.config.Task
7
+
8
+ interface SubstitutionRule : Task {
9
+ enum class SubstitutionType(val label: String) {
10
+ REGEXP_REPLACE("regexp_replace"),
11
+ TO_UPPER_CASE("to_upper_case"),
12
+ TO_LOWER_CASE("to_lower_case"),
13
+ }
14
+
15
+ @get:Config("type")
16
+ @get:ConfigDefault("\"regexp_replace\"")
17
+ val type: String
18
+
19
+ @get:Config("pattern")
20
+ @get:ConfigDefault("null")
21
+ val pattern: Optional<String>
22
+
23
+ @get:Config("to")
24
+ @get:ConfigDefault("null")
25
+ val to: Optional<String>
26
+ }
@@ -0,0 +1,33 @@
1
+ package org.embulk.filter.gsub
2
+
3
+ import org.embulk.filter.gsub.replacer.TextReplacer
4
+ import org.embulk.filter.gsub.SubstitutionRule.SubstitutionType
5
+
6
+ abstract class TextReplacerFactory {
7
+ abstract fun create(rule: SubstitutionRule): TextReplacer
8
+
9
+ companion object Factory {
10
+ fun create(rule: SubstitutionRule): TextReplacer {
11
+ val factory = createFactory(rule)
12
+ return factory.create(rule)
13
+ }
14
+
15
+ private fun createFactory(rule: SubstitutionRule): TextReplacerFactory {
16
+ val type = findSubstitutionType(rule.type)
17
+ return when (type) {
18
+ SubstitutionType.REGEXP_REPLACE -> RegexReplacerFactory()
19
+ SubstitutionType.TO_UPPER_CASE -> UpperCaseReplacerFactory()
20
+ SubstitutionType.TO_LOWER_CASE -> LowerCaseReplacerFactory()
21
+ }
22
+ }
23
+
24
+ private fun findSubstitutionType(typeName: String): SubstitutionType {
25
+ try {
26
+ return SubstitutionType.valueOf(typeName.toUpperCase())
27
+ }
28
+ catch (e: IllegalArgumentException) {
29
+ throw RuntimeException("Unknown substitution type: ${typeName}", e)
30
+ }
31
+ }
32
+ }
33
+ }
@@ -0,0 +1,23 @@
1
+ package org.embulk.filter.gsub
2
+
3
+ import org.embulk.filter.gsub.replacer.RegexFactory
4
+ import org.embulk.filter.gsub.replacer.RegexOptionConfig
5
+ import org.embulk.filter.gsub.replacer.TextReplacer
6
+ import org.embulk.filter.gsub.replacer.UpperCaseReplacer
7
+
8
+ class UpperCaseReplacerFactory : TextReplacerFactory() {
9
+ override fun create(rule: SubstitutionRule): TextReplacer {
10
+ val pattern = rule.pattern.orNull()
11
+ if (pattern != null) {
12
+ // TODO set regex options
13
+ val regexOptionConfig = RegexOptionConfig()
14
+ val factory = RegexFactory()
15
+ val regex = factory.create(pattern, regexOptionConfig)
16
+
17
+ return UpperCaseReplacer(regex)
18
+ }
19
+ else {
20
+ return UpperCaseReplacer()
21
+ }
22
+ }
23
+ }
@@ -0,0 +1,7 @@
1
+ package org.embulk.filter.gsub.replacer
2
+
3
+ class CombinedReplacer(private val replacers: List<TextReplacer>) : TextReplacer {
4
+ override fun execute(text: String): String {
5
+ return replacers.fold(text, { str, replacer -> replacer.execute(str) })
6
+ }
7
+ }
@@ -0,0 +1,17 @@
1
+ package org.embulk.filter.gsub.replacer
2
+
3
+ class LowerCaseReplacer(private val pattern: Regex?) : TextReplacer {
4
+ constructor() : this(null)
5
+
6
+ override fun execute(text: String): String {
7
+ return pattern?.let { replaceWithPattern(it, text)} ?: replaceWholeText(text)
8
+ }
9
+
10
+ private fun replaceWithPattern(pattern: Regex, text: String): String {
11
+ return pattern.replace(text, { matchResult -> matchResult.value.toLowerCase() })
12
+ }
13
+
14
+ private fun replaceWholeText(text: String): String {
15
+ return text.toLowerCase()
16
+ }
17
+ }
@@ -0,0 +1,22 @@
1
+ package org.embulk.filter.gsub.replacer
2
+
3
+ class RegexFactory {
4
+ fun create(patternString: String, regexOptionConfig: RegexOptionConfig): Regex {
5
+ val options = buildOptions(regexOptionConfig)
6
+ return Regex(patternString, options)
7
+ }
8
+
9
+ private fun buildOptions(optionConfig: RegexOptionConfig): Set<RegexOption> {
10
+ val options = HashSet<RegexOption>()
11
+
12
+ if (optionConfig.ignoreCase) {
13
+ options.add(RegexOption.IGNORE_CASE)
14
+ }
15
+
16
+ if (optionConfig.multiline) {
17
+ options.add(RegexOption.MULTILINE)
18
+ }
19
+
20
+ return options
21
+ }
22
+ }
@@ -0,0 +1,6 @@
1
+ package org.embulk.filter.gsub.replacer
2
+
3
+ class RegexOptionConfig {
4
+ var ignoreCase: Boolean = false
5
+ var multiline: Boolean = false
6
+ }
@@ -0,0 +1,7 @@
1
+ package org.embulk.filter.gsub.replacer
2
+
3
+ class RegexReplacer(private val pattern: Regex, private val to: String) : TextReplacer {
4
+ override fun execute(text: String): String {
5
+ return pattern.replace(text, to)
6
+ }
7
+ }
@@ -0,0 +1,5 @@
1
+ package org.embulk.filter.gsub.replacer
2
+
3
+ interface TextReplacer {
4
+ fun execute(text: String): String
5
+ }