embulk-filter-row 0.4.0 → 0.5.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 04f141506053778ed4ff092894f87c9f22acca3d
-  data.tar.gz: 8f7435bc29f4c0e97e44141771a4ce608edb429a
+  metadata.gz: 54615ee2184dfcb3791aa6d92922f1f64a9905a5
+  data.tar.gz: 5558f2a44121b9fdbfcee09d4f1d048ef934d20c
 SHA512:
-  metadata.gz: 5f27041d19e78633e4eedac7426d8d96e2180e047e0793f7c37b55929ba1f8da95f10966ac4643cd0dd690fe6e63ca8c82b95b0afaac64d2d96a89440f97bcf8
-  data.tar.gz: c12e1cca2eed39fa039180237808bc898e415af8a03b889c04ebe010eadcbbfd84cd82d4aaa62d18f2d17f9d71cb2382055e056c2ef30feebeceaf4ed53cc181
+  metadata.gz: 64168aceba7a817f404a6520c3e6ba5e499e9797bb6dcf97ca71eec26684a40f93f94a687b3e8284799c4ab6e09343cef416f68159adca30546c919da8f29390
+  data.tar.gz: be28eba1b0afa8edbeae6c2857ee645ce9319b2cf10eac6ce57720321789121676ce3706dc280051bd887b51a05b61d58be6913e7638670fb1b210e92a31a852

data/CHANGELOG.md

@@ -1,3 +1,10 @@
+# 0.5.0 (2017-08-23)
+
+Changes
+
+* Follow new TimestampParser API of embulk >= 0.8.29.
+  * Note that embulk-filter-row >= 0.5.0 requires embulk >= 0.8.29.
+
 # 0.4.0 (2017-05-18)
 
 Enhancements

data/README.md

@@ -6,7 +6,7 @@ A filter plugin for Embulk to filter out rows
 
 ## Configuration
 
-Versions >= 0.3.0 has `where` option to support SQL-like syntax.
+Requirement: version >= 0.3.0
 
 * **where**: Select only rows which matches with conditions written in SQL-like syntax. See [SQL-like Syntax](#sql-like-syntax)
 
@@ -23,13 +23,13 @@ filters:
   - type: row
     where: |-
       (
-        string START_WITH 'str' AND
-        number > 1.0
+        string_column START_WITH 'str' AND
+        number_column > 1.0
       )
       OR
       (
-        time = TIMESTAMP '2016-01-01 +0900' AND
-        "true" = true
+        time_column = TIMESTAMP '2016-01-01 +0900' AND
+        "true_column" = true
       )
 ```
 
@@ -37,20 +37,22 @@ See [SQL-like Syntax](#sql-like-syntax) for more details
 
 # SQL-like Syntax
 
-Versions >= 0.3.0 suppors SQL-like syntax as:
+```sql
+where: column1 = 'str'
+```
 
 ```sql
 where: |-
   (
-    string START_WITH 'str' AND
-    number > 1.0
+    string_column START_WITH 'str' AND
+    number_column > 1.0
   
   )
   OR
   (
-    time = TIMESTAMP '2016-01-01 +0900' AND
-    "true" = true AND
-    string REGEXP '^reg'
+    time_column = TIMESTAMP '2016-01-01 +0900' AND
+    "true_column" = true AND
+    string_column REGEXP '^reg'
   )
 ```
 
@@ -227,8 +229,9 @@ filters:
 ## Comparisions
 
 * [embulk-filter-calcite](https://github.com/muga/embulk-filter-calcite)
-  * embulk-filter-calcite is a pretty nice plugin which enables us to write SQL query to filter embulk records.
-  * However, based on [my benchmark (Japanese)](http://qiita.com/sonots/items/a70482d29862de87624d), embulk-filter-calcite was 1471 times slower than embulk-filter-row in iterms of string `=` operator. Use `embulk-filter-row` (this plugin) when you need performance.
+  * embulk-filter-calcite is a pretty nice plugin which enables us to write SQL query to filter embulk records, not only `WHERE` but also `SELECT`.
+  * Based on [my benchmark (Japanese)](http://qiita.com/sonots/items/a70482d29862de87624d), embulk-filter-row was faster than embulk-filter-calcite.
+  * Choose which to use as your demand.
 
 ## ToDo
 

data/build.gradle

@@ -15,13 +15,13 @@ configurations {
     provided
 }
 
-version = "0.4.0"
+version = "0.5.0"
 sourceCompatibility = 1.7
 targetCompatibility = 1.7
 
 dependencies {
-    compile  "org.embulk:embulk-core:0.8.+"
-    provided "org.embulk:embulk-core:0.8.+"
+    compile  "org.embulk:embulk-core:0.8.29+"
+    provided "org.embulk:embulk-core:0.8.29+"
 
     compile  "org.jruby.joni:joni:2.1.11"
     compile  "org.jruby.jcodings:jcodings:1.0.18"

data/src/main/java/org/embulk/filter/row/AbstractGuardColumnVisitor.java

@@ -40,7 +40,7 @@ abstract class AbstractGuardColumnVisitor
             String columnName = conditionConfig.getColumn();
             for (Column column : outputSchema.getColumns()) {
                 if (columnName.equals(column.getName())) {
-                    ConditionFactory factory = new ConditionFactory(task.getJRuby(), column, conditionConfig);
+                    ConditionFactory factory = new ConditionFactory(column, conditionConfig);
                     Condition condition = factory.createCondition();
                     conditionMap.get(columnName).add(condition);
                     break;

data/src/main/java/org/embulk/filter/row/RowFilterPlugin.java

@@ -55,7 +55,6 @@ public class RowFilterPlugin implements FilterPlugin
             FilterPlugin.Control control)
     {
         PluginTask task = config.loadConfig(PluginTask.class);
-        ParserLiteral.setJRuby(task.getJRuby());
 
         configure(task, inputSchema);
         Schema outputSchema = inputSchema;

data/src/main/java/org/embulk/filter/row/condition/ConditionFactory.java

@@ -15,11 +15,9 @@ import org.embulk.spi.type.StringType;
 import org.embulk.spi.type.TimestampType;
 import org.embulk.spi.type.Type;
 import org.joda.time.DateTimeZone;
-import org.jruby.embed.ScriptingContainer;
 
 public class ConditionFactory
 {
-    private final ScriptingContainer jruby;
     private Column column;
     private String columnName;
     private Type columnType;
@@ -27,9 +25,8 @@ public class ConditionFactory
     private String operator;
     private boolean not;
 
-    public ConditionFactory(ScriptingContainer jruby, Column column, ConditionConfig conditionConfig)
+    public ConditionFactory(Column column, ConditionConfig conditionConfig)
     {
-        this.jruby           = jruby;
         this.column          = column;
         this.columnName      = column.getName();
         this.columnType      = column.getType();
@@ -140,7 +137,7 @@ public class ConditionFactory
             String format          = (String) conditionConfig.getFormat().get();
             DateTimeZone timezone  = DateTimeZone.forID((String) conditionConfig.getTimezone().get());
 
-            TimestampParser parser = new TimestampParser(jruby, format, timezone);
+            TimestampParser parser = new TimestampParser(format, timezone);
             try {
                 Timestamp timestamp = parser.parse(argument);
                 return new TimestampCondition(operator, timestamp, not);

data/src/main/java/org/embulk/filter/row/where/ParserLiteral.java

@@ -16,20 +16,13 @@ import org.embulk.spi.type.StringType;
 import org.embulk.spi.type.TimestampType;
 import org.embulk.spi.type.Type;
 import org.joda.time.DateTimeZone;
-import org.jruby.embed.ScriptingContainer;
 import org.msgpack.value.Value;
 
 // Literal Node of AST (Abstract Syntax Tree)
 public abstract class ParserLiteral extends ParserNode
 {
-    protected static ScriptingContainer jruby;
     protected String yytext;
 
-    public static void setJRuby(ScriptingContainer jruby)
-    {
-        ParserLiteral.jruby = jruby;
-    }
-
     public boolean isBoolean()
     {
         return false;
@@ -185,7 +178,7 @@ class TimestampLiteral extends ParserLiteral
         TimestampParseException ex = null;
         for (String format : formats) {
             try {
-                TimestampParser timestampParser = new TimestampParser(jruby, format, default_timezone);
+                TimestampParser timestampParser = new TimestampParser(format, default_timezone);
                 this.val = timestampParser.parse(literal.val);
                 break;
             }

data/src/test/java/org/embulk/filter/row/condition/TestConditionFactory.java

@@ -6,7 +6,6 @@ import org.embulk.config.ConfigException;
 import org.embulk.config.TaskSource;
 import org.embulk.spi.Column;
 
-import org.jruby.embed.ScriptingContainer;
 import org.junit.Test;
 
 import static org.embulk.spi.type.Types.BOOLEAN;
@@ -63,11 +62,8 @@ public class TestConditionFactory
         }
     }
 
-    private final ScriptingContainer jruby;
-
     public TestConditionFactory()
     {
-        jruby = new ScriptingContainer();
     }
 
     @Test
@@ -83,7 +79,7 @@ public class TestConditionFactory
                 return Optional.of("IS NULL");
             }
         };
-        condition = (BooleanCondition) new ConditionFactory(jruby, column, config).createCondition();
+        condition = (BooleanCondition) new ConditionFactory(column, config).createCondition();
         assertTrue(condition.compare(null));
 
         config = new DefaultConditionConfig() {
@@ -97,7 +93,7 @@ public class TestConditionFactory
             }
         };
         try {
-            condition = (BooleanCondition) new ConditionFactory(jruby, column, config).createCondition();
+            condition = (BooleanCondition) new ConditionFactory(column, config).createCondition();
             fail("Argument is required");
         }
         catch (ConfigException e) {
@@ -113,7 +109,7 @@ public class TestConditionFactory
                 return Optional.of((Object) new Boolean(true));
             }
         };
-        condition = (BooleanCondition) new ConditionFactory(jruby, column, config).createCondition();
+        condition = (BooleanCondition) new ConditionFactory(column, config).createCondition();
         assertTrue(condition.compare(new Boolean(true)));
 
         config = new DefaultConditionConfig() {
@@ -127,7 +123,7 @@ public class TestConditionFactory
             }
         };
         try {
-            condition = (BooleanCondition) new ConditionFactory(jruby, column, config).createCondition();
+            condition = (BooleanCondition) new ConditionFactory(column, config).createCondition();
             fail("Argument type mismatch");
         }
         catch (ConfigException e) {
@@ -147,7 +143,7 @@ public class TestConditionFactory
                 return Optional.of("IS NULL");
             }
         };
-        condition = (DoubleCondition) new ConditionFactory(jruby, column, config).createCondition();
+        condition = (DoubleCondition) new ConditionFactory(column, config).createCondition();
         assertTrue(condition.compare(null));
 
         config = new DefaultConditionConfig() {
@@ -161,7 +157,7 @@ public class TestConditionFactory
             }
         };
         try {
-            condition = (DoubleCondition) new ConditionFactory(jruby, column, config).createCondition();
+            condition = (DoubleCondition) new ConditionFactory(column, config).createCondition();
             fail("Argument is required");
         }
         catch (ConfigException e) {
@@ -177,7 +173,7 @@ public class TestConditionFactory
                 return Optional.of((Object) new Double(10));
             }
         };
-        condition = (DoubleCondition) new ConditionFactory(jruby, column, config).createCondition();
+        condition = (DoubleCondition) new ConditionFactory(column, config).createCondition();
         assertTrue(condition.compare(new Double(10)));
 
         config = new DefaultConditionConfig() {
@@ -191,7 +187,7 @@ public class TestConditionFactory
             }
         };
         try {
-            condition = (DoubleCondition) new ConditionFactory(jruby, column, config).createCondition();
+            condition = (DoubleCondition) new ConditionFactory(column, config).createCondition();
             fail("Argument type mismatch");
         }
         catch (ConfigException e) {
@@ -211,7 +207,7 @@ public class TestConditionFactory
                 return Optional.of("IS NULL");
             }
         };
-        condition = (LongCondition) new ConditionFactory(jruby, column, config).createCondition();
+        condition = (LongCondition) new ConditionFactory(column, config).createCondition();
         assertTrue(condition.compare(null));
 
         config = new DefaultConditionConfig() {
@@ -225,7 +221,7 @@ public class TestConditionFactory
             }
         };
         try {
-            condition = (LongCondition) new ConditionFactory(jruby, column, config).createCondition();
+            condition = (LongCondition) new ConditionFactory(column, config).createCondition();
             fail("Argument is required");
         }
         catch (ConfigException e) {
@@ -241,7 +237,7 @@ public class TestConditionFactory
                 return Optional.of((Object) new Long(10));
             }
         };
-        condition = (LongCondition) new ConditionFactory(jruby, column, config).createCondition();
+        condition = (LongCondition) new ConditionFactory(column, config).createCondition();
         assertTrue(condition.compare(new Long(10)));
 
         config = new DefaultConditionConfig() {
@@ -255,7 +251,7 @@ public class TestConditionFactory
             }
         };
         try {
-            condition = (LongCondition) new ConditionFactory(jruby, column, config).createCondition();
+            condition = (LongCondition) new ConditionFactory(column, config).createCondition();
             fail("Argument type mismatch");
         }
         catch (ConfigException e) {
@@ -275,7 +271,7 @@ public class TestConditionFactory
                 return Optional.of("IS NULL");
             }
         };
-        condition = (StringCondition) new ConditionFactory(jruby, column, config).createCondition();
+        condition = (StringCondition) new ConditionFactory(column, config).createCondition();
         assertTrue(condition.compare(null));
 
         config = new DefaultConditionConfig() {
@@ -289,7 +285,7 @@ public class TestConditionFactory
             }
         };
         try {
-            condition = (StringCondition) new ConditionFactory(jruby, column, config).createCondition();
+            condition = (StringCondition) new ConditionFactory(column, config).createCondition();
             fail("Argument is required");
         }
         catch (ConfigException e) {
@@ -305,7 +301,7 @@ public class TestConditionFactory
                 return Optional.of((Object) "foo");
             }
         };
-        condition = (StringCondition) new ConditionFactory(jruby, column, config).createCondition();
+        condition = (StringCondition) new ConditionFactory(column, config).createCondition();
         assertTrue(condition.compare("foo"));
 
         config = new DefaultConditionConfig() {
@@ -319,7 +315,7 @@ public class TestConditionFactory
             }
         };
         try {
-            condition = (StringCondition) new ConditionFactory(jruby, column, config).createCondition();
+            condition = (StringCondition) new ConditionFactory(column, config).createCondition();
             fail("Argument type mismatch");
         }
         catch (ConfigException e) {
@@ -339,7 +335,7 @@ public class TestConditionFactory
                 return Optional.of("IS NULL");
             }
         };
-        condition = (TimestampCondition) new ConditionFactory(jruby, column, config).createCondition();
+        condition = (TimestampCondition) new ConditionFactory(column, config).createCondition();
         assertTrue(condition.compare(null));
 
         config = new DefaultConditionConfig() {
@@ -353,20 +349,12 @@ public class TestConditionFactory
             }
         };
         try {
-            condition = (TimestampCondition) new ConditionFactory(jruby, column, config).createCondition();
+            condition = (TimestampCondition) new ConditionFactory(column, config).createCondition();
             fail("Argument is required");
         }
         catch (ConfigException e) {
         }
 
-        //ToDo: How to create jruby object correctly?
-        //config = new DefaultConditionConfig() {
-        //    public Optional<String> getOperator() { return Optional.of("=="); }
-        //    public Optional<Object> getArgument() { return Optional.of((Object)"2015-07-15"); }
-        //    public Optional<String> getFormat()   { return Optional.of("%Y-%m-%d"); }
-        //};
-        //condition = (TimestampCondition)new ConditionFactory(jruby, column, config).createCondition();
-
         config = new DefaultConditionConfig() {
             public Optional<String> getOperator()
             {
@@ -378,7 +366,7 @@ public class TestConditionFactory
             }
         };
         try {
-            condition = (TimestampCondition) new ConditionFactory(jruby, column, config).createCondition();
+            condition = (TimestampCondition) new ConditionFactory(column, config).createCondition();
             fail("Argument type mismatch");
         }
         catch (ConfigException e) {

data/src/test/java/org/embulk/filter/row/where/TestParser.java

@@ -11,7 +11,6 @@ import org.embulk.spi.SchemaConfigException;
 import org.embulk.spi.time.Timestamp;
 
 import org.embulk.spi.time.TimestampParseException;
-import org.jruby.embed.ScriptingContainer;
 import org.junit.BeforeClass;
 import org.junit.Rule;
 import org.junit.Test;
@@ -32,7 +31,6 @@ import static org.junit.Assert.assertTrue;
 public class TestParser
 {
     private static EmbulkTestRuntime runtime = new EmbulkTestRuntime(); // very slow
-    private static ScriptingContainer jruby = new ScriptingContainer();
 
     private static PageReader buildPageReader(Schema schema, final Object... objects)
     {
@@ -51,8 +49,6 @@ public class TestParser
     @BeforeClass
     public static void setupBeforeClass()
     {
-        ParserLiteral.setJRuby(jruby);
-
         // {"k1":{"k1":"v"},"k2":{"k2":"v"}}
         Value k1 = ValueFactory.newString("k1");
         Value k2 = ValueFactory.newString("k2");

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: embulk-filter-row
 version: !ruby/object:Gem::Version
-  version: 0.4.0
+  version: 0.5.0
 platform: ruby
 authors:
 - Naotoshi Seo
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2017-05-18 00:00:00.000000000 Z
+date: 2017-08-23 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -96,7 +96,7 @@ files:
 - src/test/java/org/embulk/filter/row/condition/TestTimestampCondition.java
 - src/test/java/org/embulk/filter/row/where/TestParser.java
 - src/test/java/org/embulk/filter/row/where/TestYylex.java
-- classpath/embulk-filter-row-0.4.0.jar
+- classpath/embulk-filter-row-0.5.0.jar
 - classpath/jcodings-1.0.18.jar
 - classpath/joni-2.1.11.jar
 homepage: https://github.com/sonots/embulk-filter-row