Add table name option to | teragrep exec bloom (#87)

* Add table name option to | teragrep exec bloom

Fix typo in bloomOptionParameter

Add table name option to | teragrep exec bloom

* Add regex parameter to | teragrep exec bloom

src/main/antlr4/imports/COMMAND_TERAGREP_MODE.g4

@@ -91,6 +91,7 @@ COMMAND_TERAGREP_MODE_ARCHIVE: 'archive';
 COMMAND_TERAGREP_MODE_SUMMARY: 'summary' -> pushMode(DEFAULT_MODE);
 COMMAND_TERAGREP_MODE_RETENTION: 'retention=' -> pushMode(GET_SPAN);
 COMMAND_TERAGREP_MODE_CODEC: 'codec=' -> pushMode(GET_STRING);
+COMMAND_TERAGREP_MODE_TABLE: 'table' -> pushMode(GET_FIELD);
 
 COMMAND_TERAGREP_MODE_HDFS_FORMAT: 'format=';
 COMMAND_TERAGREP_MODE_CSV_FORMAT: ('csv'|'CSV');

src/main/antlr4/imports/DPLParserTransform_teragrep.g4

@@ -161,7 +161,11 @@ t_getArchiveSummaryParameter
         ;
 
 t_bloomOptionParameter
-        : COMMAND_TERAGREP_MODE_UPDATE t_estimatesParameter? t_inputParameter? | COMMAND_TERAGREP_MODE_CREATE t_estimatesParameter? t_inputParameter? | COMMAND_TERAGREP_MODE_ESTIMATE t_inputParameter? t_outputParameter?
+        : COMMAND_TERAGREP_MODE_UPDATE t_tableParameter? t_regexParameter? t_estimatesParameter? t_inputParameter? | COMMAND_TERAGREP_MODE_CREATE t_tableParameter? t_regexParameter? t_estimatesParameter? t_inputParameter? | COMMAND_TERAGREP_MODE_ESTIMATE t_inputParameter? t_outputParameter?
+        ;
+
+t_tableParameter
+        : COMMAND_TERAGREP_MODE_TABLE fieldType
         ;
 
 t_hostParameter

src/test/java/com/teragrep/pth_03/tests/TeragrepSyntaxTests.java

@@ -79,7 +79,11 @@ public class TeragrepSyntaxTests {
             "teragrep_foreachbatch",
             "teragrep_foreachbatch_transformStatement",
             "teragrep_config_set",
-            "teragrep_config_get"
+            "teragrep_config_get",
+            "teragrep_bloom_create_table",
+            "teragrep_bloom_update_table",
+            "teragrep_bloom_create_regex",
+            "teragrep_bloom_update_regex"
     })
     public void teragrepSyntaxParseTest(String arg) {
         String fileName = "src/test/resources/antlr4/commands/teragrep/" + arg + ".txt";
@@ -362,6 +366,98 @@ void testHdfsSaveAllParameters(String arg) {
         assertEquals(1, pathNodes.getLength());
     }
 
+    @ParameterizedTest
+    @ValueSource(strings = {
+            "teragrep_bloom_create_table"
+    })
+    void testBloomCreateTableName(String arg) {
+        ParserStructureTestingUtility pstu = new ParserStructureTestingUtility();
+        String fileName = "src/test/resources/antlr4/commands/teragrep/" + arg + ".txt";
+        String tableParamPath = "/root/transformStatement/teragrepTransformation/t_execParameter/t_bloomModeParameter/t_bloomOptionParameter/t_tableParameter";
+        String tableNamePath = "/root/transformStatement/teragrepTransformation/t_execParameter/t_bloomModeParameter/t_bloomOptionParameter/t_tableParameter/fieldType/value";
+        String optionModePath = "/root/transformStatement/teragrepTransformation/t_execParameter/t_bloomModeParameter/t_bloomOptionParameter/value";
+
+        NodeList tableParamNodes = Assertions.assertDoesNotThrow(() -> (NodeList) pstu.xpathQueryFile(fileName, tableParamPath, false));
+        NodeList tableNameNodes = Assertions.assertDoesNotThrow(() -> (NodeList) pstu.xpathQueryFile(fileName, tableNamePath, false));
+        NodeList createNodes = Assertions.assertDoesNotThrow(() -> (NodeList) pstu.xpathQueryFile(fileName, optionModePath, false));
+
+        assertEquals(1, tableParamNodes.getLength());
+        assertEquals(1, tableNameNodes.getLength());
+        assertEquals(1, createNodes.getLength());
+
+        assertEquals("myTable", tableNameNodes.item(0).getTextContent());
+        assertEquals("create", createNodes.item(0).getTextContent());
+    }
+
+    @ParameterizedTest
+    @ValueSource(strings = {
+            "teragrep_bloom_update_table"
+    })
+    void testBloomUpdateTableName(String arg) {
+        ParserStructureTestingUtility pstu = new ParserStructureTestingUtility();
+        String fileName = "src/test/resources/antlr4/commands/teragrep/" + arg + ".txt";
+        String tableParamPath = "/root/transformStatement/teragrepTransformation/t_execParameter/t_bloomModeParameter/t_bloomOptionParameter/t_tableParameter";
+        String tableNamePath = "/root/transformStatement/teragrepTransformation/t_execParameter/t_bloomModeParameter/t_bloomOptionParameter/t_tableParameter/fieldType/value";
+        String optionModePath = "/root/transformStatement/teragrepTransformation/t_execParameter/t_bloomModeParameter/t_bloomOptionParameter/value";
+
+        NodeList tableParamNodes = Assertions.assertDoesNotThrow(() -> (NodeList) pstu.xpathQueryFile(fileName, tableParamPath, false));
+        NodeList tableNameNodes = Assertions.assertDoesNotThrow(() -> (NodeList) pstu.xpathQueryFile(fileName, tableNamePath, false));
+        NodeList updateNodes = Assertions.assertDoesNotThrow(() -> (NodeList) pstu.xpathQueryFile(fileName, optionModePath, false));
+
+        assertEquals(1, tableParamNodes.getLength());
+        assertEquals(1, tableNameNodes.getLength());
+        assertEquals(1, updateNodes.getLength());
+
+        assertEquals("myTable", tableNameNodes.item(0).getTextContent());
+        assertEquals("update", updateNodes.item(0).getTextContent());
+    }
+
+    @ParameterizedTest
+    @ValueSource(strings = {
+            "teragrep_bloom_create_regex"
+    })
+    void testBloomCreateRegex(String arg) {
+        ParserStructureTestingUtility pstu = new ParserStructureTestingUtility();
+        String fileName = "src/test/resources/antlr4/commands/teragrep/" + arg + ".txt";
+        String regexParamPath = "/root/transformStatement/teragrepTransformation/t_execParameter/t_bloomModeParameter/t_bloomOptionParameter/t_regexParameter";
+        String regexPath = "/root/transformStatement/teragrepTransformation/t_execParameter/t_bloomModeParameter/t_bloomOptionParameter/t_regexParameter/stringType/value";
+        String optionModePath = "/root/transformStatement/teragrepTransformation/t_execParameter/t_bloomModeParameter/t_bloomOptionParameter/value";
+
+        NodeList regexParamNodes = Assertions.assertDoesNotThrow(() -> (NodeList) pstu.xpathQueryFile(fileName, regexParamPath, false));
+        NodeList regexNodes = Assertions.assertDoesNotThrow(() -> (NodeList) pstu.xpathQueryFile(fileName, regexPath, false));
+        NodeList createNodes = Assertions.assertDoesNotThrow(() -> (NodeList) pstu.xpathQueryFile(fileName, optionModePath, false));
+
+        assertEquals(1, regexParamNodes.getLength());
+        assertEquals(1, regexNodes.getLength());
+        assertEquals(1, createNodes.getLength());
+
+        assertEquals("\\w{4}", regexNodes.item(0).getTextContent());
+        assertEquals("create", createNodes.item(0).getTextContent());
+    }
+
+    @ParameterizedTest
+    @ValueSource(strings = {
+            "teragrep_bloom_update_regex"
+    })
+    void testBloomUpdateRegex(String arg) {
+        ParserStructureTestingUtility pstu = new ParserStructureTestingUtility();
+        String fileName = "src/test/resources/antlr4/commands/teragrep/" + arg + ".txt";
+        String regexParamPath = "/root/transformStatement/teragrepTransformation/t_execParameter/t_bloomModeParameter/t_bloomOptionParameter/t_regexParameter";
+        String regexPath = "/root/transformStatement/teragrepTransformation/t_execParameter/t_bloomModeParameter/t_bloomOptionParameter/t_regexParameter/stringType/value";
+        String optionModePath = "/root/transformStatement/teragrepTransformation/t_execParameter/t_bloomModeParameter/t_bloomOptionParameter/value";
+
+        NodeList regexParamNodes = Assertions.assertDoesNotThrow(() -> (NodeList) pstu.xpathQueryFile(fileName, regexParamPath, false));
+        NodeList regexNodes = Assertions.assertDoesNotThrow(() -> (NodeList) pstu.xpathQueryFile(fileName, regexPath, false));
+        NodeList updateNodes = Assertions.assertDoesNotThrow(() -> (NodeList) pstu.xpathQueryFile(fileName, optionModePath, false));
+
+        assertEquals(1, regexParamNodes.getLength());
+        assertEquals(1, regexNodes.getLength());
+        assertEquals(1, updateNodes.getLength());
+
+        assertEquals("\\w{4}", regexNodes.item(0).getTextContent());
+        assertEquals("update", updateNodes.item(0).getTextContent());
+    }
+
     @ParameterizedTest
     @ValueSource(strings = {
             "teragrep_regexextract",

src/test/resources/antlr4/commands/teragrep/teragrep_bloom_create_regex.txt

@@ -0,0 +1,46 @@
+<!-- /*
+ * Teragrep Data Processing Language Parser Library PTH-03
+ * Copyright (C) 2019, 2020, 2021, 2022, 2023  Suomen Kanuuna Oy
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU Affero General Public License as published by
+ * the Free Software Foundation, either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU Affero General Public License for more details.
+ *
+ * You should have received a copy of the GNU Affero General Public License
+ * along with this program.  If not, see <https://github.com/teragrep/teragrep/blob/main/LICENSE>.
+ *
+ *
+ * Additional permission under GNU Affero General Public License version 3
+ * section 7
+ *
+ * If you modify this Program, or any covered work, by linking or combining it
+ * with other code, such other code is not for that reason alone subject to any
+ * of the requirements of the GNU Affero GPL version 3 as long as this Program
+ * is the same Program as licensed from Suomen Kanuuna Oy without any additional
+ * modifications.
+ *
+ * Supplemented terms under GNU Affero General Public License version 3
+ * section 7
+ *
+ * Origin of the software must be attributed to Suomen Kanuuna Oy. Any modified
+ * versions must be marked as "Modified version of" The Program.
+ *
+ * Names of the licensors and authors may not be used for publicity purposes.
+ *
+ * No rights are granted for use of trade names, trademarks, or service marks
+ * which are in The Program if any.
+ *
+ * Licensee must indemnify licensors and authors for any liability that these
+ * contractual assumptions impose on licensors and authors.
+ *
+ * To the extent this program is licensed as part of the Commercial versions of
+ * Teragrep, the applicable Commercial License may apply to this file if you as
+ * a licensee so wish it.
+ */ -->
+ | teragrep exec bloom create regex \w{4}

src/test/resources/antlr4/commands/teragrep/teragrep_bloom_create_table.txt

@@ -0,0 +1,46 @@
+<!-- /*
+ * Teragrep Data Processing Language Parser Library PTH-03
+ * Copyright (C) 2019, 2020, 2021, 2022, 2023  Suomen Kanuuna Oy
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU Affero General Public License as published by
+ * the Free Software Foundation, either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU Affero General Public License for more details.
+ *
+ * You should have received a copy of the GNU Affero General Public License
+ * along with this program.  If not, see <https://github.com/teragrep/teragrep/blob/main/LICENSE>.
+ *
+ *
+ * Additional permission under GNU Affero General Public License version 3
+ * section 7
+ *
+ * If you modify this Program, or any covered work, by linking or combining it
+ * with other code, such other code is not for that reason alone subject to any
+ * of the requirements of the GNU Affero GPL version 3 as long as this Program
+ * is the same Program as licensed from Suomen Kanuuna Oy without any additional
+ * modifications.
+ *
+ * Supplemented terms under GNU Affero General Public License version 3
+ * section 7
+ *
+ * Origin of the software must be attributed to Suomen Kanuuna Oy. Any modified
+ * versions must be marked as "Modified version of" The Program.
+ *
+ * Names of the licensors and authors may not be used for publicity purposes.
+ *
+ * No rights are granted for use of trade names, trademarks, or service marks
+ * which are in The Program if any.
+ *
+ * Licensee must indemnify licensors and authors for any liability that these
+ * contractual assumptions impose on licensors and authors.
+ *
+ * To the extent this program is licensed as part of the Commercial versions of
+ * Teragrep, the applicable Commercial License may apply to this file if you as
+ * a licensee so wish it.
+ */ -->
+ | teragrep exec bloom create table myTable

src/test/resources/antlr4/commands/teragrep/teragrep_bloom_update_regex.txt

@@ -0,0 +1,46 @@
+<!-- /*
+ * Teragrep Data Processing Language Parser Library PTH-03
+ * Copyright (C) 2019, 2020, 2021, 2022, 2023  Suomen Kanuuna Oy
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU Affero General Public License as published by
+ * the Free Software Foundation, either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU Affero General Public License for more details.
+ *
+ * You should have received a copy of the GNU Affero General Public License
+ * along with this program.  If not, see <https://github.com/teragrep/teragrep/blob/main/LICENSE>.
+ *
+ *
+ * Additional permission under GNU Affero General Public License version 3
+ * section 7
+ *
+ * If you modify this Program, or any covered work, by linking or combining it
+ * with other code, such other code is not for that reason alone subject to any
+ * of the requirements of the GNU Affero GPL version 3 as long as this Program
+ * is the same Program as licensed from Suomen Kanuuna Oy without any additional
+ * modifications.
+ *
+ * Supplemented terms under GNU Affero General Public License version 3
+ * section 7
+ *
+ * Origin of the software must be attributed to Suomen Kanuuna Oy. Any modified
+ * versions must be marked as "Modified version of" The Program.
+ *
+ * Names of the licensors and authors may not be used for publicity purposes.
+ *
+ * No rights are granted for use of trade names, trademarks, or service marks
+ * which are in The Program if any.
+ *
+ * Licensee must indemnify licensors and authors for any liability that these
+ * contractual assumptions impose on licensors and authors.
+ *
+ * To the extent this program is licensed as part of the Commercial versions of
+ * Teragrep, the applicable Commercial License may apply to this file if you as
+ * a licensee so wish it.
+ */ -->
+ | teragrep exec bloom update regex \w{4}

src/test/resources/antlr4/commands/teragrep/teragrep_bloom_update_table.txt

@@ -0,0 +1,46 @@
+<!-- /*
+ * Teragrep Data Processing Language Parser Library PTH-03
+ * Copyright (C) 2019, 2020, 2021, 2022, 2023  Suomen Kanuuna Oy
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU Affero General Public License as published by
+ * the Free Software Foundation, either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU Affero General Public License for more details.
+ *
+ * You should have received a copy of the GNU Affero General Public License
+ * along with this program.  If not, see <https://github.com/teragrep/teragrep/blob/main/LICENSE>.
+ *
+ *
+ * Additional permission under GNU Affero General Public License version 3
+ * section 7
+ *
+ * If you modify this Program, or any covered work, by linking or combining it
+ * with other code, such other code is not for that reason alone subject to any
+ * of the requirements of the GNU Affero GPL version 3 as long as this Program
+ * is the same Program as licensed from Suomen Kanuuna Oy without any additional
+ * modifications.
+ *
+ * Supplemented terms under GNU Affero General Public License version 3
+ * section 7
+ *
+ * Origin of the software must be attributed to Suomen Kanuuna Oy. Any modified
+ * versions must be marked as "Modified version of" The Program.
+ *
+ * Names of the licensors and authors may not be used for publicity purposes.
+ *
+ * No rights are granted for use of trade names, trademarks, or service marks
+ * which are in The Program if any.
+ *
+ * Licensee must indemnify licensors and authors for any liability that these
+ * contractual assumptions impose on licensors and authors.
+ *
+ * To the extent this program is licensed as part of the Commercial versions of
+ * Teragrep, the applicable Commercial License may apply to this file if you as
+ * a licensee so wish it.
+ */ -->
+ | teragrep exec bloom update table myTable