Return-Path: X-Original-To: apmail-pig-commits-archive@www.apache.org Delivered-To: apmail-pig-commits-archive@www.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id E387E737A for ; Sat, 5 Nov 2011 21:15:44 +0000 (UTC) Received: (qmail 17831 invoked by uid 500); 5 Nov 2011 21:15:44 -0000 Delivered-To: apmail-pig-commits-archive@pig.apache.org Received: (qmail 17802 invoked by uid 500); 5 Nov 2011 21:15:44 -0000 Mailing-List: contact commits-help@pig.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: dev@pig.apache.org Delivered-To: mailing list commits@pig.apache.org Received: (qmail 17795 invoked by uid 99); 5 Nov 2011 21:15:44 -0000 Received: from nike.apache.org (HELO nike.apache.org) (192.87.106.230) by apache.org (qpsmtpd/0.29) with ESMTP; Sat, 05 Nov 2011 21:15:44 +0000 X-ASF-Spam-Status: No, hits=-2000.0 required=5.0 tests=ALL_TRUSTED X-Spam-Check-By: apache.org Received: from [140.211.11.4] (HELO eris.apache.org) (140.211.11.4) by apache.org (qpsmtpd/0.29) with ESMTP; Sat, 05 Nov 2011 21:15:41 +0000 Received: from eris.apache.org (localhost [127.0.0.1]) by eris.apache.org (Postfix) with ESMTP id 7421C23888FE for ; Sat, 5 Nov 2011 21:15:20 +0000 (UTC) Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Subject: svn commit: r1198059 - in /pig/branches/branch-0.10: CHANGES.txt src/org/apache/pig/newplan/logical/optimizer/LogicalPlanOptimizer.java test/org/apache/pig/test/TestPlanGeneration.java Date: Sat, 05 Nov 2011 21:15:20 -0000 To: commits@pig.apache.org From: daijy@apache.org X-Mailer: svnmailer-1.0.8-patched Message-Id: <20111105211520.7421C23888FE@eris.apache.org> X-Virus-Checked: Checked by ClamAV on apache.org Author: daijy Date: Sat Nov 5 21:15:19 2011 New Revision: 1198059 URL: http://svn.apache.org/viewvc?rev=1198059&view=rev Log: PIG-2339: HCatLoader loads all the partitions in a partitioned table even though a filter clause on the partitions is specified in the Pig script Modified: pig/branches/branch-0.10/CHANGES.txt pig/branches/branch-0.10/src/org/apache/pig/newplan/logical/optimizer/LogicalPlanOptimizer.java pig/branches/branch-0.10/test/org/apache/pig/test/TestPlanGeneration.java Modified: pig/branches/branch-0.10/CHANGES.txt URL: http://svn.apache.org/viewvc/pig/branches/branch-0.10/CHANGES.txt?rev=1198059&r1=1198058&r2=1198059&view=diff ============================================================================== --- pig/branches/branch-0.10/CHANGES.txt (original) +++ pig/branches/branch-0.10/CHANGES.txt Sat Nov 5 21:15:19 2011 @@ -278,6 +278,9 @@ PIG-2320: Error: "projection with nothin PIG-2346: TypeCastInsert should not insert Foreach if there is no as statement (daijy) +PIG-2339: HCatLoader loads all the partitions in a partitioned table even though + a filter clause on the partitions is specified in the Pig script (daijy) + Release 0.9.1 - Unreleased IMPROVEMENTS Modified: pig/branches/branch-0.10/src/org/apache/pig/newplan/logical/optimizer/LogicalPlanOptimizer.java URL: http://svn.apache.org/viewvc/pig/branches/branch-0.10/src/org/apache/pig/newplan/logical/optimizer/LogicalPlanOptimizer.java?rev=1198059&r1=1198058&r2=1198059&view=diff ============================================================================== --- pig/branches/branch-0.10/src/org/apache/pig/newplan/logical/optimizer/LogicalPlanOptimizer.java (original) +++ pig/branches/branch-0.10/src/org/apache/pig/newplan/logical/optimizer/LogicalPlanOptimizer.java Sat Nov 5 21:15:19 2011 @@ -91,15 +91,6 @@ public class LogicalPlanOptimizer extend checkAndAddRule(s, r); if (!s.isEmpty()) ls.add(s); - - // Partition filter set - // This set of rules push partition filter to LoadFunc - s = new HashSet(); - // Optimize partition filter - r = new PartitionFilterOptimizer("PartitionFilterOptimizer"); - checkAndAddRule(s, r); - if (!s.isEmpty()) - ls.add(s); // Limit Set // This set of rules push up limit @@ -141,6 +132,15 @@ public class LogicalPlanOptimizer extend if (!s.isEmpty()) ls.add(s); + // Partition filter set + // This set of rules push partition filter to LoadFunc + s = new HashSet(); + // Optimize partition filter + r = new PartitionFilterOptimizer("PartitionFilterOptimizer"); + checkAndAddRule(s, r); + if (!s.isEmpty()) + ls.add(s); + // PushDownForEachFlatten set s = new HashSet(); // Add the PushDownForEachFlatten Modified: pig/branches/branch-0.10/test/org/apache/pig/test/TestPlanGeneration.java URL: http://svn.apache.org/viewvc/pig/branches/branch-0.10/test/org/apache/pig/test/TestPlanGeneration.java?rev=1198059&r1=1198058&r2=1198059&view=diff ============================================================================== --- pig/branches/branch-0.10/test/org/apache/pig/test/TestPlanGeneration.java (original) +++ pig/branches/branch-0.10/test/org/apache/pig/test/TestPlanGeneration.java Sat Nov 5 21:15:19 2011 @@ -251,4 +251,64 @@ public class TestPlanGeneration extends Assert.assertTrue(lp.getSuccessors(loStore)==null); } + + public static class PartitionedLoader extends PigStorage implements LoadMetadata { + + Schema schema; + String[] partCols; + static Expression partFilter = null; + + public PartitionedLoader(String schemaString, String commaSepPartitionCols) + throws ParserException { + schema = Utils.getSchemaFromString(schemaString); + partCols = commaSepPartitionCols.split(","); + } + + @Override + public ResourceSchema getSchema(String location, Job job) + throws IOException { + return new ResourceSchema(schema); + } + + @Override + public ResourceStatistics getStatistics(String location, + Job job) throws IOException { + return null; + } + + @Override + public void setPartitionFilter(Expression partitionFilter) + throws IOException { + partFilter = partitionFilter; + } + + @Override + public String[] getPartitionKeys(String location, Job job) + throws IOException { + return partCols; + } + + public Expression getPartFilter() { + return partFilter; + } + + } + + @Test + // See PIG-2339 + public void testPartitionFilterOptimizer() throws Exception { + String query = "a = load 'foo' using " + PartitionedLoader.class.getName() + + "('name:chararray, dt:chararray', 'dt');\n" + + "b = filter a by dt=='2011';\n" + + "store b into 'output';"; + + LogicalPlan lp = Util.parse(query, pc); + Util.optimizeNewLP(lp); + + LOLoad loLoad = (LOLoad)lp.getSources().get(0); + LOForEach loForEach = (LOForEach)lp.getSuccessors(loLoad).get(0); + LOStore loStore = (LOStore)lp.getSuccessors(loForEach).get(0); + Assert.assertTrue(((PartitionedLoader)loLoad.getLoadFunc()).getPartFilter()!=null); + Assert.assertTrue(loStore.getAlias().equals("b")); + } }