paul-rogers commented on a change in pull request #1618: DRILL-6950: Row set-based scan framework URL: https://github.com/apache/drill/pull/1618#discussion_r251219014
########## File path: exec/java-exec/src/main/java/org/apache/drill/exec/physical/impl/scan/file/FileMetadataColumnsParser.java ########## @@ -0,0 +1,166 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ +package org.apache.drill.exec.physical.impl.scan.file; + +import java.util.regex.Matcher; +import java.util.regex.Pattern; + +import org.apache.drill.exec.physical.impl.scan.project.ColumnProjection; +import org.apache.drill.exec.physical.impl.scan.project.ScanLevelProjection; +import org.apache.drill.exec.physical.impl.scan.project.ScanLevelProjection.ScanProjectionParser; +import org.apache.drill.exec.physical.rowSet.project.RequestedTuple.RequestedColumn; + +public class FileMetadataColumnsParser implements ScanProjectionParser { + + static final org.slf4j.Logger logger = org.slf4j.LoggerFactory.getLogger(FileMetadataColumnsParser.class); + + // Internal + + private final FileMetadataManager metadataManager; + private final Pattern partitionPattern; + private ScanLevelProjection builder; + + // Output + + boolean hasMetadata; + + public FileMetadataColumnsParser(FileMetadataManager metadataManager) { + this.metadataManager = metadataManager; + partitionPattern = Pattern.compile(metadataManager.partitionDesignator + "(\\d+)", Pattern.CASE_INSENSITIVE); + } + + @Override + public void bind(ScanLevelProjection builder) { + this.builder = builder; + } + + @Override + public boolean parse(RequestedColumn inCol) { + Matcher m = partitionPattern.matcher(inCol.name()); + if (m.matches()) { + return buildPartitionColumn(m, inCol); + } + + FileMetadataColumnDefn defn = metadataManager.fileMetadataColIndex.get(inCol.name()); + if (defn != null) { + return buildMetadataColumn(defn, inCol); + } + if (inCol.isWildcard()) { + buildWildcard(); + + // Don't consider this a match. + } + return false; + } + + private boolean buildPartitionColumn(Matcher m, RequestedColumn inCol) { + + // If the projected column is a map or array, then it shadows the + // partition column. Example: dir0.x, dir0[2]. + + if (! inCol.isSimple()) { + logger.warn("Partition column {} is shadowed by a projected {}", + inCol.name(), inCol.summary()); + return false; + } + if (builder.hasWildcard()) { + wildcardAndMetadataError(); + } + + // Partition column + + builder.addMetadataColumn( + new PartitionColumn( + inCol.name(), + Integer.parseInt(m.group(1)))); + hasMetadata = true; + return true; + } + + private boolean buildMetadataColumn(FileMetadataColumnDefn defn, + RequestedColumn inCol) { + + // If the projected column is a map or array, then it shadows the + // metadata column. Example: filename.x, filename[2]. + + if (! inCol.isSimple()) { + logger.warn("File metadata column {} is shadowed by a projected {}", + inCol.name(), inCol.summary()); + return false; + } + if (builder.hasWildcard()) { + wildcardAndMetadataError(); + } + + // File metadata (implicit) column + + builder.addMetadataColumn(new FileMetadataColumn(inCol.name(), defn)); + hasMetadata = true; + return true; + } + + private void buildWildcard() { + if (hasMetadata) { + wildcardAndMetadataError(); + } + if (metadataManager.useLegacyWildcardExpansion) { + + // Star column: this is a SELECT * query. + + // Old-style wildcard handling inserts all metadata columns in + // the scanner, removes them in Project. + // Fill in the file metadata columns. Can do here because the + // set is constant across all files. + + expandWildcard(); + hasMetadata = true; + } + } + + protected void expandWildcard() { + + // Legacy wildcard expansion: include the file metadata and + // file partitions for this file. + // This is a disadvantage for a * query: files at different directory + // levels will have different numbers of columns. Would be better to + // return this data as an array at some point. + // Append this after the *, keeping the * for later expansion. + + for (FileMetadataColumnDefn iCol : metadataManager.fileMetadataColDefns()) { + builder.addMetadataColumn(new FileMetadataColumn( + iCol.colName(), iCol)); + } + for (int i = 0; i < metadataManager.partitionCount(); i++) { + builder.addMetadataColumn(new PartitionColumn( + metadataManager.partitionName(i), i)); + } + } + + @Override + public void validate() { } + + @Override + public void validateColumn(ColumnProjection outCol) { } + + private void wildcardAndMetadataError() { + throw new IllegalArgumentException("Cannot select file metadata columns and `*` together"); Review comment: Interesting... I recall we had a debate about this about 14 months ago. At that time, the wildcard matched not just all table columns, but also the implicit cols. Something about Calcite can't differentiate between the two and did not allow "SELECT *, a, b, c". The code you pointed to its a year old, meaning it is after we had the discussion mentioned above. Does Calcite now allow the combined syntax? More important, did the code change to not have the wildcard match implicit columns? If the semantics changed I'll need to do a bit of thinking to get things aligned with the new rules. Let's pin down those rules first. So, I'll fix this in a separate commit to this PR. ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services
