anmolanmol1234 commented on code in PR #8611: URL: https://github.com/apache/hadoop/pull/8611#discussion_r3665116934
########## hadoop-tools/hadoop-azure/src/main/java/org/apache/hadoop/fs/azurebfs/contracts/services/ArrowListBlobParser.java: ########## @@ -0,0 +1,590 @@ +/** + * 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.hadoop.fs.azurebfs.contracts.services; + +import java.io.IOException; +import java.io.InputStream; +import java.nio.ByteBuffer; +import java.nio.channels.ReadableByteChannel; +import java.nio.charset.StandardCharsets; +import java.util.HashMap; +import java.util.List; +import java.util.Locale; +import java.util.Map; + +import org.apache.arrow.memory.BufferAllocator; +import org.apache.arrow.memory.RootAllocator; +import org.apache.arrow.vector.BitVector; +import org.apache.arrow.vector.FieldVector; +import org.apache.arrow.vector.VarCharVector; +import org.apache.arrow.vector.VectorSchemaRoot; +import org.apache.arrow.vector.complex.MapVector; +import org.apache.arrow.vector.ipc.ArrowStreamReader; +import org.apache.arrow.vector.types.pojo.Schema; +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +import org.apache.hadoop.fs.Path; +import org.apache.hadoop.fs.azurebfs.utils.DateTimeUtils; + +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.ARROW_COL_ACL; +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.ARROW_COL_CONTENT_LENGTH; +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.ARROW_COL_COPY_COMPLETION_TIME; +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.ARROW_COL_COPY_ID; +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.ARROW_COL_COPY_PROGRESS; +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.ARROW_COL_COPY_SOURCE; +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.ARROW_COL_COPY_STATUS; +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.ARROW_COL_COPY_STATUS_DESCRIPTION; +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.ARROW_COL_CREATION_TIME; +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.ARROW_COL_ETAG; +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.ARROW_COL_GROUP; +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.ARROW_COL_IS_DIRECTORY; +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.ARROW_COL_LAST_MODIFIED; +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.ARROW_COL_METADATA; +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.ARROW_COL_NAME; +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.ARROW_COL_OWNER; +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.ARROW_COL_PERMISSIONS; +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.ARROW_COL_RESOURCE_TYPE; +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.ARROW_METADATA_NEXT_MARKER; +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.ARROW_RESOURCE_TYPE_BLOB_PREFIX; +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.DIRECTORY; +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.FORWARD_SLASH; +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.ROOT_PATH; +import static org.apache.hadoop.fs.azurebfs.constants.AbfsHttpConstants.XML_TAG_HDI_ISFOLDER; +import static org.apache.hadoop.fs.azurebfs.constants.HttpHeaderConfigurations.X_MS_META_HDI_ISFOLDER; +import static org.apache.hadoop.fs.azurebfs.services.AbfsErrors.ERR_ARROW_LIST_PARSING; + +/** + * Photon response parser implementation for Apache Arrow based ListBlobs + * responses. The parser reads the Arrow IPC stream returned by the service, + * iterates through the record batches and converts each row into a + * {@link BlobListResultEntrySchema}, producing a {@link BlobListResultSchema} + * that keeps the downstream FileStatus conversion path unchanged. + * <p> + * Arrow specific objects are confined to this class; the output contract is + * {@link BlobListResultSchema} so existing listing behavior is preserved. + * <p> + * Column matching is case-insensitive. Unknown columns are ignored and missing + * columns are treated as absent values so that additive schema changes do not + * break parsing. The continuation token is read from the Arrow schema custom + * metadata (key {@code NextMarker}). + */ +public class ArrowListBlobParser implements ListBlobResponseParser { + + private static final Logger LOG = + LoggerFactory.getLogger(ArrowListBlobParser.class); + + /** + * Upper bound, in bytes, on the heap staging buffer allocated per read when + * copying into a direct (non-array-backed) {@link ByteBuffer}. Caps transient + * allocations for large reader requests; the reader simply issues additional + * reads for anything beyond this chunk. + */ + private static final int NON_INTERRUPTIBLE_READ_CHUNK = 8192; + + /** + * Base URL for which the ListBlobs API is called, used to build the absolute + * URL for each entry (mirrors the XML parser behavior). + */ + private final String url; + + /** + * Maximum off-heap (direct) memory in bytes the Arrow allocator may use while + * parsing a single response. Bounds the memory an oversized or malformed + * response can consume; exceeding it fails the parse with an + * {@link org.apache.arrow.memory.OutOfMemoryException} that is surfaced as an + * {@link IOException}. + */ + private final long memoryLimitBytes; + + /** + * @param url base URL for which the ListBlobs API is called. + * @param memoryLimitBytes maximum off-heap memory in bytes the Arrow allocator + * may use while parsing a single response. + */ + public ArrowListBlobParser(final String url, final long memoryLimitBytes) { + this.url = url; + this.memoryLimitBytes = memoryLimitBytes; + } + + @Override + public BlobListResultSchema parse(final InputStream responseStream) + throws IOException { + BlobListResultSchema listResultSchema = new BlobListResultSchema(); + try (BufferAllocator allocator = new RootAllocator(memoryLimitBytes); + ArrowStreamReader reader = + new ArrowStreamReader(nonInterruptibleChannel(responseStream), + allocator)) { + VectorSchemaRoot root = reader.getVectorSchemaRoot(); + Schema schema = root.getSchema(); + + // Continuation token is carried in the Arrow schema custom metadata. + // The service emits an empty string when there is no continuation token, + // whereas the XML path leaves it null; normalize an absent or empty + // marker to null so both formats produce identical values. + Map<String, String> customMetadata = schema.getCustomMetadata(); + if (customMetadata != null) { + String nextMarker = customMetadata.get(ARROW_METADATA_NEXT_MARKER); + listResultSchema.setNextMarker( + (nextMarker == null || nextMarker.isEmpty()) ? null : nextMarker); + } + + while (reader.loadNextBatch()) { + BatchColumns columns = BatchColumns.resolve(root); Review Comment: varChar() returns null when a column is missing or present-but-not-a-VarCharVector, so if that happens for Name, buildEntry() returns null for every row and parse() completes normally with zero entries — no exception, and PHOTON_PARSE_FAILURE_COUNT never fires. Callers can't distinguish that from a genuinely empty directory, so the failure surfaces later as missing data in Spark/Hive rather than as a listing error. Could we throw here when columns.name == null? Name is the one column buildEntry() can't proceed without, so the "missing columns are tolerated" rule shouldn't extend to it. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
