Samrat002 commented on code in PR #27788: URL: https://github.com/apache/flink/pull/27788#discussion_r3023394345
########## flink-filesystems/flink-s3-fs-native/src/main/java/org/apache/flink/fs/s3native/BucketConfigProvider.java: ########## @@ -0,0 +1,177 @@ +/* + * 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.flink.fs.s3native; + +import org.apache.flink.annotation.Internal; +import org.apache.flink.annotation.VisibleForTesting; +import org.apache.flink.configuration.Configuration; +import org.apache.flink.configuration.IllegalConfigurationException; + +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +import javax.annotation.Nullable; + +import java.util.Collections; +import java.util.HashMap; +import java.util.Map; + +/** + * Parses bucket-specific S3 configuration using format {@code s3.bucket.<bucket-name>.<property>}. + * + * <p>Enables per-bucket overrides for endpoints, credentials, encryption, and IAM roles. Bucket + * names containing dots are supported; properties are matched by longest suffix first. + * + * <p>Immutable and thread-safe after construction. + */ +@Internal +final class BucketConfigProvider { + + private static final Logger LOG = LoggerFactory.getLogger(BucketConfigProvider.class); + + static final String BUCKET_CONFIG_PREFIX = "s3.bucket."; + + /** + * Known bucket-level properties, sorted by descending length so that the longest match wins. + */ + private static final String[] KNOWN_PROPERTIES = + new String[] { + "assume-role.session-duration", + "assume-role.session-name", + "assume-role.external-id", + "credentials.provider", + "path-style-access", + "sse.kms-key-id", + "assume-role.arn", + "access-key", + "secret-key", + "sse.type", + "endpoint", + "region" + }; + + private final Map<String, S3BucketConfig> bucketConfigs; + + BucketConfigProvider(Configuration flinkConfig) { + this.bucketConfigs = Collections.unmodifiableMap(parseBucketConfigs(flinkConfig)); + } + + @Nullable + S3BucketConfig getBucketConfig(String bucketName) { + return bucketConfigs.get(bucketName); + } + + @VisibleForTesting + boolean hasBucketConfig(String bucketName) { + return bucketConfigs.containsKey(bucketName); + } + + @VisibleForTesting + int size() { + return bucketConfigs.size(); + } + + private static Map<String, S3BucketConfig> parseBucketConfigs(Configuration flinkConfig) { + Map<String, Map<String, String>> rawConfigs = new HashMap<>(); + + for (String key : flinkConfig.keySet()) { + if (!key.startsWith(BUCKET_CONFIG_PREFIX)) { + continue; + } + String suffix = key.substring(BUCKET_CONFIG_PREFIX.length()); + String value = flinkConfig.getString(key, null); + if (value == null) { + continue; + } + + for (String prop : KNOWN_PROPERTIES) { + if (suffix.endsWith("." + prop)) { + String bucketName = suffix.substring(0, suffix.length() - prop.length() - 1); + if (!bucketName.isEmpty()) { Review Comment: > Should we check that bucket name consists of valid characters? I'd prefer to leave that to the AWS SDK, which gives clear error messages and handles partition-specific naming rules. Validating here would couple us to rules that may change. For 'all' buckets the global config (s3.region, s3.endpoint, etc.) already serves as the default. Bucket-level config is purely for overrides. > Another question. Is it generally a legit situation to have empty bucket name? Should we log a warning to highlight the mistake if not? Good point on the empty bucket name. I have added a warning log for that case to help users catch typos like s3.bucket..region -- 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]
