[
https://issues.apache.org/jira/browse/HADOOP-19254?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17904591#comment-17904591
]
ASF GitHub Bot commented on HADOOP-19254:
-----------------------------------------
HarshitGupta11 commented on code in PR #7197:
URL: https://github.com/apache/hadoop/pull/7197#discussion_r1878512626
##########
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/shell/BulkDeleteCommand.java:
##########
@@ -73,12 +105,40 @@ protected void processArguments(LinkedList<PathData> args)
throws IOException {
BufferedReader br = new BufferedReader(new
InputStreamReader(localFile.open(new Path(fileName))));
String line;
while((line = br.readLine()) != null) {
- pathList.add(new Path(line));
+ if(!line.startsWith("#")) {
+ pathList.add(new Path(line));
+ }
}
} else {
pathList.addAll(this.childArgs.stream().map(Path::new).collect(Collectors.toList()));
}
BulkDelete bulkDelete = basePath.fs.createBulkDelete(basePath.path);
- bulkDelete.bulkDelete(pathList);
+ deleteInBatches(bulkDelete, pathList);
+ }
+
+ static class Batch<T> {
Review Comment:
Done
##########
hadoop-common-project/hadoop-common/src/main/java/org/apache/hadoop/fs/shell/BulkDeleteCommand.java:
##########
@@ -73,12 +105,40 @@ protected void processArguments(LinkedList<PathData> args)
throws IOException {
BufferedReader br = new BufferedReader(new
InputStreamReader(localFile.open(new Path(fileName))));
String line;
while((line = br.readLine()) != null) {
- pathList.add(new Path(line));
+ if(!line.startsWith("#")) {
+ pathList.add(new Path(line));
+ }
}
} else {
pathList.addAll(this.childArgs.stream().map(Path::new).collect(Collectors.toList()));
}
BulkDelete bulkDelete = basePath.fs.createBulkDelete(basePath.path);
- bulkDelete.bulkDelete(pathList);
+ deleteInBatches(bulkDelete, pathList);
+ }
+
+ static class Batch<T> {
+ List<T> data;
+ int batchSize;
+ int currentLocation;
+
+ Batch(List<T> data, int batchSize) {
+ this.data = Collections.unmodifiableList(data);
+ this.batchSize = batchSize;
+ this.currentLocation = 0;
+ }
+
+ boolean hasNext() {
+ return this.currentLocation < this.data.size();
+ }
+
+ List<T> next() {
+ List<T> ret = new ArrayList<>();
+ int i = currentLocation;
+ for(; i < Math.min(currentLocation + batchSize, data.size()); i++)
{
Review Comment:
Made the change
> Implement bulk delete command as hadoop fs command operation
> -------------------------------------------------------------
>
> Key: HADOOP-19254
> URL: https://issues.apache.org/jira/browse/HADOOP-19254
> Project: Hadoop Common
> Issue Type: Improvement
> Components: fs
> Affects Versions: 3.4.1
> Reporter: Mukund Thakur
> Assignee: Harshit Gupta
> Priority: Major
> Labels: pull-request-available
>
> {code}
> hadoop fs -bulkdelete <base-url> <file>
> {code}
> Key uses
> * QE: Testing from python and other scripting languages
> * cluster maintenance: actual bulk deletion operations from the store
> one thought there: we MUST qualify paths with / elements: if a passed in path
> ends in /, it means "delete a marker", not "delete a dir"'. and if it doesn't
> have one then it's an object.. This makes it possible to be used to delete
> surplus markers or where there is a file above another file...cloudstore
> listobjects finds this
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
