dimitris.pierrakos
/
dnet-hadoop
forked from D-Net/dnet-hadoop
1
0
Fork 0
Code Pull Requests Activity

prevent the code to cicle indefinetly

This commit is contained in:
Miriam Baglioni 2020-10-30 15:01:25 +01:00
parent f747e303ac
commit b828587252
1 changed files with 5 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
dhp-workflows/dhp-graph-mapper/src/main/java/eu/dnetlib/dhp/oa/graph/dump/MakeTar.java
View File

@ -88,16 +88,16 @@ public class MakeTar implements Serializable {
write(fileSystem, inputPath, outputPath + ".tar", dir_name);
} else {
int partNum = 0;
long remainingBytes = sourceSize % bytesPerSplit;
RemoteIterator<LocatedFileStatus> fileStatusListIterator = fileSystem
.listFiles(
new Path(inputPath), true);
while (sourceSize > 0) {
TarArchiveOutputStream ar = getTar(fileSystem, outputPath + "_" + partNum + ".tar");
boolean next = fileStatusListIterator.hasNext();
while (sourceSize > 0 && next) {
TarArchiveOutputStream ar = getTar(fileSystem, outputPath + "_" + (partNum + 1) + ".tar");
long current_size = 0;
while (fileStatusListIterator.hasNext() && current_size < bytesPerSplit) {
while (next && current_size < bytesPerSplit) {
LocatedFileStatus fileStatus = fileStatusListIterator.next();
Path p = fileStatus.getPath();
@ -125,6 +125,7 @@ public class MakeTar implements Serializable {
ar.closeArchiveEntry();
}
next = fileStatusListIterator.hasNext();
}
sourceSize = sourceSize - current_size;