Steven Christou created IO-453:
----------------------------------
Summary: Regression in FileUtils.readFileToString from 2.0.1
Key: IO-453
URL: https://issues.apache.org/jira/browse/IO-453
Project: Commons IO
Issue Type: Bug
Components: Utilities
Affects Versions: 2.4, 2.3, 2.2, 2.1
Reporter: Steven Christou
FileUtils.readFileToString has changed it's behavior to make a call from
{{IOUtils.toByteArray(in)}} to {{IOUtils.toByteArray(in, file.length())}} in
{{FileUtils.readFileToString}}. This is a regression because if the file.length
= 0, then it will return 0. According to the javadocs for
[File#length|http://docs.oracle.com/javase/6/docs/api/java/io/File.html#length()],
it is possible to return 0 if it is a System dependent entities, so even
though the File.length might return 0, the stream is still open and
Steps to reproduce (Ubuntu):
1. Execute {{nohup sleep 10000 &}} in a terminal, and get the process id of the
sleep command ({{ps -ef | grep sleep}}).
2. Call {{FileUtils.readFileToString(new File("/proc/$PID/environ"));}} where
$PID is the process ID from step 1.
You will notice that in 2.0.1 it returns several elements, however in 2.1 it
will return nothing.
See [~ndeloof]'s comment in
https://github.com/apache/commons-io/commit/53a40a6d9dcaaa616b404255406edc30fe2d524c.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
