aboutsummaryrefslogtreecommitdiff
path: root/python/pyspark/files.py
diff options
context:
space:
mode:
authorJosh Rosen <joshrosen@eecs.berkeley.edu>2013-01-21 16:42:24 -0800
committerJosh Rosen <joshrosen@eecs.berkeley.edu>2013-01-21 17:34:17 -0800
commitef711902c1f42db14c8ddd524195f0a9efb56e65 (patch)
treee770a7439d3983c13346cbd81aa1eeeef23e2571 /python/pyspark/files.py
parent506077c9938cd411842fe42404aa6b74b45b23a1 (diff)
downloadspark-ef711902c1f42db14c8ddd524195f0a9efb56e65.tar.gz
spark-ef711902c1f42db14c8ddd524195f0a9efb56e65.tar.bz2
spark-ef711902c1f42db14c8ddd524195f0a9efb56e65.zip
Don't download files to master's working directory.
This should avoid exceptions caused by existing files with different contents. I also removed some unused code.
Diffstat (limited to 'python/pyspark/files.py')
-rw-r--r--python/pyspark/files.py24
1 files changed, 24 insertions, 0 deletions
diff --git a/python/pyspark/files.py b/python/pyspark/files.py
new file mode 100644
index 0000000000..de1334f046
--- /dev/null
+++ b/python/pyspark/files.py
@@ -0,0 +1,24 @@
+import os
+
+
+class SparkFiles(object):
+ """
+ Resolves paths to files added through
+ L{addFile()<pyspark.context.SparkContext.addFile>}.
+
+ SparkFiles contains only classmethods; users should not create SparkFiles
+ instances.
+ """
+
+ _root_directory = None
+
+ def __init__(self):
+ raise NotImplementedError("Do not construct SparkFiles objects")
+
+ @classmethod
+ def get(cls, filename):
+ """
+ Get the absolute path of a file added through C{addFile()}.
+ """
+ path = os.path.join(SparkFiles._root_directory, filename)
+ return os.path.abspath(path)