file backend isn't using context manager correctly in locking code

[sqlalchemy-bot]

Migrated issue, originally created by Michael Bayer (zzzeek)

#!python

diff -r 2421b6447c65442f6ae0d410629382701da035c3 dogpile/cache/backends/file.py
--- a/dogpile/cache/backends/file.py	Tue Nov 20 17:48:17 2012 -0500
+++ b/dogpile/cache/backends/file.py	Mon Dec 10 14:11:24 2012 -0500
@@ -213,14 +213,18 @@
     @contextmanager
     def read(self):
         self.acquire_read_lock(True)
-        yield
-        self.release_read_lock()
+        try:
+            yield
+        finally:
+            self.release_read_lock()
 
     @contextmanager
     def write(self):
         self.acquire_write_lock(True)
-        yield
-        self.release_write_lock()
+        try:
+            yield
+        finally:
+            self.release_write_lock()
 
     def acquire_read_lock(self, wait):
         return self._acquire(wait, os.O_RDONLY, fcntl.LOCK_SH)

[sqlalchemy-bot]

Michael Bayer (zzzeek) wrote:

c9c29f4

[sqlalchemy-bot]

Changes by Michael Bayer (zzzeek):

• changed status to closed