Detect corrupt downloads

build.py

@@ -62,6 +62,7 @@ SRC = [path
        if path.endswith('.js')]
 
 PLOVR_JAR = 'bin/plovr-eba786b34df9.jar'
+PLOVR_JAR_MD5 = '20eac8ccc4578676511cf7ccbfc65100'
 
 
 def report_sizes(t):
@@ -205,7 +206,7 @@ pake.virtual('plovr', PLOVR_JAR)
 
 @pake.target(PLOVR_JAR, clean=False)
 def plovr_jar(t):
-    t.download('https://plovr.googlecode.com/files/' + os.path.basename(PLOVR_JAR))
+    t.download('https://plovr.googlecode.com/files/' + os.path.basename(PLOVR_JAR), md5=PLOVR_JAR_MD5)
 
 
 @pake.target('gh-pages', 'hostexamples', 'doc', phony=True)

pake.py

@@ -2,6 +2,7 @@
 
 import collections
 import contextlib
+import hashlib
 import logging
 import optparse
 import os
@@ -126,8 +127,12 @@ class Target(object):
     def debug(self, *args, **kwargs):
         self.logger.debug(*args, **kwargs)
 
-    def download(self, url):
+    def download(self, url, md5=None):
         content = urllib2.urlopen(url).read()
+        if md5 and hashlib.md5(content).hexdigest() != md5:
+            raise pake.BuildError(t, 'corrupt download')
+        # FIXME Python on Windoze corrupts the content when writing it
+        # FIXME probably something to do with encodings
         with open(self.name, 'w') as f:
             f.write(content)