Merge pull request #99 from mattlqx/catalog

Add catalog functionality to prevent re-downloads

Author: joekiller

README.md

@@ -10,7 +10,9 @@ An Amazon Web Services account and something in S3 to fetch.
 
 Multi-part S3 uploads do not put the MD5 of the content in the ETag header. If x-amz-meta-digest is provided in User-Defined Metadata on the S3 Object it is processed as if it were a Digest header (RFC 3230).
 
-The MD5 of the local file will be checked against the MD5 from x-amz-meta-digest if it is present.  It not it will check against the ETag.  If there is no match or the local file is absent it will be downloaded.
+The MD5 of the local file will be checked against the MD5 from x-amz-meta-digest if it is present.  If not it will check against the ETag.  If there is no match or the local file is absent it will be downloaded.
+
+By default, a catalog file in Chef's cache path will be kept for all downloaded files tracking their etag and md5 at time of download. If either of these don't match, the file will be downloaded. To disable this behavior, set `node['s3_file']['use_catalog']` to `false`.
 
 If credentials are not provided, s3_file will attempt to use the first instance profile associated with the instance. See documentation at http://docs.aws.amazon.com/IAM/latest/UserGuide/instance-profiles.html for more on instance profiles.
 
@@ -45,7 +47,7 @@ Example:
     	decryption_key "my SHA256 digest key"
     	decrypted_file_checksum "SHA256 hex digest of decrypted file"
     end
-	
+
 #MD5 and Multi-Part Upload
 s3_file compares the MD5 hash of a local file, if present, and the ETag header of the S3 object.  If they do not match, then the remote object will be downloaded and notifiations will be fired.
 

attributes/default.rb

@@ -1,2 +1,5 @@
 default['s3_file']['mime-types']['version'] = '2.6.2'
 default['s3_file']['rest-client']['version'] = '1.7.3'
+
+# Keep a catalog of each downloaded file's etag and md5 at time of download.
+default['s3_file']['use_catalog'] = true

libraries/s3_file.rb

@@ -293,6 +293,15 @@ def self.verify_sha256_checksum(checksum, file)
 
   def self.verify_md5_checksum(checksum, file)
     s3_md5 = checksum
+    local_md5 = buffered_md5_checksum(file)
+
+    Chef::Log.debug "md5 of remote object is #{s3_md5}"
+    Chef::Log.debug "md5 of local object is #{local_md5.hexdigest}"
+
+    local_md5.hexdigest == s3_md5
+  end
+
+  def self.buffered_md5_checksum(file)
     local_md5 = Digest::MD5.new
 
     # buffer the checksum which should save RAM consumption
@@ -301,11 +310,23 @@ def self.verify_md5_checksum(checksum, file)
         local_md5.update buffer
       end
     end
+    local_md5
+  end
 
-    Chef::Log.debug "md5 of remote object is #{s3_md5}"
-    Chef::Log.debug "md5 of local object is #{local_md5.hexdigest}"
+  def self.verify_etag(etag, file)
+    catalog.fetch(file, nil) == etag
+  end
 
-    local_md5.hexdigest == s3_md5
+  def self.catalog_path
+    File.join(Chef::Config[:file_cache_path], 's3_file_etags.json')
+  end
+
+  def self.catalog
+    File.exist?(catalog_path) ? JSON.parse(IO.read(catalog_path)) : {}
+  end
+
+  def self.write_catalog(data)
+    File.open(catalog_path, 'w', 0644) { |f| f.write(JSON.dump(data)) }
   end
 
   def self.client

providers/default.rb

@@ -34,11 +34,11 @@
   end
 
   if ::File.exists?(new_resource.path)
+    s3_etag = S3FileLib::get_md5_from_s3(new_resource.bucket, new_resource.s3_url, remote_path, aws_access_key_id, aws_secret_access_key, token)
+
     if decryption_key.nil?
       if new_resource.decrypted_file_checksum.nil?
-        s3_md5 = S3FileLib::get_md5_from_s3(new_resource.bucket, new_resource.s3_url, remote_path, aws_access_key_id, aws_secret_access_key, token)
-
-        if S3FileLib::verify_md5_checksum(s3_md5, new_resource.path)
+        if S3FileLib::verify_md5_checksum(s3_etag, new_resource.path)
           Chef::Log.debug 'Skipping download, md5sum of local file matches file in S3.'
           download = false
         end
@@ -59,6 +59,16 @@
         end
       end
     end
+
+    # Don't download if content and etag match prior download
+    if node['s3_file']['use_catalog']
+      catalog_data = S3FileLib::catalog.fetch(new_resource.path, nil)
+      existing_file_md5 = S3FileLib::buffered_md5_checksum(new_resource.path)
+      if catalog_data && existing_file_md5 == catalog_data['local_md5'] && s3_etag == catalog_data['etag']
+        Chef::Log.debug 'Skipping download, md5 of local file and etag matches prior download.'
+        download = false
+      end
+    end
   end
 
   if download
@@ -78,16 +88,34 @@
         raise e
       end
 
-      ::FileUtils.mv(decrypted_file.path, new_resource.path)
+      downloaded_file = decrypted_file
     else
-      ::FileUtils.mv(response.file.path, new_resource.path)
+      downloaded_file = response.file
+    end
+
+    # Write etag and md5 to catalog for future reference
+    if node['s3_file']['use_catalog']
+      catalog = S3FileLib::catalog
+      catalog[new_resource.path] = {
+        'etag' => response.headers[:etag].gsub('"',''),
+        'local_md5' => S3FileLib::buffered_md5_checksum(downloaded_file.path)
+      }
+      S3FileLib::write_catalog(catalog)
+    end
+
+    # Take ownership and permissions from existing object
+    if ::File.exist?(new_resource.path)
+      stat = ::File::Stat.new(new_resource.path)
+      ::FileUtils.chown(stat.uid, stat.gid, downloaded_file)
+      ::FileUtils.chmod(stat.mode, downloaded_file)
     end
+    ::FileUtils.mv(downloaded_file.path, new_resource.path)
   end
 
   f = file new_resource.path do
     action :create
-    owner new_resource.owner || ENV['user']
-    group new_resource.group || ENV['user']
+    owner new_resource.owner || ENV['USER']
+    group new_resource.group || ENV['USER']
     mode new_resource.mode || '0644'
   end