fingerprint 1.3.0 → 1.3.1

data/README.md

@@ -26,6 +26,7 @@ Todo
 * Supporting tools for signing fingerprints easily.
 * Support indexing specific files as well as whole directories (maybe?).
 * Support general filenames for `--archive`, e.g. along with `-n`, maybe support a file called `index.fingerprint` by default: improved visibility for end user.
+* Because fingerprint is currently IO bound in terms of performance, single-threaded checksumming is fine, but for SSD and other fast storage, it might be possible to improve speed somewhat by using a map-reduce style approach.
 
 License
 -------
@@ -48,4 +49,4 @@ FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
 AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
 LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
 OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
-THE SOFTWARE.
+THE SOFTWARE.

data/bin/fingerprint

@@ -23,6 +23,8 @@
 require 'optparse'
 require 'pathname'
 require 'fingerprint'
+require 'lockfile'
+require 'fileutils'
 
 OPTIONS = {
 	:root => "./",
@@ -30,9 +32,10 @@ OPTIONS = {
 	:output => $stdout,
 	:verbose => false,
 	:force => false,
-	:name => "._index.fingerprint",
+	:name => "index.fingerprint",
 	:extended => false,
 	:checksums => Fingerprint::DEFAULT_CHECKSUMS,
+	:lockfile => true
 }
 
 ARGV.options do |o|
@@ -80,6 +83,10 @@ ARGV.options do |o|
 	o.on("--progress", "Print percentage progress to standard error.") do
 		OPTIONS[:progress] = true
 	end
+	
+	o.on("--no-lockfile", "Don't use a lockfile to validate access to the fingerprint.") do
+		OPTIONS[:lockfile] = false
+	end
 
 	o.separator ""
 	
@@ -116,22 +123,61 @@ if OPTIONS[:checksums].size == 0
 	OPTIONS[:checksums] = ['MD5', 'SHA2.256']
 end
 
+# Run some block with the given lock, if requested.
+def with_lock
+	lockfile_path = nil
+	if OPTIONS[:lockfile] == true
+		lockfile_path = Pathname.new(OPTIONS[:root]) + (OPTIONS[:name] + ".lock")
+	elsif OPTION[:lockfile] != false
+		lockfile_path = OPTIONS[:lockfile]
+	end
+	
+	if lockfile_path
+		begin
+			Lockfile.new(lockfile_path, :retries => 0) do
+				yield
+			end
+		rescue Lockfile::MaxTriesLockError
+			$stderr.puts "Could not acquire lock #{lockfile_path}."
+			
+			return false
+		end
+	else
+		yield
+	end
+	
+	return true
+end
+
 case (OPTIONS[:mode])
 	when :analyze
-		output_file = Pathname.new(OPTIONS[:root]) + OPTIONS[:name]
-		
-		if output_file.exist? && !OPTIONS[:force]
-			$stderr.puts "Output file #{output_file} already exists. Aborting."
-			exit(2)
-		end
+		result = with_lock do
+			output_file = Pathname.new(OPTIONS[:root]) + OPTIONS[:name]
 		
-		options = OPTIONS.dup
-		options[:excludes] = [OPTIONS[:name]]
+			if output_file.exist? && !OPTIONS[:force]
+				$stderr.puts "Output file #{output_file} already exists. Aborting."
+				exit(2)
+			end
 		
-		File.open(output_file, "w") do |io|
-			options[:output] = io
+			options = OPTIONS.dup
+			options[:excludes] = [OPTIONS[:name]]
 			
-			Fingerprint::Scanner.scan_paths([OPTIONS[:root]], options)
+			finished = false
+			begin
+				File.open(output_file, "w") do |io|
+					options[:output] = io
+			
+					Fingerprint::Scanner.scan_paths([OPTIONS[:root]], options)
+				end
+				finished = true
+			ensure
+				FileUtils.rm(output_file) unless finished
+			end
+		end
+		
+		unless result
+			# Lockfile failure
+			exit(4)
 		end
 	when :verify
 		error_count = 0

data/lib/fingerprint/record.rb

@@ -32,7 +32,7 @@ module Fingerprint
 	}
 
 	class Record
-		def initialize(mode, path, metadata)
+		def initialize(mode, path, metadata = {})
 			@mode = mode
 			@path = path
 

data/lib/fingerprint/scanner.rb

@@ -46,7 +46,9 @@ module Fingerprint
 			@options = options
 
 			@digests = {}
-			
+
+			@progress = nil
+
 			unless @options[:checksums] and @options[:checksums].size > 0
 				@options[:checksums] = DEFAULT_CHECKSUMS
 			end
@@ -75,6 +77,8 @@ module Fingerprint
 
 		# This code won't handle multiple threads..
 		def digests_for(path)
+			total = 0
+
 			@digests.each do |key, digest|
 				digest.reset
 			end
@@ -82,6 +86,10 @@ module Fingerprint
 			File.open(path, "rb") do |file|
 				buf = ""
 				while file.read(1024 * 1024 * 10, buf)
+					total += buf.size
+					
+					@progress.call(total) if @progress
+					
 					@digests.each do |key, digest|
 						digest << buf
 					end
@@ -104,6 +112,7 @@ module Fingerprint
 			if type == :file
 				metadata['file.size'] = stat.size
 				digests = digests_for(path)
+				metadata.merge!(digests)
 			end
 
 			# Extended information
@@ -129,7 +138,9 @@ module Fingerprint
 		# Output a file and associated metadata.
 		def file_record_for(path)
 			metadata = metadata_for(:file, path)
-			metadata.merge!(digests_for(path))
+			
+			# Should this be here or in metadata_for?
+			# metadata.merge!(digests_for(path))
 			
 			Record.new(:file, path, metadata)
 		end
@@ -183,6 +194,10 @@ module Fingerprint
 				@roots.each do |root|
 					Dir.chdir(root) do
 						Find.find("./") do |path|
+							if @options[:progress]
+								$stderr.puts "# Scanning: #{path}"
+							end
+							
 							if File.directory?(path)
 								if excluded?(path)
 									Find.prune # Ignore this directory
@@ -199,11 +214,21 @@ module Fingerprint
 				end
 			end
 			
+			if @options[:progress]
+				@progress = lambda do |read_size|
+					$stderr.puts "# Progress: File #{processed_count} / #{total_count}; Byte #{processed_size + read_size} / #{total_size} = #{sprintf('%0.3f%', (processed_size + read_size).to_f / total_size.to_f * 100.0)} (#{read_size}, #{processed_size}, #{total_size})"
+				end
+			end
+			
 			@roots.each do |root|
 				Dir.chdir(root) do
 					recordset << header_for(root)
 					
 					Find.find("./") do |path|
+						if @options[:progress]
+							$stderr.puts "# Path: #{path}"
+						end
+						
 						if File.directory?(path)
 							if excluded?(path)
 								excluded_count += 1
@@ -221,10 +246,10 @@ module Fingerprint
 						else
 							# Skip anything that isn't a valid file (e.g. pipes, sockets, symlinks).
 							if valid_file?(path)
+								recordset << file_record_for(path)
+
 								processed_count += 1
 								processed_size += File.size(path)
-
-								recordset << file_record_for(path)
 							else
 								excluded_count += 1
 								
@@ -235,9 +260,7 @@ module Fingerprint
 						end
 						
 						# Print out a progress summary if requested
-						if @options[:progress]
-							$stderr.puts "# Progress: File #{processed_count} / #{total_count} = #{sprintf('%0.2f%', processed_count.to_f / total_count.to_f * 100.0)}; Byte #{processed_size} / #{total_size} = #{sprintf('%0.2f%', processed_size.to_f / total_size.to_f * 100.0)}"
-						end
+						@progress.call(0) if @progress
 					end
 				end
 			end

data/lib/fingerprint/version.rb

@@ -23,7 +23,7 @@ module Fingerprint
   module VERSION
     MAJOR = 1
     MINOR = 3
-    TINY  = 0
+    TINY  = 1
 
     STRING = [MAJOR, MINOR, TINY].join('.')
   end

metadata

@@ -1,13 +1,13 @@
 --- !ruby/object:Gem::Specification 
 name: fingerprint
 version: !ruby/object:Gem::Version 
-  hash: 27
+  hash: 25
   prerelease: 
   segments: 
   - 1
   - 3
-  - 0
-  version: 1.3.0
+  - 1
+  version: 1.3.1
 platform: ruby
 authors: 
 - Samuel Williams
@@ -15,9 +15,22 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2011-08-18 00:00:00 Z
-dependencies: []
-
+date: 2011-10-05 00:00:00 Z
+dependencies: 
+- !ruby/object:Gem::Dependency 
+  name: lockfile
+  prerelease: false
+  requirement: &id001 !ruby/object:Gem::Requirement 
+    none: false
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
+        hash: 3
+        segments: 
+        - 0
+        version: "0"
+  type: :runtime
+  version_requirements: *id001
 description: 
 email: samuel@oriontransfer.org
 executables: