Bumped version to 0.3.2

* Runtime: Store the actual archive on commit
* Registry: Improve the checksum process
- Registry: Fix error 400 on push

CHANGELOG.md

@@ -1,14 +1,55 @@
 # Changelog
 
-## 0.2.2 (2012-05-03)
+## 0.3.2 (2013-05-09)
+ * Runtime: Store the actual archive on commit
+ * Registry: Improve the checksum process
+ * Registry: Use the size to have a good progress bar while pushing
+ * Registry: Use the actual archive if it exists in order to speed up the push
+ - Registry: Fix error 400 on push
+
+## 0.3.1 (2013-05-08)
+ + Builder: Implement the autorun capability within docker builder
+ + Builder: Add caching to docker builder
+ + Builder: Add support for docker builder with native API as top level command
+ + Runtime: Add go version to debug infos
+ + Builder: Implement ENV within docker builder
+ + Registry: Add docker search top level command in order to search a repository
+ + Images: output graph of images to dot (graphviz)
+ + Documentation: new introduction and high-level overview
+ + Documentation: Add the documentation for docker builder
+ + Website: new high-level overview
+ - Makefile: Swap "go get" for "go get -d", especially to compile on go1.1rc
+ - Images: fix ByParent function
+ - Builder: Check the command existance prior create and add Unit tests for the case
+ - Registry: Fix pull for official images with specific tag
+ - Registry: Fix issue when login in with a different user and trying to push
+ - Documentation: CSS fix for docker documentation to make REST API docs look better.
+ - Documentation: Fixed CouchDB example page header mistake
+ - Documentation: fixed README formatting
+ * Registry: Improve checksum - async calculation
+ * Runtime: kernel version - don't show the dash if flavor is empty
+ * Documentation: updated www.docker.io website.
+ * Builder: use any whitespaces instead of tabs
+ * Packaging: packaging ubuntu; issue #510: Use goland-stable PPA package to build docker
+
+## 0.3.0 (2013-05-06)
+ + Registry: Implement the new registry
+ + Documentation: new example: sharing data between 2 couchdb databases
+ - Runtime: Fix the command existance check
+ - Runtime: strings.Split may return an empty string on no match
+ - Runtime: Fix an index out of range crash if cgroup memory is not
+ * Documentation: Various improvments
+ * Vagrant: Use only one deb line in /etc/apt
+
+## 0.2.2 (2013-05-03)
  + Support for data volumes ('docker run -v=PATH')
  + Share data volumes between containers ('docker run -volumes-from')
  + Improved documentation
  * Upgrade to Go 1.0.3
  * Various upgrades to the dev environment for contributors
 
-## 0.2.1 (2012-05-01)
- + 'docker commit -run' bundles a layer with default runtime options: command, ports etc. 
+## 0.2.1 (2013-05-01)
+ + 'docker commit -run' bundles a layer with default runtime options: command, ports etc.
  * Improve install process on Vagrant
  + New Dockerfile operation: "maintainer"
  + New Dockerfile operation: "expose"
@@ -17,7 +58,7 @@
  + 'docker -d -r': restart crashed containers at daemon startup
  * Runtime: improve test coverage
 
-## 0.2.0 (2012-04-23)
+## 0.2.0 (2013-04-23)
  - Runtime: ghost containers can be killed and waited for
  * Documentation: update install intructions
  - Packaging: fix Vagrantfile
@@ -25,13 +66,12 @@
  + Add a changelog
  - Various bugfixes
 
-
 ## 0.1.8 (2013-04-22)
  - Dynamically detect cgroup capabilities
  - Issue stability warning on kernels <3.8
  - 'docker push' buffers on disk instead of memory
  - Fix 'docker diff' for removed files
- - Fix 'docker stop' for ghost containers 
+ - Fix 'docker stop' for ghost containers
  - Fix handling of pidfile
  - Various bugfixes and stability improvements
 
@@ -52,7 +92,7 @@
  - Improve diagnosis of missing system capabilities
  - Allow disabling memory limits at compile time
  - Add debian packaging
- - Documentation: installing on Arch Linux 
+ - Documentation: installing on Arch Linux
  - Documentation: running Redis on docker
  - Fixed lxc 0.9 compatibility
  - Automatically load aufs module

Makefile

@@ -39,7 +39,7 @@ $(DOCKER_BIN): $(DOCKER_DIR)
 $(DOCKER_DIR):
 	@mkdir -p $(dir $@)
 	@if [ -h $@ ]; then rm -f $@; fi; ln -sf $(CURDIR)/ $@
-	@(cd $(DOCKER_MAIN); go get $(GO_OPTIONS))
+	@(cd $(DOCKER_MAIN); go get -d $(GO_OPTIONS))
 
 whichrelease:
 	echo $(RELEASE_VERSION)

README.md

@@ -1,37 +1,94 @@
-Docker: the Linux container runtime
-===================================
+Docker: the Linux container engine
+==================================
 
-Docker complements LXC with a high-level API which operates at the process level. It runs unix processes with strong guarantees of isolation and repeatability across servers.
+Docker is an open-source engine which automates the deployment of applications as highly portable, self-sufficient containers.
 
-Docker is a great building block for automating distributed systems: large-scale web deployments, database clusters, continuous deployment systems, private PaaS, service-oriented architectures, etc.
+Docker containers are both *hardware-agnostic* and *platform-agnostic*. This means that they can run anywhere, from your
+laptop to the largest EC2 compute instance and everything in between - and they don't require that you use a particular
+language, framework or packaging system. That makes them great building blocks for deploying and scaling web apps, databases
+and backend services without depending on a particular stack or provider.
+
+Docker is an open-source implementation of the deployment engine which powers [dotCloud](http://dotcloud.com), a popular Platform-as-a-Service.
+It benefits directly from the experience accumulated over several years of large-scale operation and support of hundreds of thousands
+of applications and databases.
 
 ![Docker L](docs/sources/static_files/lego_docker.jpg "Docker")
 
-* *Heterogeneous payloads*: any combination of binaries, libraries, configuration files, scripts, virtualenvs, jars, gems, tarballs, you name it. No more juggling between domain-specific tools. Docker can deploy and run them all.
+## Better than VMs
 
-* *Any server*: docker can run on any x64 machine with a modern linux kernel - whether it's a laptop, a bare metal server or a VM. This makes it perfect for multi-cloud deployments.
+A common method for distributing applications and sandbox their execution is to use virtual machines, or VMs. Typical VM formats
+are VMWare's vmdk, Oracle Virtualbox's vdi, and Amazon EC2's ami. In theory these formats should allow every developer to
+automatically package their application into a "machine" for easy distribution and deployment. In practice, that almost never
+happens, for a few reasons:
 
-* *Isolation*: docker isolates processes from each other and from the underlying host, using lightweight containers.
+  * *Size*: VMs are very large which makes them impractical to store and transfer.
+  * *Performance*: running VMs consumes significant CPU and memory, which makes them impractical in many scenarios, for example local development of multi-tier applications, and
+  	large-scale deployment of cpu and memory-intensive applications on large numbers of machines.
+  * *Portability*: competing VM environments don't play well with each other. Although conversion tools do exist, they are limited and add even more overhead.
+  * *Hardware-centric*: VMs were designed with machine operators in mind, not software developers. As a result, they offer very limited tooling for what developers need most:
+  	building, testing and running their software. For example, VMs offer no facilities for application versioning, monitoring, configuration, logging or service discovery.
 
-* *Repeatability*: because containers are isolated in their own filesystem, they behave the same regardless of where, when, and alongside what they run.
+By contrast, Docker relies on a different sandboxing method known as *containerization*. Unlike traditional virtualization,
+containerization takes place at the kernel level. Most modern operating system kernels now support the primitives necessary
+for containerization, including Linux with [openvz](http://openvz.org), [vserver](http://linux-vserver.org) and more recently [lxc](http://lxc.sourceforge.net),
+	Solaris with [zones](http://docs.oracle.com/cd/E26502_01/html/E29024/preface-1.html#scrolltoc) and FreeBSD with [Jails](http://www.freebsd.org/doc/handbook/jails.html).
+
+Docker builds on top of these low-level primitives to offer developers a portable format and runtime environment that solves
+all 4 problems. Docker containers are small (and their transfer can be optimized with layers), they have basically zero memory and cpu overhead,
+the are completely portable and are designed from the ground up with an application-centric design.
+
+The best part: because docker operates at the OS level, it can still be run inside a VM!
+
+## Plays well with others
+
+Docker does not require that you buy into a particular programming language, framework, packaging system or configuration language.
+
+Is your application a unix process? Does it use files, tcp connections, environment variables, standard unix streams and command-line
+arguments as inputs and outputs? Then docker can run it.
+
+Can your application's build be expressed a sequence of such commands? Then docker can build it.
 
 
-Notable features
------------------
+## Escape dependency hell
 
-* Filesystem isolation: each process container runs in a completely separate root filesystem.
+A common problem for developers is the difficulty of managing all their application's dependencies in a simple and automated way.
 
-* Resource isolation: system resources like cpu and memory can be allocated differently to each process container, using cgroups.
+This is usually difficult for several reasons:
 
-* Network isolation: each process container runs in its own network namespace, with a virtual interface and IP address of its own.
+  * *Cross-platform dependencies*. Modern applications often depend on a combination of system libraries and binaries, language-specific packages, framework-specific modules,
+  	internal components developed for another project, etc. These dependencies live in different "worlds" and require different tools - these tools typically don't work
+	well with each other, requiring awkward custom integrations.
 
-* Copy-on-write: root filesystems are created using copy-on-write, which makes deployment extremely fast, memory-cheap and disk-cheap.
+  * Conflicting dependencies. Different applications may depend on different versions of the same dependency. Packaging tools handle these situations with various degrees of ease -
+  	but they all handle them in different and incompatible ways, which again forces the developer to do extra work.
+  
+  * Custom dependencies. A developer may need to prepare a custom version of his application's dependency. Some packaging systems can handle custom versions of a dependency,
+  	others can't - and all of them handle it differently.
 
-* Logging: the standard streams (stdout/stderr/stdin) of each process container are collected and logged for real-time or batch retrieval.
 
-* Change management: changes to a container's filesystem can be committed into a new image and re-used to create more containers. No templating or manual configuration required.
+Docker solves dependency hell by giving the developer a simple way to express *all* his application's dependencies in one place,
+and streamline the process of assembling them. If this makes you think of [XKCD 927](http://xkcd.com/927/), don't worry. Docker doesn't
+*replace* your favorite packaging systems. It simply orchestrates their use in a simple and repeatable way. How does it do that? With layers.
+
+Docker defines a build as running a sequence unix commands, one after the other, in the same container. Build commands modify the contents of the container
+(usually by installing new files on the filesystem), the next command modifies it some more, etc. Since each build command inherits the result of the previous
+commands, the *order* in which the commands are executed expresses *dependencies*.
+
+Here's a typical docker build process:
+
+```bash
+from	ubuntu:12.10
+run	apt-get update
+run	apt-get install python
+run	apt-get install python-pip
+run	pip install django
+run	apt-get install curl
+run	curl http://github.com/shykes/helloflask/helloflask/master.tar.gz | tar -zxv
+run	cd master && pip install -r requirements.txt
+```
+
+Note that Docker doesn't care *how* dependencies are built - as long as they can be built by running a unix command in a container.
 
-* Interactive shell: docker can allocate a pseudo-tty and attach to the standard input of any container, for example to run a throwaway interactive shell.
 
 Install instructions
 ==================

Vagrantfile

@@ -1,8 +1,8 @@
 # -*- mode: ruby -*-
 # vi: set ft=ruby :
 
-BOX_NAME = "ubuntu"
-BOX_URI = "http://files.vagrantup.com/precise64.box"
+BOX_NAME = ENV['BOX_NAME'] || "ubuntu"
+BOX_URI = ENV['BOX_URI'] || "http://files.vagrantup.com/precise64.box"
 PPA_KEY = "E61D797F63561DC6"
 
 Vagrant::Config.run do |config|

auth/auth.go

@@ -3,7 +3,6 @@ package auth
 import (
 	"encoding/base64"
 	"encoding/json"
-	"errors"
 	"fmt"
 	"io/ioutil"
 	"net/http"
@@ -16,7 +15,7 @@ import (
 const CONFIGFILE = ".dockercfg"
 
 // the registry server we want to login against
-const REGISTRY_SERVER = "https://registry.docker.io"
+const INDEX_SERVER = "https://index.docker.io"
 
 type AuthConfig struct {
 	Username string `json:"username"`
@@ -76,6 +75,9 @@ func LoadConfig(rootPath string) (*AuthConfig, error) {
 		return nil, err
 	}
 	arr := strings.Split(string(b), "\n")
+	if len(arr) < 2 {
+		return nil, fmt.Errorf("The Auth config file is empty")
+	}
 	origAuth := strings.Split(arr[0], " = ")
 	origEmail := strings.Split(arr[1], " = ")
 	authConfig, err := DecodeAuth(origAuth[1])
@@ -89,9 +91,14 @@ func LoadConfig(rootPath string) (*AuthConfig, error) {
 
 // save the auth config
 func saveConfig(rootPath, authStr string, email string) error {
+	confFile := path.Join(rootPath, CONFIGFILE)
+	if len(email) == 0 {
+		os.Remove(confFile)
+		return nil
+	}
 	lines := "auth = " + authStr + "\n" + "email = " + email + "\n"
 	b := []byte(lines)
-	err := ioutil.WriteFile(path.Join(rootPath, CONFIGFILE), b, 0600)
+	err := ioutil.WriteFile(confFile, b, 0600)
 	if err != nil {
 		return err
 	}
@@ -101,40 +108,38 @@ func saveConfig(rootPath, authStr string, email string) error {
 // try to register/login to the registry server
 func Login(authConfig *AuthConfig) (string, error) {
 	storeConfig := false
+	client := &http.Client{}
 	reqStatusCode := 0
 	var status string
-	var errMsg string
 	var reqBody []byte
 	jsonBody, err := json.Marshal(authConfig)
 	if err != nil {
-		errMsg = fmt.Sprintf("Config Error: %s", err)
-		return "", errors.New(errMsg)
+		return "", fmt.Errorf("Config Error: %s", err)
 	}
 
 	// using `bytes.NewReader(jsonBody)` here causes the server to respond with a 411 status.
 	b := strings.NewReader(string(jsonBody))
-	req1, err := http.Post(REGISTRY_SERVER+"/v1/users", "application/json; charset=utf-8", b)
+	req1, err := http.Post(INDEX_SERVER+"/v1/users/", "application/json; charset=utf-8", b)
 	if err != nil {
-		errMsg = fmt.Sprintf("Server Error: %s", err)
-		return "", errors.New(errMsg)
+		return "", fmt.Errorf("Server Error: %s", err)
 	}
-
 	reqStatusCode = req1.StatusCode
 	defer req1.Body.Close()
 	reqBody, err = ioutil.ReadAll(req1.Body)
 	if err != nil {
-		errMsg = fmt.Sprintf("Server Error: [%#v] %s", reqStatusCode, err)
-		return "", errors.New(errMsg)
+		return "", fmt.Errorf("Server Error: [%#v] %s", reqStatusCode, err)
 	}
 
 	if reqStatusCode == 201 {
-		status = "Account Created\n"
+		status = "Account created. Please use the confirmation link we sent" +
+			" to your e-mail to activate it.\n"
 		storeConfig = true
+	} else if reqStatusCode == 403 {
+		return "", fmt.Errorf("Login: Your account hasn't been activated. " +
+			"Please check your e-mail for a confirmation link.")
 	} else if reqStatusCode == 400 {
-		// FIXME: This should be 'exists', not 'exist'. Need to change on the server first.
-		if string(reqBody) == "Username or email already exist" {
-			client := &http.Client{}
-			req, err := http.NewRequest("GET", REGISTRY_SERVER+"/v1/users", nil)
+		if string(reqBody) == "\"Username or email already exists\"" {
+			req, err := http.NewRequest("GET", INDEX_SERVER+"/v1/users/", nil)
 			req.SetBasicAuth(authConfig.Username, authConfig.Password)
 			resp, err := client.Do(req)
 			if err != nil {
@@ -148,17 +153,18 @@ func Login(authConfig *AuthConfig) (string, error) {
 			if resp.StatusCode == 200 {
 				status = "Login Succeeded\n"
 				storeConfig = true
+			} else if resp.StatusCode == 401 {
+				saveConfig(authConfig.rootPath, "", "")
+				return "", fmt.Errorf("Wrong login/password, please try again")
 			} else {
-				status = fmt.Sprintf("Login: %s", body)
-				return "", errors.New(status)
+				return "", fmt.Errorf("Login: %s (Code: %d; Headers: %s)", body,
+					resp.StatusCode, resp.Header)
 			}
 		} else {
-			status = fmt.Sprintf("Registration: %s", reqBody)
-			return "", errors.New(status)
+			return "", fmt.Errorf("Registration: %s", reqBody)
 		}
 	} else {
-		status = fmt.Sprintf("[%s] : %s", reqStatusCode, reqBody)
-		return "", errors.New(status)
+		return "", fmt.Errorf("Unexpected status code [%d] : %s", reqStatusCode, reqBody)
 	}
 	if storeConfig {
 		authStr := EncodeAuth(authConfig)

builder.go

@@ -0,0 +1,463 @@
+package docker
+
+import (
+	"bufio"
+	"encoding/json"
+	"fmt"
+	"io"
+	"os"
+	"path"
+	"strings"
+	"time"
+)
+
+type Builder struct {
+	runtime      *Runtime
+	repositories *TagStore
+	graph        *Graph
+}
+
+func NewBuilder(runtime *Runtime) *Builder {
+	return &Builder{
+		runtime:      runtime,
+		graph:        runtime.graph,
+		repositories: runtime.repositories,
+	}
+}
+
+func (builder *Builder) mergeConfig(userConf, imageConf *Config) {
+	if userConf.Hostname != "" {
+		userConf.Hostname = imageConf.Hostname
+	}
+	if userConf.User != "" {
+		userConf.User = imageConf.User
+	}
+	if userConf.Memory == 0 {
+		userConf.Memory = imageConf.Memory
+	}
+	if userConf.MemorySwap == 0 {
+		userConf.MemorySwap = imageConf.MemorySwap
+	}
+	if userConf.PortSpecs == nil || len(userConf.PortSpecs) == 0 {
+		userConf.PortSpecs = imageConf.PortSpecs
+	}
+	if !userConf.Tty {
+		userConf.Tty = userConf.Tty
+	}
+	if !userConf.OpenStdin {
+		userConf.OpenStdin = imageConf.OpenStdin
+	}
+	if !userConf.StdinOnce {
+		userConf.StdinOnce = imageConf.StdinOnce
+	}
+	if userConf.Env == nil || len(userConf.Env) == 0 {
+		userConf.Env = imageConf.Env
+	}
+	if userConf.Cmd == nil || len(userConf.Cmd) == 0 {
+		userConf.Cmd = imageConf.Cmd
+	}
+	if userConf.Dns == nil || len(userConf.Dns) == 0 {
+		userConf.Dns = imageConf.Dns
+	}
+}
+
+func (builder *Builder) Create(config *Config) (*Container, error) {
+	// Lookup image
+	img, err := builder.repositories.LookupImage(config.Image)
+	if err != nil {
+		return nil, err
+	}
+
+	if img.Config != nil {
+		builder.mergeConfig(config, img.Config)
+	}
+
+	if config.Cmd == nil || len(config.Cmd) == 0 {
+		return nil, fmt.Errorf("No command specified")
+	}
+
+	// Generate id
+	id := GenerateId()
+	// Generate default hostname
+	// FIXME: the lxc template no longer needs to set a default hostname
+	if config.Hostname == "" {
+		config.Hostname = id[:12]
+	}
+
+	container := &Container{
+		// FIXME: we should generate the ID here instead of receiving it as an argument
+		Id:              id,
+		Created:         time.Now(),
+		Path:            config.Cmd[0],
+		Args:            config.Cmd[1:], //FIXME: de-duplicate from config
+		Config:          config,
+		Image:           img.Id, // Always use the resolved image id
+		NetworkSettings: &NetworkSettings{},
+		// FIXME: do we need to store this in the container?
+		SysInitPath: sysInitPath,
+	}
+	container.root = builder.runtime.containerRoot(container.Id)
+	// Step 1: create the container directory.
+	// This doubles as a barrier to avoid race conditions.
+	if err := os.Mkdir(container.root, 0700); err != nil {
+		return nil, err
+	}
+
+	// If custom dns exists, then create a resolv.conf for the container
+	if len(config.Dns) > 0 {
+		container.ResolvConfPath = path.Join(container.root, "resolv.conf")
+		f, err := os.Create(container.ResolvConfPath)
+		if err != nil {
+			return nil, err
+		}
+		defer f.Close()
+		for _, dns := range config.Dns {
+			if _, err := f.Write([]byte("nameserver " + dns + "\n")); err != nil {
+				return nil, err
+			}
+		}
+	} else {
+		container.ResolvConfPath = "/etc/resolv.conf"
+	}
+
+	// Step 2: save the container json
+	if err := container.ToDisk(); err != nil {
+		return nil, err
+	}
+	// Step 3: register the container
+	if err := builder.runtime.Register(container); err != nil {
+		return nil, err
+	}
+	return container, nil
+}
+
+// Commit creates a new filesystem image from the current state of a container.
+// The image can optionally be tagged into a repository
+func (builder *Builder) Commit(container *Container, repository, tag, comment, author string, config *Config) (*Image, error) {
+	// FIXME: freeze the container before copying it to avoid data corruption?
+	// FIXME: this shouldn't be in commands.
+	rwTar, err := container.ExportRw()
+	if err != nil {
+		return nil, err
+	}
+	// Create a new image from the container's base layers + a new layer from container changes
+	img, err := builder.graph.Create(rwTar, container, comment, author, config)
+	if err != nil {
+		return nil, err
+	}
+	// Register the image if needed
+	if repository != "" {
+		if err := builder.repositories.Set(repository, tag, img.Id, true); err != nil {
+			return img, err
+		}
+	}
+	return img, nil
+}
+
+func (builder *Builder) clearTmp(containers, images map[string]struct{}) {
+	for c := range containers {
+		tmp := builder.runtime.Get(c)
+		builder.runtime.Destroy(tmp)
+		Debugf("Removing container %s", c)
+	}
+	for i := range images {
+		builder.runtime.graph.Delete(i)
+		Debugf("Removing image %s", i)
+	}
+}
+
+func (builder *Builder) getCachedImage(image *Image, config *Config) (*Image, error) {
+	// Retrieve all images
+	images, err := builder.graph.All()
+	if err != nil {
+		return nil, err
+	}
+
+	// Store the tree in a map of map (map[parentId][childId])
+	imageMap := make(map[string]map[string]struct{})
+	for _, img := range images {
+		if _, exists := imageMap[img.Parent]; !exists {
+			imageMap[img.Parent] = make(map[string]struct{})
+		}
+		imageMap[img.Parent][img.Id] = struct{}{}
+	}
+
+	// Loop on the children of the given image and check the config
+	for elem := range imageMap[image.Id] {
+		img, err := builder.graph.Get(elem)
+		if err != nil {
+			return nil, err
+		}
+		if CompareConfig(&img.ContainerConfig, config) {
+			return img, nil
+		}
+	}
+	return nil, nil
+}
+
+func (builder *Builder) Build(dockerfile io.Reader, stdout io.Writer) (*Image, error) {
+	var (
+		image, base   *Image
+		config        *Config
+		maintainer    string
+		env           map[string]string   = make(map[string]string)
+		tmpContainers map[string]struct{} = make(map[string]struct{})
+		tmpImages     map[string]struct{} = make(map[string]struct{})
+	)
+	defer builder.clearTmp(tmpContainers, tmpImages)
+
+	file := bufio.NewReader(dockerfile)
+	for {
+		line, err := file.ReadString('\n')
+		if err != nil {
+			if err == io.EOF {
+				break
+			}
+			return nil, err
+		}
+		line = strings.Replace(strings.TrimSpace(line), "	", " ", 1)
+		// Skip comments and empty line
+		if len(line) == 0 || line[0] == '#' {
+			continue
+		}
+		tmp := strings.SplitN(line, " ", 2)
+		if len(tmp) != 2 {
+			return nil, fmt.Errorf("Invalid Dockerfile format")
+		}
+		instruction := strings.Trim(tmp[0], " ")
+		arguments := strings.Trim(tmp[1], " ")
+		switch strings.ToLower(instruction) {
+		case "from":
+			fmt.Fprintf(stdout, "FROM %s\n", arguments)
+			image, err = builder.runtime.repositories.LookupImage(arguments)
+			if err != nil {
+				if builder.runtime.graph.IsNotExist(err) {
+
+					var tag, remote string
+					if strings.Contains(arguments, ":") {
+						remoteParts := strings.Split(arguments, ":")
+						tag = remoteParts[1]
+						remote = remoteParts[0]
+					} else {
+						remote = arguments
+					}
+
+					if err := builder.runtime.graph.PullRepository(stdout, remote, tag, builder.runtime.repositories, builder.runtime.authConfig); err != nil {
+						return nil, err
+					}
+
+					image, err = builder.runtime.repositories.LookupImage(arguments)
+					if err != nil {
+						return nil, err
+					}
+				} else {
+					return nil, err
+				}
+			}
+			config = &Config{}
+
+			break
+		case "maintainer":
+			fmt.Fprintf(stdout, "MAINTAINER %s\n", arguments)
+			maintainer = arguments
+			break
+		case "run":
+			fmt.Fprintf(stdout, "RUN %s\n", arguments)
+			if image == nil {
+				return nil, fmt.Errorf("Please provide a source image with `from` prior to run")
+			}
+			config, err := ParseRun([]string{image.Id, "/bin/sh", "-c", arguments}, nil, builder.runtime.capabilities)
+			if err != nil {
+				return nil, err
+			}
+
+			for key, value := range env {
+				config.Env = append(config.Env, fmt.Sprintf("%s=%s", key, value))
+			}
+
+			if cache, err := builder.getCachedImage(image, config); err != nil {
+				return nil, err
+			} else if cache != nil {
+				image = cache
+				fmt.Fprintf(stdout, "===> %s\n", image.ShortId())
+				break
+			}
+
+			Debugf("Env -----> %v ------ %v\n", config.Env, env)
+
+			// Create the container and start it
+			c, err := builder.Create(config)
+			if err != nil {
+				return nil, err
+			}
+
+			if os.Getenv("DEBUG") != "" {
+				out, _ := c.StdoutPipe()
+				err2, _ := c.StderrPipe()
+				go io.Copy(os.Stdout, out)
+				go io.Copy(os.Stdout, err2)
+			}
+
+			if err := c.Start(); err != nil {
+				return nil, err
+			}
+			tmpContainers[c.Id] = struct{}{}
+
+			// Wait for it to finish
+			if result := c.Wait(); result != 0 {
+				return nil, fmt.Errorf("!!! '%s' return non-zero exit code '%d'. Aborting.", arguments, result)
+			}
+
+			// Commit the container
+			base, err = builder.Commit(c, "", "", "", maintainer, nil)
+			if err != nil {
+				return nil, err
+			}
+			tmpImages[base.Id] = struct{}{}
+
+			fmt.Fprintf(stdout, "===> %s\n", base.ShortId())
+
+			// use the base as the new image
+			image = base
+
+			break
+		case "env":
+			tmp := strings.SplitN(arguments, " ", 2)
+			if len(tmp) != 2 {
+				return nil, fmt.Errorf("Invalid ENV format")
+			}
+			key := strings.Trim(tmp[0], " ")
+			value := strings.Trim(tmp[1], " ")
+			fmt.Fprintf(stdout, "ENV %s %s\n", key, value)
+			env[key] = value
+			if image != nil {
+				fmt.Fprintf(stdout, "===> %s\n", image.ShortId())
+			} else {
+				fmt.Fprintf(stdout, "===> <nil>\n")
+			}
+			break
+		case "cmd":
+			fmt.Fprintf(stdout, "CMD %s\n", arguments)
+
+			// Create the container and start it
+			c, err := builder.Create(&Config{Image: image.Id, Cmd: []string{"", ""}})
+			if err != nil {
+				return nil, err
+			}
+			if err := c.Start(); err != nil {
+				return nil, err
+			}
+			tmpContainers[c.Id] = struct{}{}
+
+			cmd := []string{}
+			if err := json.Unmarshal([]byte(arguments), &cmd); err != nil {
+				return nil, err
+			}
+			config.Cmd = cmd
+
+			// Commit the container
+			base, err = builder.Commit(c, "", "", "", maintainer, config)
+			if err != nil {
+				return nil, err
+			}
+			tmpImages[base.Id] = struct{}{}
+
+			fmt.Fprintf(stdout, "===> %s\n", base.ShortId())
+			image = base
+			break
+		case "expose":
+			ports := strings.Split(arguments, " ")
+
+			fmt.Fprintf(stdout, "EXPOSE %v\n", ports)
+			if image == nil {
+				return nil, fmt.Errorf("Please provide a source image with `from` prior to copy")
+			}
+
+			// Create the container and start it
+			c, err := builder.Create(&Config{Image: image.Id, Cmd: []string{"", ""}})
+			if err != nil {
+				return nil, err
+			}
+			if err := c.Start(); err != nil {
+				return nil, err
+			}
+			tmpContainers[c.Id] = struct{}{}
+
+			config.PortSpecs = append(ports, config.PortSpecs...)
+
+			// Commit the container
+			base, err = builder.Commit(c, "", "", "", maintainer, config)
+			if err != nil {
+				return nil, err
+			}
+			tmpImages[base.Id] = struct{}{}
+
+			fmt.Fprintf(stdout, "===> %s\n", base.ShortId())
+			image = base
+			break
+		case "insert":
+			if image == nil {
+				return nil, fmt.Errorf("Please provide a source image with `from` prior to copy")
+			}
+			tmp = strings.SplitN(arguments, " ", 2)
+			if len(tmp) != 2 {
+				return nil, fmt.Errorf("Invalid INSERT format")
+			}
+			sourceUrl := strings.Trim(tmp[0], " ")
+			destPath := strings.Trim(tmp[1], " ")
+			fmt.Fprintf(stdout, "COPY %s to %s in %s\n", sourceUrl, destPath, base.ShortId())
+
+			file, err := Download(sourceUrl, stdout)
+			if err != nil {
+				return nil, err
+			}
+			defer file.Body.Close()
+
+			config, err := ParseRun([]string{base.Id, "echo", "insert", sourceUrl, destPath}, nil, builder.runtime.capabilities)
+			if err != nil {
+				return nil, err
+			}
+			c, err := builder.Create(config)
+			if err != nil {
+				return nil, err
+			}
+
+			if err := c.Start(); err != nil {
+				return nil, err
+			}
+
+			// Wait for echo to finish
+			if result := c.Wait(); result != 0 {
+				return nil, fmt.Errorf("!!! '%s' return non-zero exit code '%d'. Aborting.", arguments, result)
+			}
+
+			if err := c.Inject(file.Body, destPath); err != nil {
+				return nil, err
+			}
+
+			base, err = builder.Commit(c, "", "", "", maintainer, nil)
+			if err != nil {
+				return nil, err
+			}
+			fmt.Fprintf(stdout, "===> %s\n", base.ShortId())
+
+			image = base
+
+			break
+		default:
+			fmt.Fprintf(stdout, "Skipping unknown instruction %s\n", strings.ToUpper(instruction))
+		}
+	}
+	if image != nil {
+		// The build is successful, keep the temporary containers and images
+		for i := range tmpImages {
+			delete(tmpImages, i)
+		}
+		for i := range tmpContainers {
+			delete(tmpContainers, i)
+		}
+		fmt.Fprintf(stdout, "Build finished. image id: %s\n", image.ShortId())
+		return image, nil
+	}
+	return nil, fmt.Errorf("An error occured during the build\n")
+}

builder_test.go

@@ -0,0 +1,88 @@
+package docker
+
+import (
+	"strings"
+	"testing"
+)
+
+const Dockerfile = `
+# VERSION		0.1
+# DOCKER-VERSION	0.2
+
+from   ` + unitTestImageName + `
+run    sh -c 'echo root:testpass > /tmp/passwd'
+run    mkdir -p /var/run/sshd
+insert https://raw.github.com/dotcloud/docker/master/CHANGELOG.md /tmp/CHANGELOG.md
+`
+
+func TestBuild(t *testing.T) {
+	runtime, err := newTestRuntime()
+	if err != nil {
+		t.Fatal(err)
+	}
+	defer nuke(runtime)
+
+	builder := NewBuilder(runtime)
+
+	img, err := builder.Build(strings.NewReader(Dockerfile), &nopWriter{})
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	container, err := builder.Create(
+		&Config{
+			Image: img.Id,
+			Cmd:   []string{"cat", "/tmp/passwd"},
+		},
+	)
+	if err != nil {
+		t.Fatal(err)
+	}
+	defer runtime.Destroy(container)
+
+	output, err := container.Output()
+	if err != nil {
+		t.Fatal(err)
+	}
+	if string(output) != "root:testpass\n" {
+		t.Fatalf("Unexpected output. Read '%s', expected '%s'", output, "root:testpass\n")
+	}
+
+	container2, err := builder.Create(
+		&Config{
+			Image: img.Id,
+			Cmd:   []string{"ls", "-d", "/var/run/sshd"},
+		},
+	)
+	if err != nil {
+		t.Fatal(err)
+	}
+	defer runtime.Destroy(container2)
+
+	output, err = container2.Output()
+	if err != nil {
+		t.Fatal(err)
+	}
+	if string(output) != "/var/run/sshd\n" {
+		t.Fatal("/var/run/sshd has not been created")
+	}
+
+	container3, err := builder.Create(
+		&Config{
+			Image: img.Id,
+			Cmd:   []string{"cat", "/tmp/CHANGELOG.md"},
+		},
+	)
+	if err != nil {
+		t.Fatal(err)
+	}
+	defer runtime.Destroy(container3)
+
+	output, err = container3.Output()
+	if err != nil {
+		t.Fatal(err)
+	}
+	if len(output) == 0 {
+		t.Fatal("/tmp/CHANGELOG.md has not been copied")
+	}
+}

commands.go

@@ -6,10 +6,12 @@ import (
 	"fmt"
 	"github.com/dotcloud/docker/auth"
 	"github.com/dotcloud/docker/rcli"
+	"github.com/shin-/cookiejar"
 	"io"
 	"log"
 	"net/http"
 	"net/url"
+	"os"
 	"path/filepath"
 	"runtime"
 	"strconv"
@@ -19,7 +21,7 @@ import (
 	"unicode"
 )
 
-const VERSION = "0.2.2"
+const VERSION = "0.3.2"
 
 var (
 	GIT_COMMIT string
@@ -34,6 +36,7 @@ func (srv *Server) Help() string {
 	help := "Usage: docker COMMAND [arg...]\n\nA self-sufficient runtime for linux containers.\n\nCommands:\n"
 	for _, cmd := range [][]string{
 		{"attach", "Attach to a running container"},
+		{"build", "Build a container from Dockerfile via stdin"},
 		{"commit", "Create a new image from a container's changes"},
 		{"diff", "Inspect changes on a container's filesystem"},
 		{"export", "Stream the contents of a container as a tar archive"},
@@ -41,6 +44,7 @@ func (srv *Server) Help() string {
 		{"images", "List images"},
 		{"import", "Create a new filesystem image from the contents of a tarball"},
 		{"info", "Display system-wide information"},
+		{"insert", "Insert a file in an image"},
 		{"inspect", "Return low-level information on a container"},
 		{"kill", "Kill a running container"},
 		{"login", "Register or Login to the docker registry server"},
@@ -53,6 +57,7 @@ func (srv *Server) Help() string {
 		{"rm", "Remove a container"},
 		{"rmi", "Remove an image"},
 		{"run", "Run a command in a new container"},
+		{"search", "Search for an image in the docker index"},
 		{"start", "Start a stopped container"},
 		{"stop", "Stop a running container"},
 		{"tag", "Tag an image into a repository"},
@@ -64,6 +69,67 @@ func (srv *Server) Help() string {
 	return help
 }
 
+func (srv *Server) CmdInsert(stdin io.ReadCloser, stdout rcli.DockerConn, args ...string) error {
+	stdout.Flush()
+	cmd := rcli.Subcmd(stdout, "insert", "IMAGE URL PATH", "Insert a file from URL in the IMAGE at PATH")
+	if err := cmd.Parse(args); err != nil {
+		return nil
+	}
+	if cmd.NArg() != 3 {
+		cmd.Usage()
+		return nil
+	}
+	imageId := cmd.Arg(0)
+	url := cmd.Arg(1)
+	path := cmd.Arg(2)
+
+	img, err := srv.runtime.repositories.LookupImage(imageId)
+	if err != nil {
+		return err
+	}
+	file, err := Download(url, stdout)
+	if err != nil {
+		return err
+	}
+	defer file.Body.Close()
+
+	config, err := ParseRun([]string{img.Id, "echo", "insert", url, path}, nil, srv.runtime.capabilities)
+	if err != nil {
+		return err
+	}
+
+	b := NewBuilder(srv.runtime)
+	c, err := b.Create(config)
+	if err != nil {
+		return err
+	}
+
+	if err := c.Inject(ProgressReader(file.Body, int(file.ContentLength), stdout, "Downloading %v/%v (%v)"), path); err != nil {
+		return err
+	}
+	// FIXME: Handle custom repo, tag comment, author
+	img, err = b.Commit(c, "", "", img.Comment, img.Author, nil)
+	if err != nil {
+		return err
+	}
+	fmt.Fprintf(stdout, "%s\n", img.Id)
+	return nil
+}
+
+func (srv *Server) CmdBuild(stdin io.ReadCloser, stdout rcli.DockerConn, args ...string) error {
+	stdout.Flush()
+	cmd := rcli.Subcmd(stdout, "build", "-", "Build a container from Dockerfile via stdin")
+	if err := cmd.Parse(args); err != nil {
+		return nil
+	}
+	img, err := NewBuilder(srv.runtime).Build(stdin, stdout)
+	if err != nil {
+		return err
+	}
+	fmt.Fprintf(stdout, "%s\n", img.ShortId())
+	return nil
+}
+
 // 'docker login': login / register a user to registry service.
 func (srv *Server) CmdLogin(stdin io.ReadCloser, stdout rcli.DockerConn, args ...string) error {
 	// Read a line on raw terminal with support for simple backspace
@@ -154,6 +220,7 @@ func (srv *Server) CmdLogin(stdin io.ReadCloser, stdout rcli.DockerConn, args ..
 	if err != nil {
 		fmt.Fprintf(stdout, "Error: %s\r\n", err)
 	} else {
+		srv.runtime.graph.getHttpClient().Jar = cookiejar.NewCookieJar()
 		srv.runtime.authConfig = newAuthConfig
 	}
 	if status != "" {
@@ -217,10 +284,12 @@ func (srv *Server) CmdInfo(stdin io.ReadCloser, stdout io.Writer, args ...string
 		len(srv.runtime.List()),
 		VERSION,
 		imgcount)
+	fmt.Fprintf(stdout, "Go version: %s\n", runtime.Version())
 
-	if !rcli.DEBUG_FLAG {
+	if os.Getenv("DEBUG") == "" {
 		return nil
 	}
+
 	fmt.Fprintln(stdout, "debug mode enabled")
 	fmt.Fprintf(stdout, "fds: %d\ngoroutines: %d\n", getTotalUsedFds(), runtime.NumGoroutine())
 	return nil
@@ -521,6 +590,7 @@ func (srv *Server) CmdImport(stdin io.ReadCloser, stdout rcli.DockerConn, args .
 
 func (srv *Server) CmdPush(stdin io.ReadCloser, stdout rcli.DockerConn, args ...string) error {
 	cmd := rcli.Subcmd(stdout, "push", "NAME", "Push an image or a repository to the registry")
+	registry := cmd.String("registry", "", "Registry host to push the image to")
 	if err := cmd.Parse(args); err != nil {
 		return nil
 	}
@@ -531,8 +601,8 @@ func (srv *Server) CmdPush(stdin io.ReadCloser, stdout rcli.DockerConn, args ...
 		return nil
 	}
 
-	// If the login failed, abort
-	if srv.runtime.authConfig == nil || srv.runtime.authConfig.Username == "" {
+	// If the login failed AND we're using the index, abort
+	if *registry == "" && (srv.runtime.authConfig == nil || srv.runtime.authConfig.Username == "") {
 		if err := srv.CmdLogin(stdin, stdout, args...); err != nil {
 			return err
 		}
@@ -555,9 +625,6 @@ func (srv *Server) CmdPush(stdin io.ReadCloser, stdout rcli.DockerConn, args ...
 	Debugf("Pushing [%s] to [%s]\n", local, remote)
 
 	// Try to get the image
-	// FIXME: Handle lookup
-	// FIXME: Also push the tags in case of ./docker push myrepo:mytag
-	//	img, err := srv.runtime.LookupImage(cmd.Arg(0))
 	img, err := srv.runtime.graph.Get(local)
 	if err != nil {
 		Debugf("The push refers to a repository [%s] (len: %d)\n", local, len(srv.runtime.repositories.Repositories[local]))
@@ -571,7 +638,7 @@ func (srv *Server) CmdPush(stdin io.ReadCloser, stdout rcli.DockerConn, args ...
 
 		return err
 	}
-	err = srv.runtime.graph.PushImage(stdout, img, srv.runtime.authConfig)
+	err = srv.runtime.graph.PushImage(stdout, img, *registry, nil)
 	if err != nil {
 		return err
 	}
@@ -580,6 +647,8 @@ func (srv *Server) CmdPush(stdin io.ReadCloser, stdout rcli.DockerConn, args ...
 
 func (srv *Server) CmdPull(stdin io.ReadCloser, stdout io.Writer, args ...string) error {
 	cmd := rcli.Subcmd(stdout, "pull", "NAME", "Pull an image or a repository from the registry")
+	tag := cmd.String("t", "", "Download tagged image in repository")
+	registry := cmd.String("registry", "", "Registry to download from. Necessary if image is pulled by ID")
 	if err := cmd.Parse(args); err != nil {
 		return nil
 	}
@@ -589,15 +658,20 @@ func (srv *Server) CmdPull(stdin io.ReadCloser, stdout io.Writer, args ...string
 		return nil
 	}
 
+	if strings.Contains(remote, ":") {
+		remoteParts := strings.Split(remote, ":")
+		tag = &remoteParts[1]
+		remote = remoteParts[0]
+	}
+
 	// FIXME: CmdPull should be a wrapper around Runtime.Pull()
-	if srv.runtime.graph.LookupRemoteImage(remote, srv.runtime.authConfig) {
-		if err := srv.runtime.graph.PullImage(stdout, remote, srv.runtime.authConfig); err != nil {
+	if *registry != "" {
+		if err := srv.runtime.graph.PullImage(stdout, remote, *registry, nil); err != nil {
 			return err
 		}
 		return nil
 	}
-	// FIXME: Allow pull repo:tag
-	if err := srv.runtime.graph.PullRepository(stdout, remote, "", srv.runtime.repositories, srv.runtime.authConfig); err != nil {
+	if err := srv.runtime.graph.PullRepository(stdout, remote, *tag, srv.runtime.repositories, srv.runtime.authConfig); err != nil {
 		return err
 	}
 	return nil
@@ -608,85 +682,126 @@ func (srv *Server) CmdImages(stdin io.ReadCloser, stdout io.Writer, args ...stri
 	//limit := cmd.Int("l", 0, "Only show the N most recent versions of each image")
 	quiet := cmd.Bool("q", false, "only show numeric IDs")
 	flAll := cmd.Bool("a", false, "show all images")
+	flViz := cmd.Bool("viz", false, "output graph in graphviz format")
 	if err := cmd.Parse(args); err != nil {
 		return nil
 	}
-	if cmd.NArg() > 1 {
-		cmd.Usage()
-		return nil
-	}
-	var nameFilter string
-	if cmd.NArg() == 1 {
-		nameFilter = cmd.Arg(0)
-	}
-	w := tabwriter.NewWriter(stdout, 20, 1, 3, ' ', 0)
-	if !*quiet {
-		fmt.Fprintln(w, "REPOSITORY\tTAG\tID\tCREATED")
-	}
-	var allImages map[string]*Image
-	var err error
-	if *flAll {
-		allImages, err = srv.runtime.graph.Map()
-	} else {
-		allImages, err = srv.runtime.graph.Heads()
-	}
-	if err != nil {
-		return err
-	}
-	for name, repository := range srv.runtime.repositories.Repositories {
-		if nameFilter != "" && name != nameFilter {
-			continue
+
+	if *flViz {
+		images, _ := srv.runtime.graph.All()
+		if images == nil {
+			return nil
 		}
-		for tag, id := range repository {
-			image, err := srv.runtime.graph.Get(id)
+
+		fmt.Fprintf(stdout, "digraph docker {\n")
+
+		var parentImage *Image
+		var err error
+		for _, image := range images {
+			parentImage, err = image.GetParent()
 			if err != nil {
-				log.Printf("Warning: couldn't load %s from %s/%s: %s", id, name, tag, err)
+				fmt.Errorf("Error while getting parent image: %v", err)
+				return nil
+			}
+			if parentImage != nil {
+				fmt.Fprintf(stdout, "  \"%s\" -> \"%s\"\n", parentImage.ShortId(), image.ShortId())
+			} else {
+				fmt.Fprintf(stdout, "  base -> \"%s\" [style=invis]\n", image.ShortId())
+			}
+		}
+
+		reporefs := make(map[string][]string)
+
+		for name, repository := range srv.runtime.repositories.Repositories {
+			for tag, id := range repository {
+				reporefs[TruncateId(id)] = append(reporefs[TruncateId(id)], fmt.Sprintf("%s:%s", name, tag))
+			}
+		}
+
+		for id, repos := range reporefs {
+			fmt.Fprintf(stdout, "  \"%s\" [label=\"%s\\n%s\",shape=box,fillcolor=\"paleturquoise\",style=\"filled,rounded\"];\n", id, id, strings.Join(repos, "\\n"))
+		}
+
+		fmt.Fprintf(stdout, "  base [style=invisible]\n")
+		fmt.Fprintf(stdout, "}\n")
+	} else {
+		if cmd.NArg() > 1 {
+			cmd.Usage()
+			return nil
+		}
+		var nameFilter string
+		if cmd.NArg() == 1 {
+			nameFilter = cmd.Arg(0)
+		}
+		w := tabwriter.NewWriter(stdout, 20, 1, 3, ' ', 0)
+		if !*quiet {
+			fmt.Fprintln(w, "REPOSITORY\tTAG\tID\tCREATED")
+		}
+		var allImages map[string]*Image
+		var err error
+		if *flAll {
+			allImages, err = srv.runtime.graph.Map()
+		} else {
+			allImages, err = srv.runtime.graph.Heads()
+		}
+		if err != nil {
+			return err
+		}
+		for name, repository := range srv.runtime.repositories.Repositories {
+			if nameFilter != "" && name != nameFilter {
 				continue
 			}
-			delete(allImages, id)
-			if !*quiet {
-				for idx, field := range []string{
-					/* REPOSITORY */ name,
-					/* TAG */ tag,
-					/* ID */ TruncateId(id),
-					/* CREATED */ HumanDuration(time.Now().Sub(image.Created)) + " ago",
-				} {
-					if idx == 0 {
-						w.Write([]byte(field))
-					} else {
-						w.Write([]byte("\t" + field))
-					}
+			for tag, id := range repository {
+				image, err := srv.runtime.graph.Get(id)
+				if err != nil {
+					log.Printf("Warning: couldn't load %s from %s/%s: %s", id, name, tag, err)
+					continue
+				}
+				delete(allImages, id)
+				if !*quiet {
+					for idx, field := range []string{
+						/* REPOSITORY */ name,
+						/* TAG */ tag,
+						/* ID */ TruncateId(id),
+						/* CREATED */ HumanDuration(time.Now().Sub(image.Created)) + " ago",
+					} {
+						if idx == 0 {
+							w.Write([]byte(field))
+						} else {
+							w.Write([]byte("\t" + field))
+						}
+					}
+					w.Write([]byte{'\n'})
+				} else {
+					stdout.Write([]byte(image.ShortId() + "\n"))
 				}
-				w.Write([]byte{'\n'})
-			} else {
-				stdout.Write([]byte(image.ShortId() + "\n"))
 			}
 		}
-	}
-	// Display images which aren't part of a
-	if nameFilter == "" {
-		for id, image := range allImages {
-			if !*quiet {
-				for idx, field := range []string{
-					/* REPOSITORY */ "<none>",
-					/* TAG */ "<none>",
-					/* ID */ TruncateId(id),
-					/* CREATED */ HumanDuration(time.Now().Sub(image.Created)) + " ago",
-				} {
-					if idx == 0 {
-						w.Write([]byte(field))
-					} else {
-						w.Write([]byte("\t" + field))
+		// Display images which aren't part of a
+		if nameFilter == "" {
+			for id, image := range allImages {
+				if !*quiet {
+					for idx, field := range []string{
+						/* REPOSITORY */ "<none>",
+						/* TAG */ "<none>",
+						/* ID */ TruncateId(id),
+						/* CREATED */ HumanDuration(time.Now().Sub(image.Created)) + " ago",
+					} {
+						if idx == 0 {
+							w.Write([]byte(field))
+						} else {
+							w.Write([]byte("\t" + field))
+						}
 					}
+					w.Write([]byte{'\n'})
+				} else {
+					stdout.Write([]byte(image.ShortId() + "\n"))
 				}
-				w.Write([]byte{'\n'})
-			} else {
-				stdout.Write([]byte(image.ShortId() + "\n"))
 			}
 		}
-	}
-	if !*quiet {
-		w.Flush()
+		if !*quiet {
+			w.Flush()
+		}
 	}
 	return nil
 }
@@ -771,7 +886,12 @@ func (srv *Server) CmdCommit(stdin io.ReadCloser, stdout io.Writer, args ...stri
 		}
 	}
 
-	img, err := srv.runtime.Commit(containerName, repository, tag, *flComment, *flAuthor, config)
+	container := srv.runtime.Get(containerName)
+	if container == nil {
+		return fmt.Errorf("No such container: %s", containerName)
+	}
+
+	img, err := NewBuilder(srv.runtime).Commit(container, repository, tag, *flComment, *flAuthor, config)
 	if err != nil {
 		return err
 	}
@@ -885,6 +1005,34 @@ func (srv *Server) CmdAttach(stdin io.ReadCloser, stdout rcli.DockerConn, args .
 	return <-container.Attach(stdin, nil, stdout, stdout)
 }
 
+func (srv *Server) CmdSearch(stdin io.ReadCloser, stdout rcli.DockerConn, args ...string) error {
+	cmd := rcli.Subcmd(stdout, "search", "NAME", "Search the docker index for images")
+	if err := cmd.Parse(args); err != nil {
+		return nil
+	}
+	if cmd.NArg() != 1 {
+		cmd.Usage()
+		return nil
+	}
+	term := cmd.Arg(0)
+	results, err := srv.runtime.graph.SearchRepositories(stdout, term)
+	if err != nil {
+		return err
+	}
+	fmt.Fprintf(stdout, "Found %d results matching your query (\"%s\")\n", results.NumResults, results.Query)
+	w := tabwriter.NewWriter(stdout, 20, 1, 3, ' ', 0)
+	fmt.Fprintf(w, "NAME\tDESCRIPTION\n")
+	for _, repo := range results.Results {
+		description := repo["description"]
+		if len(description) > 45 {
+			description = Trunc(description, 42) + "..."
+		}
+		fmt.Fprintf(w, "%s\t%s\n", repo["name"], description)
+	}
+	w.Flush()
+	return nil
+}
+
 // Ports type - Used to parse multiple -p flags
 type ports []int
 
@@ -989,8 +1137,10 @@ func (srv *Server) CmdRun(stdin io.ReadCloser, stdout rcli.DockerConn, args ...s
 	// or tell the client there is no options
 	stdout.Flush()
 
+	b := NewBuilder(srv.runtime)
+
 	// Create new container
-	container, err := srv.runtime.Create(config)
+	container, err := b.Create(config)
 	if err != nil {
 		// If container not found, try to pull it
 		if srv.runtime.graph.IsNotExist(err) {
@@ -998,7 +1148,7 @@ func (srv *Server) CmdRun(stdin io.ReadCloser, stdout rcli.DockerConn, args ...s
 			if err = srv.CmdPull(stdin, stdout, config.Image); err != nil {
 				return err
 			}
-			if container, err = srv.runtime.Create(config); err != nil {
+			if container, err = b.Create(config); err != nil {
 				return err
 			}
 		} else {

commands_test.go

@@ -73,6 +73,77 @@ func cmdWait(srv *Server, container *Container) error {
 	return closeWrap(stdout, stdoutPipe)
 }
 
+func cmdImages(srv *Server, args ...string) (string, error) {
+	stdout, stdoutPipe := io.Pipe()
+
+	go func() {
+		if err := srv.CmdImages(nil, stdoutPipe, args...); err != nil {
+			return
+		}
+
+		// force the pipe closed, so that the code below gets an EOF
+		stdoutPipe.Close()
+	}()
+
+	output, err := ioutil.ReadAll(stdout)
+	if err != nil {
+		return "", err
+	}
+
+	// Cleanup pipes
+	return string(output), closeWrap(stdout, stdoutPipe)
+}
+
+// TestImages checks that 'docker images' displays information correctly
+func TestImages(t *testing.T) {
+
+	runtime, err := newTestRuntime()
+	if err != nil {
+		t.Fatal(err)
+	}
+	defer nuke(runtime)
+
+	srv := &Server{runtime: runtime}
+
+	output, err := cmdImages(srv)
+
+	if !strings.Contains(output, "REPOSITORY") {
+		t.Fatal("'images' should have a header")
+	}
+	if !strings.Contains(output, "docker-ut") {
+		t.Fatal("'images' should show the docker-ut image")
+	}
+	if !strings.Contains(output, "e9aa60c60128") {
+		t.Fatal("'images' should show the docker-ut image id")
+	}
+
+	output, err = cmdImages(srv, "-q")
+
+	if strings.Contains(output, "REPOSITORY") {
+		t.Fatal("'images -q' should not have a header")
+	}
+	if strings.Contains(output, "docker-ut") {
+		t.Fatal("'images' should not show the docker-ut image name")
+	}
+	if !strings.Contains(output, "e9aa60c60128") {
+		t.Fatal("'images' should show the docker-ut image id")
+	}
+
+	output, err = cmdImages(srv, "-viz")
+
+	if !strings.HasPrefix(output, "digraph docker {") {
+		t.Fatal("'images -v' should start with the dot header")
+	}
+	if !strings.HasSuffix(output, "}\n") {
+		t.Fatal("'images -v' should end with a '}'")
+	}
+	if !strings.Contains(output, "base -> \"e9aa60c60128\" [style=invis]") {
+		t.Fatal("'images -v' should have the docker-ut image id node")
+	}
+
+	// todo: add checks for -a
+}
+
 // TestRunHostname checks that 'docker run -h' correctly sets a custom hostname
 func TestRunHostname(t *testing.T) {
 	runtime, err := newTestRuntime()
@@ -339,7 +410,7 @@ func TestAttachDisconnect(t *testing.T) {
 
 	srv := &Server{runtime: runtime}
 
-	container, err := runtime.Create(
+	container, err := NewBuilder(runtime).Create(
 		&Config{
 			Image:     GetTestImage(runtime).Id,
 			Memory:    33554432,

container.go

@@ -178,6 +178,23 @@ func (settings *NetworkSettings) PortMappingHuman() string {
 	return strings.Join(mapping, ", ")
 }
 
+// Inject the io.Reader at the given path. Note: do not close the reader
+func (container *Container) Inject(file io.Reader, pth string) error {
+	// Make sure the directory exists
+	if err := os.MkdirAll(path.Join(container.rwPath(), path.Dir(pth)), 0755); err != nil {
+		return err
+	}
+	// FIXME: Handle permissions/already existing dest
+	dest, err := os.Create(path.Join(container.rwPath(), pth))
+	if err != nil {
+		return err
+	}
+	if _, err := io.Copy(dest, file); err != nil {
+		return err
+	}
+	return nil
+}
+
 func (container *Container) Cmd() *exec.Cmd {
 	return container.cmd
 }
@@ -736,6 +753,14 @@ func (container *Container) ExportRw() (Archive, error) {
 	return Tar(container.rwPath(), Uncompressed)
 }
 
+func (container *Container) RwChecksum() (string, error) {
+	rwData, err := Tar(container.rwPath(), Xz)
+	if err != nil {
+		return "", err
+	}
+	return HashData(rwData)
+}
+
 func (container *Container) Export() (Archive, error) {
 	if err := container.EnsureMounted(); err != nil {
 		return nil, err

container_test.go

@@ -20,7 +20,7 @@ func TestIdFormat(t *testing.T) {
 		t.Fatal(err)
 	}
 	defer nuke(runtime)
-	container1, err := runtime.Create(
+	container1, err := NewBuilder(runtime).Create(
 		&Config{
 			Image: GetTestImage(runtime).Id,
 			Cmd:   []string{"/bin/sh", "-c", "echo hello world"},
@@ -44,7 +44,7 @@ func TestMultipleAttachRestart(t *testing.T) {
 		t.Fatal(err)
 	}
 	defer nuke(runtime)
-	container, err := runtime.Create(
+	container, err := NewBuilder(runtime).Create(
 		&Config{
 			Image: GetTestImage(runtime).Id,
 			Cmd: []string{"/bin/sh", "-c",
@@ -148,8 +148,10 @@ func TestDiff(t *testing.T) {
 	}
 	defer nuke(runtime)
 
+	builder := NewBuilder(runtime)
+
 	// Create a container and remove a file
-	container1, err := runtime.Create(
+	container1, err := builder.Create(
 		&Config{
 			Image: GetTestImage(runtime).Id,
 			Cmd:   []string{"/bin/rm", "/etc/passwd"},
@@ -190,7 +192,7 @@ func TestDiff(t *testing.T) {
 	}
 
 	// Create a new container from the commited image
-	container2, err := runtime.Create(
+	container2, err := builder.Create(
 		&Config{
 			Image: img.Id,
 			Cmd:   []string{"cat", "/etc/passwd"},
@@ -223,7 +225,9 @@ func TestCommitAutoRun(t *testing.T) {
 		t.Fatal(err)
 	}
 	defer nuke(runtime)
-	container1, err := runtime.Create(
+
+	builder := NewBuilder(runtime)
+	container1, err := builder.Create(
 		&Config{
 			Image: GetTestImage(runtime).Id,
 			Cmd:   []string{"/bin/sh", "-c", "echo hello > /world"},
@@ -254,8 +258,7 @@ func TestCommitAutoRun(t *testing.T) {
 	}
 
 	// FIXME: Make a TestCommit that stops here and check docker.root/layers/img.id/world
-
-	container2, err := runtime.Create(
+	container2, err := builder.Create(
 		&Config{
 			Image: img.Id,
 		},
@@ -301,7 +304,10 @@ func TestCommitRun(t *testing.T) {
 		t.Fatal(err)
 	}
 	defer nuke(runtime)
-	container1, err := runtime.Create(
+
+	builder := NewBuilder(runtime)
+
+	container1, err := builder.Create(
 		&Config{
 			Image: GetTestImage(runtime).Id,
 			Cmd:   []string{"/bin/sh", "-c", "echo hello > /world"},
@@ -333,7 +339,7 @@ func TestCommitRun(t *testing.T) {
 
 	// FIXME: Make a TestCommit that stops here and check docker.root/layers/img.id/world
 
-	container2, err := runtime.Create(
+	container2, err := builder.Create(
 		&Config{
 			Image: img.Id,
 			Cmd:   []string{"cat", "/world"},
@@ -380,7 +386,7 @@ func TestStart(t *testing.T) {
 		t.Fatal(err)
 	}
 	defer nuke(runtime)
-	container, err := runtime.Create(
+	container, err := NewBuilder(runtime).Create(
 		&Config{
 			Image:     GetTestImage(runtime).Id,
 			Memory:    33554432,
@@ -419,7 +425,7 @@ func TestRun(t *testing.T) {
 		t.Fatal(err)
 	}
 	defer nuke(runtime)
-	container, err := runtime.Create(
+	container, err := NewBuilder(runtime).Create(
 		&Config{
 			Image: GetTestImage(runtime).Id,
 			Cmd:   []string{"ls", "-al"},
@@ -447,7 +453,7 @@ func TestOutput(t *testing.T) {
 		t.Fatal(err)
 	}
 	defer nuke(runtime)
-	container, err := runtime.Create(
+	container, err := NewBuilder(runtime).Create(
 		&Config{
 			Image: GetTestImage(runtime).Id,
 			Cmd:   []string{"echo", "-n", "foobar"},
@@ -472,7 +478,7 @@ func TestKillDifferentUser(t *testing.T) {
 		t.Fatal(err)
 	}
 	defer nuke(runtime)
-	container, err := runtime.Create(&Config{
+	container, err := NewBuilder(runtime).Create(&Config{
 		Image: GetTestImage(runtime).Id,
 		Cmd:   []string{"tail", "-f", "/etc/resolv.conf"},
 		User:  "daemon",
@@ -520,7 +526,7 @@ func TestKill(t *testing.T) {
 		t.Fatal(err)
 	}
 	defer nuke(runtime)
-	container, err := runtime.Create(&Config{
+	container, err := NewBuilder(runtime).Create(&Config{
 		Image: GetTestImage(runtime).Id,
 		Cmd:   []string{"cat", "/dev/zero"},
 	},
@@ -566,7 +572,9 @@ func TestExitCode(t *testing.T) {
 	}
 	defer nuke(runtime)
 
-	trueContainer, err := runtime.Create(&Config{
+	builder := NewBuilder(runtime)
+
+	trueContainer, err := builder.Create(&Config{
 		Image: GetTestImage(runtime).Id,
 		Cmd:   []string{"/bin/true", ""},
 	})
@@ -581,7 +589,7 @@ func TestExitCode(t *testing.T) {
 		t.Errorf("Unexpected exit code %d (expected 0)", trueContainer.State.ExitCode)
 	}
 
-	falseContainer, err := runtime.Create(&Config{
+	falseContainer, err := builder.Create(&Config{
 		Image: GetTestImage(runtime).Id,
 		Cmd:   []string{"/bin/false", ""},
 	})
@@ -603,7 +611,7 @@ func TestRestart(t *testing.T) {
 		t.Fatal(err)
 	}
 	defer nuke(runtime)
-	container, err := runtime.Create(&Config{
+	container, err := NewBuilder(runtime).Create(&Config{
 		Image: GetTestImage(runtime).Id,
 		Cmd:   []string{"echo", "-n", "foobar"},
 	},
@@ -636,7 +644,7 @@ func TestRestartStdin(t *testing.T) {
 		t.Fatal(err)
 	}
 	defer nuke(runtime)
-	container, err := runtime.Create(&Config{
+	container, err := NewBuilder(runtime).Create(&Config{
 		Image: GetTestImage(runtime).Id,
 		Cmd:   []string{"cat"},
 
@@ -715,8 +723,10 @@ func TestUser(t *testing.T) {
 	}
 	defer nuke(runtime)
 
+	builder := NewBuilder(runtime)
+
 	// Default user must be root
-	container, err := runtime.Create(&Config{
+	container, err := builder.Create(&Config{
 		Image: GetTestImage(runtime).Id,
 		Cmd:   []string{"id"},
 	},
@@ -734,7 +744,7 @@ func TestUser(t *testing.T) {
 	}
 
 	// Set a username
-	container, err = runtime.Create(&Config{
+	container, err = builder.Create(&Config{
 		Image: GetTestImage(runtime).Id,
 		Cmd:   []string{"id"},
 
@@ -754,7 +764,7 @@ func TestUser(t *testing.T) {
 	}
 
 	// Set a UID
-	container, err = runtime.Create(&Config{
+	container, err = builder.Create(&Config{
 		Image: GetTestImage(runtime).Id,
 		Cmd:   []string{"id"},
 
@@ -774,7 +784,7 @@ func TestUser(t *testing.T) {
 	}
 
 	// Set a different user by uid
-	container, err = runtime.Create(&Config{
+	container, err = builder.Create(&Config{
 		Image: GetTestImage(runtime).Id,
 		Cmd:   []string{"id"},
 
@@ -796,7 +806,7 @@ func TestUser(t *testing.T) {
 	}
 
 	// Set a different user by username
-	container, err = runtime.Create(&Config{
+	container, err = builder.Create(&Config{
 		Image: GetTestImage(runtime).Id,
 		Cmd:   []string{"id"},
 
@@ -823,7 +833,9 @@ func TestMultipleContainers(t *testing.T) {
 	}
 	defer nuke(runtime)
 
-	container1, err := runtime.Create(&Config{
+	builder := NewBuilder(runtime)
+
+	container1, err := builder.Create(&Config{
 		Image: GetTestImage(runtime).Id,
 		Cmd:   []string{"cat", "/dev/zero"},
 	},
@@ -833,7 +845,7 @@ func TestMultipleContainers(t *testing.T) {
 	}
 	defer runtime.Destroy(container1)
 
-	container2, err := runtime.Create(&Config{
+	container2, err := builder.Create(&Config{
 		Image: GetTestImage(runtime).Id,
 		Cmd:   []string{"cat", "/dev/zero"},
 	},
@@ -879,7 +891,7 @@ func TestStdin(t *testing.T) {
 		t.Fatal(err)
 	}
 	defer nuke(runtime)
-	container, err := runtime.Create(&Config{
+	container, err := NewBuilder(runtime).Create(&Config{
 		Image: GetTestImage(runtime).Id,
 		Cmd:   []string{"cat"},
 
@@ -926,7 +938,7 @@ func TestTty(t *testing.T) {
 		t.Fatal(err)
 	}
 	defer nuke(runtime)
-	container, err := runtime.Create(&Config{
+	container, err := NewBuilder(runtime).Create(&Config{
 		Image: GetTestImage(runtime).Id,
 		Cmd:   []string{"cat"},
 
@@ -973,7 +985,7 @@ func TestEnv(t *testing.T) {
 		t.Fatal(err)
 	}
 	defer nuke(runtime)
-	container, err := runtime.Create(&Config{
+	container, err := NewBuilder(runtime).Create(&Config{
 		Image: GetTestImage(runtime).Id,
 		Cmd:   []string{"/usr/bin/env"},
 	},
@@ -1047,7 +1059,7 @@ func TestLXCConfig(t *testing.T) {
 	memMin := 33554432
 	memMax := 536870912
 	mem := memMin + rand.Intn(memMax-memMin)
-	container, err := runtime.Create(&Config{
+	container, err := NewBuilder(runtime).Create(&Config{
 		Image: GetTestImage(runtime).Id,
 		Cmd:   []string{"/bin/true"},
 
@@ -1074,7 +1086,7 @@ func BenchmarkRunSequencial(b *testing.B) {
 	}
 	defer nuke(runtime)
 	for i := 0; i < b.N; i++ {
-		container, err := runtime.Create(&Config{
+		container, err := NewBuilder(runtime).Create(&Config{
 			Image: GetTestImage(runtime).Id,
 			Cmd:   []string{"echo", "-n", "foo"},
 		},
@@ -1109,7 +1121,7 @@ func BenchmarkRunParallel(b *testing.B) {
 		complete := make(chan error)
 		tasks = append(tasks, complete)
 		go func(i int, complete chan error) {
-			container, err := runtime.Create(&Config{
+			container, err := NewBuilder(runtime).Create(&Config{
 				Image: GetTestImage(runtime).Id,
 				Cmd:   []string{"echo", "-n", "foo"},
 			},

contrib/docker-build/docker-build

@@ -89,7 +89,7 @@ def main():
 			# Skip comments and empty lines
 			if line == "" or line[0] == "#":
 				continue
-			op, param = line.split("	", 1)
+			op, param = line.split(None, 1)
 			print op.upper() + " " + param
 			if op == "from":
 				base = param

contrib/install.sh

@@ -36,9 +36,9 @@ else
 fi
 
 echo "Downloading docker binary and uncompressing into /usr/local/bin..."
-curl -s http://get.docker.io/builds/$(uname -s)/$(uname -m)/docker-master.tgz |
+curl -s http://get.docker.io/builds/$(uname -s)/$(uname -m)/docker-latest.tgz |
 tar -C /usr/local/bin --strip-components=1 -zxf- \
-docker-master/docker
+docker-latest/docker
 
 if [ -f /etc/init/dockerd.conf ]
 then

docs/Makefile

@@ -46,23 +46,24 @@ clean:
 docs:
 	-rm -rf $(BUILDDIR)/*
 	$(SPHINXBUILD) -b dirhtml $(ALLSPHINXOPTS) $(BUILDDIR)/html
-	cp sources/index.html $(BUILDDIR)/html/
-	cp -r sources/gettingstarted $(BUILDDIR)/html/
-	cp sources/dotcloud.yml $(BUILDDIR)/html/
-	cp sources/CNAME $(BUILDDIR)/html/
-	cp sources/.nojekyll $(BUILDDIR)/html/
-	cp sources/nginx.conf $(BUILDDIR)/html/
 	@echo
-	@echo "Build finished. The HTML pages are in $(BUILDDIR)/html."
+	@echo "Build finished. The documentation pages are now in $(BUILDDIR)/html."
+
+
+site:
+	cp -r website $(BUILDDIR)/
+	cp -r theme/docker/static/ $(BUILDDIR)/website/
+	@echo
+	@echo "The Website pages are in $(BUILDDIR)/site."
 
 connect:
-	@echo pushing changes to staging site
-	@cd _build/html/ ; \
-	@dotcloud list ; \
+	@echo connecting dotcloud to www.docker.io website, make sure to use user 1
+	@cd _build/website/ ; \
+	dotcloud list ; \
 	dotcloud connect dockerwebsite
 
 push:
-	@cd _build/html/ ; \
+	@cd _build/website/ ; \
 	dotcloud push
 
 github-deploy: docs

docs/README.md

@@ -15,6 +15,7 @@ Installation
 
 * Work in your own fork of the code, we accept pull requests.
 * Install sphinx: ``pip install sphinx``
+* Install sphinx httpdomain contrib package ``sphinxcontrib-httpdomain``
 * If pip is not available you can probably install it using your favorite package manager as **python-pip**
 
 Usage

docs/requirements.txt

@@ -0,0 +1,2 @@
+Sphinx==1.1.3
+sphinxcontrib-httpdomain==1.1.8

docs/sources/CNAME

@@ -1 +0,0 @@
-docker.io

docs/sources/builder/basics.rst

@@ -0,0 +1,130 @@
+==============
+Docker Builder
+==============
+
+.. contents:: Table of Contents
+
+1. Format
+=========
+
+The Docker builder format is quite simple:
+
+    ``instruction arguments``
+
+The first instruction must be `FROM`
+
+All instruction are to be placed in a file named `Dockerfile`
+
+In order to place comments within a Dockerfile, simply prefix the line with "`#`"
+
+2. Instructions
+===============
+
+Docker builder comes with a set of instructions:
+
+1. FROM: Set from what image to build
+2. RUN: Execute a command
+3. INSERT: Insert a remote file (http) into the image
+
+2.1 FROM
+--------
+    ``FROM <image>``
+
+The `FROM` instruction must be the first one in order for Builder to know from where to run commands.
+
+`FROM` can also be used in order to build multiple images within a single Dockerfile
+
+2.2 MAINTAINER
+--------------
+    ``MAINTAINER <name>``
+
+The `MAINTAINER` instruction allow you to set the Author field of the generated images.
+This instruction is never automatically reset.
+
+2.3 RUN
+-------
+    ``RUN <command>``
+
+The `RUN` instruction is the main one, it allows you to execute any commands on the `FROM` image and to save the results.
+You can use as many `RUN` as you want within a Dockerfile, the commands will be executed on the result of the previous command.
+
+
+2.4 CMD
+-------
+    ``CMD <command>``
+
+The `CMD` instruction sets the command to be executed when running the image.
+It is equivalent to do `docker commit -run '{"Cmd": <command>}'` outside the builder.
+
+.. note::
+    Do not confuse `RUN` with `CMD`. `RUN` actually run a command and save the result, `CMD` does not execute anything.
+
+2.5 EXPOSE
+----------
+    ``EXPOSE <port> [<port>...]``
+
+The `EXPOSE` instruction sets ports to be publicly exposed when running the image.
+This is equivalent to do `docker commit -run '{"PortSpecs": ["<port>", "<port2>"]}'` outside the builder.
+
+2.6 ENV
+-------
+    ``ENV <key> <value>``
+
+The `ENV` instruction set as environment variable `<key>` with the value `<value>`. This value will be passed to all future ``RUN`` instructions.
+
+.. note::
+    The environment variables are local to the Dockerfile, they will not be set as autorun.
+
+2.7 INSERT
+----------
+
+    ``INSERT <file url> <path>``
+
+The `INSERT` instruction will download the file at the given url and place it within the image at the given path.
+
+.. note::
+    The path must include the file name.
+
+
+3. Dockerfile Examples
+======================
+
+::
+
+    # Nginx
+    #
+    # VERSION               0.0.1
+    # DOCKER-VERSION        0.2
+    
+    from      ubuntu
+    maintainer Guillaume J. Charmes "guillaume@dotcloud.com"
+    
+    # make sure the package repository is up to date
+    run echo "deb http://archive.ubuntu.com/ubuntu precise main universe" > /etc/apt/sources.list
+    run apt-get update
+    
+    run apt-get install -y inotify-tools nginx apache2 openssh-server
+    insert https://raw.github.com/creack/docker-vps/master/nginx-wrapper.sh /usr/sbin/nginx-wrapper
+
+::
+
+    # Firefox over VNC
+    #
+    # VERSION               0.3
+    # DOCKER-VERSION        0.2
+    
+    from ubuntu
+    # make sure the package repository is up to date
+    run echo "deb http://archive.ubuntu.com/ubuntu precise main universe" > /etc/apt/sources.list
+    run apt-get update
+    
+    # Install vnc, xvfb in order to create a 'fake' display and firefox
+    run apt-get install -y x11vnc xvfb firefox
+    run mkdir /.vnc
+    # Setup a password
+    run x11vnc -storepasswd 1234 ~/.vnc/passwd
+    # Autostart firefox (might not be the best way to do it, but it does the trick)
+    run bash -c 'echo "firefox" >> /.bashrc'
+    
+    expose 5900
+    cmd    ["x11vnc", "-forever", "-usepw", "-create"]

docs/sources/builder/index.rst

@@ -0,0 +1,14 @@
+:title: docker documentation
+:description: Documentation for docker builder
+:keywords: docker, builder, dockerfile
+
+
+Builder
+=======
+
+Contents:
+
+.. toctree::
+  :maxdepth: 2
+
+  basics

docs/sources/commandline/cli.rst

@@ -27,6 +27,7 @@ Available Commands
    :maxdepth: 1
 
    command/attach
+   command/build
    command/commit
    command/diff
    command/export
@@ -46,6 +47,7 @@ Available Commands
    command/rm
    command/rmi
    command/run
+   command/search
    command/start
    command/stop
    command/tag

docs/sources/commandline/command/build.rst

@@ -0,0 +1,9 @@
+========================================================
+``build`` -- Build a container from Dockerfile via stdin
+========================================================
+
+::
+
+    Usage: docker build -
+    Example: cat Dockerfile | docker build -
+    Build a new image from the Dockerfile passed via stdin

docs/sources/commandline/command/images.rst

@@ -10,3 +10,13 @@
 
       -a=false: show all images
       -q=false: only show numeric IDs
+      -viz=false: output in graphviz format
+
+Displaying images visually
+--------------------------
+
+::
+
+    docker images -viz | dot -Tpng -o docker.png
+
+.. image:: images/docker_images.gif

docs/sources/commandline/command/run.rst

@@ -18,4 +18,5 @@
       -t=false: Allocate a pseudo-tty
       -u="": Username or UID
       -d=[]: Set custom dns servers for the container
-      -v=[]: Creates a new volumes and mount it at the specified path. A container ID can be passed instead of a path in order to mount all volumes from the given container.
+      -v=[]: Creates a new volumes and mount it at the specified path.
+      -volumes-from="": Mount all volumes from the given container.

docs/sources/commandline/command/search.rst

@@ -0,0 +1,10 @@
+===================================================================
+``search`` -- Search for an image in the docker index
+===================================================================
+
+::
+
+    Usage: docker search TERM
+
+    Searches for the TERM parameter on the Docker index and prints out a list of repositories
+    that match.

docs/sources/commandline/index.rst

@@ -9,7 +9,7 @@ Commands
 Contents:
 
 .. toctree::
-  :maxdepth: 2
+  :maxdepth: 3
 
   basics
   workingwithrepository

docs/sources/concepts/containers.rst

@@ -5,124 +5,4 @@
 
 :note: This version of the introduction is temporary, just to make sure we don't break the links from the website when the documentation is updated
 
-
-Introduction
-============
-
-Docker - The Linux container runtime
-------------------------------------
-
-Docker complements LXC with a high-level API which operates at the process level. It runs unix processes with strong guarantees of isolation and repeatability across servers.
-
-Docker is a great building block for automating distributed systems: large-scale web deployments, database clusters, continuous deployment systems, private PaaS, service-oriented architectures, etc.
-
-
-- **Heterogeneous payloads** Any combination of binaries, libraries, configuration files, scripts, virtualenvs, jars, gems, tarballs, you name it. No more juggling between domain-specific tools. Docker can deploy and run them all.
-- **Any server** Docker can run on any x64 machine with a modern linux kernel - whether it's a laptop, a bare metal server or a VM. This makes it perfect for multi-cloud deployments.
-- **Isolation** docker isolates processes from each other and from the underlying host, using lightweight containers.
-- **Repeatability** Because containers are isolated in their own filesystem, they behave the same regardless of where, when, and alongside what they run.
-
-
-
-What is a Standard Container?
------------------------------
-
-Docker defines a unit of software delivery called a Standard Container. The goal of a Standard Container is to encapsulate a software component and all its dependencies in
-a format that is self-describing and portable, so that any compliant runtime can run it without extra dependency, regardless of the underlying machine and the contents of the container.
-
-The spec for Standard Containers is currently work in progress, but it is very straightforward. It mostly defines 1) an image format, 2) a set of standard operations, and 3) an execution environment.
-
-A great analogy for this is the shipping container. Just like Standard Containers are a fundamental unit of software delivery, shipping containers (http://bricks.argz.com/ins/7823-1/12) are a fundamental unit of physical delivery.
-
-Standard operations
-~~~~~~~~~~~~~~~~~~~
-
-Just like shipping containers, Standard Containers define a set of STANDARD OPERATIONS. Shipping containers can be lifted, stacked, locked, loaded, unloaded and labelled. Similarly, standard containers can be started, stopped, copied, snapshotted, downloaded, uploaded and tagged.
-
-
-Content-agnostic
-~~~~~~~~~~~~~~~~~~~
-
-Just like shipping containers, Standard Containers are CONTENT-AGNOSTIC: all standard operations have the same effect regardless of the contents. A shipping container will be stacked in exactly the same way whether it contains Vietnamese powder coffee or spare Maserati parts. Similarly, Standard Containers are started or uploaded in the same way whether they contain a postgres database, a php application with its dependencies and application server, or Java build artifacts.
-
-
-Infrastructure-agnostic
-~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-Both types of containers are INFRASTRUCTURE-AGNOSTIC: they can be transported to thousands of facilities around the world, and manipulated by a wide variety of equipment. A shipping container can be packed in a factory in Ukraine, transported by truck to the nearest routing center, stacked onto a train, loaded into a German boat by an Australian-built crane, stored in a warehouse at a US facility, etc. Similarly, a standard container can be bundled on my laptop, uploaded to S3, downloaded, run and snapshotted by a build server at Equinix in Virginia, uploaded to 10 staging servers in a home-made Openstack cluster, then sent to 30 production instances across 3 EC2 regions.
-
-
-Designed for automation
-~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-Because they offer the same standard operations regardless of content and infrastructure, Standard Containers, just like their physical counterpart, are extremely well-suited for automation. In fact, you could say automation is their secret weapon.
-
-Many things that once required time-consuming and error-prone human effort can now be programmed. Before shipping containers, a bag of powder coffee was hauled, dragged, dropped, rolled and stacked by 10 different people in 10 different locations by the time it reached its destination. 1 out of 50 disappeared. 1 out of 20 was damaged. The process was slow, inefficient and cost a fortune - and was entirely different depending on the facility and the type of goods.
-
-Similarly, before Standard Containers, by the time a software component ran in production, it had been individually built, configured, bundled, documented, patched, vendored, templated, tweaked and instrumented by 10 different people on 10 different computers. Builds failed, libraries conflicted, mirrors crashed, post-it notes were lost, logs were misplaced, cluster updates were half-broken. The process was slow, inefficient and cost a fortune - and was entirely different depending on the language and infrastructure provider.
-
-
-Industrial-grade delivery
-~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-There are 17 million shipping containers in existence, packed with every physical good imaginable. Every single one of them can be loaded on the same boats, by the same cranes, in the same facilities, and sent anywhere in the World with incredible efficiency. It is embarrassing to think that a 30 ton shipment of coffee can safely travel half-way across the World in *less time* than it takes a software team to deliver its code from one datacenter to another sitting 10 miles away.
-
-With Standard Containers we can put an end to that embarrassment, by making INDUSTRIAL-GRADE DELIVERY of software a reality.
-
-
-Standard Container Specification
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-(TODO)
-
-Image format
-~~~~~~~~~~~~
-
-Standard operations
-~~~~~~~~~~~~~~~~~~~
-
--  Copy
--  Run
--  Stop
--  Wait
--  Commit
--  Attach standard streams
--  List filesystem changes
--  ...
-
-Execution environment
-~~~~~~~~~~~~~~~~~~~~~
-
-Root filesystem
-^^^^^^^^^^^^^^^
-
-Environment variables
-^^^^^^^^^^^^^^^^^^^^^
-
-Process arguments
-^^^^^^^^^^^^^^^^^
-
-Networking
-^^^^^^^^^^
-
-Process namespacing
-^^^^^^^^^^^^^^^^^^^
-
-Resource limits
-^^^^^^^^^^^^^^^
-
-Process monitoring
-^^^^^^^^^^^^^^^^^^
-
-Logging
-^^^^^^^
-
-Signals
-^^^^^^^
-
-Pseudo-terminal allocation
-^^^^^^^^^^^^^^^^^^^^^^^^^^
-
-Security
-^^^^^^^^
-
+This document has been moved to  :ref:`introduction`, please update your bookmarks.

docs/sources/concepts/introduction.rst

@@ -2,7 +2,7 @@
 :description: An introduction to docker and standard containers?
 :keywords: containers, lxc, concepts, explanation
 
-
+.. _introduction:
 
 Introduction
 ============

docs/sources/conf.py

@@ -25,7 +25,7 @@ import sys, os
 
 # Add any Sphinx extension module names here, as strings. They can be extensions
 # coming with Sphinx (named 'sphinx.ext.*') or your custom ones.
-extensions = []
+extensions = ['sphinxcontrib.httpdomain']
 
 # Add any paths that contain templates here, relative to this directory.
 templates_path = ['_templates']

docs/sources/examples/couchdb_data_volumes.rst

@@ -2,9 +2,9 @@
 :description: Sharing data between 2 couchdb databases
 :keywords: docker, example, package installation, networking, couchdb, data volumes
 
-.. _running_redis_service:
+.. _running_couchdb_service:
 
-Create a redis service
+Create a CouchDB service
 ======================
 
 .. include:: example_header.inc

docs/sources/index.rst

@@ -16,7 +16,9 @@ This documentation has the following resources:
    contributing/index
    commandline/index
    registry/index
+   index/index
+   builder/index
    faq
 
 
-.. image:: http://www.docker.io/_static/lego_docker.jpg
+.. image:: http://www.docker.io/_static/lego_docker.jpg

docs/sources/index/index.rst

@@ -0,0 +1,15 @@
+:title: Docker Index documentation
+:description: Documentation for docker Index
+:keywords: docker, index, api
+
+
+
+Index
+=====
+
+Contents:
+
+.. toctree::
+   :maxdepth: 2
+
+   search

docs/sources/index/search.rst

@@ -0,0 +1,38 @@
+=======================
+Docker Index Search API
+=======================
+
+Search
+------
+
+.. http:get:: /v1/search
+
+   Search the Index given a search term. It accepts :http:method:`get` only.
+
+   **Example request**:
+
+   .. sourcecode:: http
+
+      GET /v1/search?q=search_term HTTP/1.1
+      Host: example.com
+      Accept: application/json
+
+   **Example response**:
+
+   .. sourcecode:: http
+
+      HTTP/1.1 200 OK
+      Vary: Accept
+      Content-Type: application/json
+
+      {"query":"search_term",
+        "num_results": 2,
+        "results" : [
+           {"name": "dotcloud/base", "description": "A base ubuntu64  image..."},
+           {"name": "base2", "description": "A base ubuntu64  image..."},
+         ]
+       }
+   
+   :query q: what you want to search for
+   :statuscode 200: no error
+   :statuscode 500: server error

docs/sources/installation/binaries.rst

@@ -26,9 +26,9 @@ Install the docker binary:
 
 ::
 
-    wget http://get.docker.io/builds/Linux/x86_64/docker-master.tgz
-    tar -xf docker-master.tgz
-    sudo cp ./docker-master /usr/local/bin
+    wget http://get.docker.io/builds/Linux/x86_64/docker-latest.tgz
+    tar -xf docker-latest.tgz
+    sudo cp ./docker-latest/docker /usr/local/bin
 
 Note: docker currently only supports 64-bit Linux hosts.
 
@@ -50,4 +50,4 @@ Run your first container!
 
 
 
-Continue with the :ref:`hello_world` example.
+Continue with the :ref:`hello_world` example.

docs/sources/installation/ubuntulinux.rst

@@ -18,7 +18,7 @@ The linux-image-extra package is only needed on standard Ubuntu EC2 AMIs in orde
 
 .. code-block:: bash
 
-   sudo apt-get install linux-image-extra-`uname -r`
+   sudo apt-get install linux-image-extra-`uname -r` lxc bsdtar
 
 
 Installation
@@ -48,7 +48,7 @@ Now install it, you will see another warning that the package cannot be authenti
 
 .. code-block:: bash
 
-   sudo apt-get install lxc-docker
+    curl get.docker.io | sudo sh -x
 
 
 Verify it worked

docs/sources/installation/upgrading.rst

@@ -11,7 +11,7 @@ Get the latest docker binary:
 
 ::
 
-  wget http://get.docker.io/builds/$(uname -s)/$(uname -m)/docker-master.tgz
+  wget http://get.docker.io/builds/$(uname -s)/$(uname -m)/docker-latest.tgz
 
 
 
@@ -19,7 +19,7 @@ Unpack it to your current dir
 
 ::
 
-   tar -xf docker-master.tgz
+   tar -xf docker-latest.tgz
 
 
 Stop your current daemon. How you stop your daemon depends on how you started it.
@@ -38,4 +38,4 @@ Now start the daemon
    sudo ./docker -d &
 
 
-Alternatively you can replace the docker binary in ``/usr/local/bin``
+Alternatively you can replace the docker binary in ``/usr/local/bin``

docs/sources/registry/api.rst

@@ -84,7 +84,9 @@ It’s possible to run docker pull https://<registry>/repositories/samalba/busyb
 
 Currently registry redirects to s3 urls for downloads, going forward all downloads need to be streamed through the registry. The Registry will then abstract the calls to S3 by a top-level class which implements sub-classes for S3 and local storage.
 
-Token is only returned when it is a private repo, public repos do not require tokens to be returned. The Registry will still contact the Index to make sure the pull is authorized (“is it ok to download this repos without a Token?”).
+Token is only returned when the 'X-Docker-Token' header is sent with request.
+
+Basic Auth is required to pull private repos. Basic auth isn't required for pulling public repos, but if one is provided, it needs to be valid and for an active account.
 
 API (pulling repository foo/bar):
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
@@ -426,6 +428,8 @@ You have 3 options:
         - X-Docker-Token: true
 
     In this case, along with the 200 response, you’ll get a new token (if user auth is ok):
+    If authorization isn't correct you get a 401 response.
+    If account isn't active you will get a 403 response.
 
     **Response**:
         - 200 OK

docs/theme/docker/layout.html

@@ -6,6 +6,7 @@
 <head>
     <meta charset="utf-8">
     <meta http-equiv="X-UA-Compatible" content="IE=edge,chrome=1">
+    <meta name="google-site-verification" content="UxV66EKuPe87dgnH1sbrldrx6VsoWMrx5NjwkgUFxXI" />
 
     <title>Docker - {{ meta['title'] if meta and meta['title'] else title }}</title>
 
@@ -74,7 +75,7 @@
             </div>
 
             <div style="margin-left: -12px; float: left;">
-                <a href="{{ pathto('./', 1) }}"><img style="margin-top: 12px; height: 38px" src="{{ pathto('_static/img/docker-letters-logo.gif', 1) }}"></a>
+                <a href="http://www.docker.io"><img style="margin-top: 12px; height: 38px" src="{{ pathto('_static/img/docker-letters-logo.gif', 1) }}"></a>
             </div>
         </div>
 

docs/theme/docker/static/css/main.css

@@ -82,7 +82,7 @@ h4 {
 .btn-custom {
   background-color: #292929 !important;
   background-repeat: repeat-x;
-  filter: progid:DXImageTransform.Microsoft.gradient(startColorstr="#515151", endColorstr="#282828");
+  filter: progid:dximagetransform.microsoft.gradient(startColorstr="#515151", endColorstr="#282828");
   background-image: -khtml-gradient(linear, left top, left bottom, from(#515151), to(#282828));
   background-image: -moz-linear-gradient(top, #515151, #282828);
   background-image: -ms-linear-gradient(top, #515151, #282828);
@@ -330,3 +330,7 @@ section.header {
 @media (max-width: 480px) {
   
 }
+/* Misc fixes */
+table th {
+  text-align: left;
+}

docs/theme/docker/static/css/main.less

@@ -449,4 +449,9 @@ section.header {
 @media (max-width: 480px) {
 
 
+}
+
+/* Misc fixes */
+table th {
+  text-align: left;
 }

docs/website/gettingstarted/index.html

@@ -13,15 +13,15 @@
     <meta name="viewport" content="width=device-width">
 
     <!-- twitter bootstrap -->
-    <link rel="stylesheet" href="../_static/css/bootstrap.min.css">
-    <link rel="stylesheet" href="../_static/css/bootstrap-responsive.min.css">
+    <link rel="stylesheet" href="../static/css/bootstrap.min.css">
+    <link rel="stylesheet" href="../static/css/bootstrap-responsive.min.css">
 
     <!-- main style file -->
-    <link rel="stylesheet" href="../_static/css/main.css">
+    <link rel="stylesheet" href="../static/css/main.css">
 
     <!-- vendor scripts -->
-    <script src="../_static/js/vendor/jquery-1.9.1.min.js" type="text/javascript" ></script>
-    <script src="../_static/js/vendor/modernizr-2.6.2-respond-1.1.0.min.js" type="text/javascript" ></script>
+    <script src="../static/js/vendor/jquery-1.9.1.min.js" type="text/javascript" ></script>
+    <script src="../static/js/vendor/modernizr-2.6.2-respond-1.1.0.min.js" type="text/javascript" ></script>
 
 </head>
 
@@ -35,8 +35,8 @@
             <div style="float: right" class="pull-right">
                 <ul class="nav">
                     <li><a href="../">Introduction</a></li>
-                    <li class="active"><a href="./">Getting started</a></li>
-                    <li class=""><a href="http://docs.docker.io/en/latest/concepts/containers/">Documentation</a></li>
+                    <li class="active"><a href="">Getting started</a></li>
+                    <li class=""><a href="http://docs.docker.io/en/latest/concepts/introduction/">Documentation</a></li>
                 </ul>
 
                 <div class="social links" style="float: right; margin-top: 14px; margin-left: 12px">
@@ -46,7 +46,7 @@
             </div>
 
             <div style="margin-left: -12px; float: left;">
-                <a href="../index.html"><img style="margin-top: 12px; height: 38px" src="../_static/img/docker-letters-logo.gif"></a>
+                <a href="../index.html"><img style="margin-top: 12px; height: 38px" src="../static/img/docker-letters-logo.gif"></a>
             </div>
         </div>
     </div>
@@ -186,7 +186,7 @@
 
 
 <!-- bootstrap javascipts -->
-<script src="../_static/js/vendor/bootstrap.min.js" type="text/javascript"></script>
+<script src="../static/js/vendor/bootstrap.min.js" type="text/javascript"></script>
 
 <!-- Google analytics -->
 <script type="text/javascript">

docs/website/index.html

@@ -7,21 +7,45 @@
 <head>
     <meta charset="utf-8">
     <meta http-equiv="X-UA-Compatible" content="IE=edge,chrome=1">
-    <title>Docker - the Linux container runtime</title>
+    <meta name="google-site-verification" content="UxV66EKuPe87dgnH1sbrldrx6VsoWMrx5NjwkgUFxXI" />
+    <title>Docker - the Linux container engine</title>
 
     <meta name="description" content="Docker encapsulates heterogeneous payloads in standard containers">
     <meta name="viewport" content="width=device-width">
 
     <!-- twitter bootstrap -->
-    <link rel="stylesheet" href="_static/css/bootstrap.min.css">
-    <link rel="stylesheet" href="_static/css/bootstrap-responsive.min.css">
+    <link rel="stylesheet" href="static/css/bootstrap.min.css">
+    <link rel="stylesheet" href="static/css/bootstrap-responsive.min.css">
 
     <!-- main style file -->
-    <link rel="stylesheet" href="_static/css/main.css">
+    <link rel="stylesheet" href="static/css/main.css">
 
     <!-- vendor scripts -->
-    <script src="_static/js/vendor/jquery-1.9.1.min.js" type="text/javascript" ></script>
-    <script src="_static/js/vendor/modernizr-2.6.2-respond-1.1.0.min.js" type="text/javascript" ></script>
+    <script src="static/js/vendor/jquery-1.9.1.min.js" type="text/javascript" ></script>
+    <script src="static/js/vendor/modernizr-2.6.2-respond-1.1.0.min.js" type="text/javascript" ></script>
+
+    <style>
+        .indexlabel {
+            float: left;
+            width: 150px;
+            display: block;
+            padding: 10px 20px 10px;
+            font-size: 20px;
+            font-weight: 200;
+            background-color: #a30000;
+            color: white;
+            height: 22px;
+        }
+        .searchbutton {
+            font-size: 20px;
+            height: 40px;
+        }
+
+        .debug {
+            border: 1px red dotted;
+        }
+
+    </style>
 
 </head>
 
@@ -34,9 +58,9 @@
 
             <div class="pull-right" >
                 <ul class="nav">
-                    <li class="active"><a href="./">Introduction</a></li>
-                    <li ><a href="gettingstarted/">Getting started</a></li>
-                    <li class=""><a href="http://docs.docker.io/en/latest/concepts/containers/">Documentation</a></li>
+                    <li class="active"><a href="../sources">Introduction</a></li>
+                    <li ><a href="gettingstarted">Getting started</a></li>
+                    <li class=""><a href="http://docs.docker.io/en/latest/concepts/introduction/">Documentation</a></li>
                 </ul>
 
                 <div class="social links" style="float: right; margin-top: 14px; margin-left: 12px">
@@ -49,49 +73,40 @@
 </div>
 
 
-<div class="container">
-    <div class="row">
-        <div class="text-center">
-            <img src="_static/img/docker-letters-logo.gif">
-        </div>
-    </div>
-</div>
-
-<div class="container">
+<div class="container" style="margin-top: 30px;">
     <div class="row">
 
         <div class="span12">
             <section class="contentblock header">
 
-                <div class="span6" style="margin:10px 0px 0px 30px; float: right; ">
+                <div class="span5" style="margin-bottom: 15px;">
+                    <div style="text-align: center;" >
+                        <img src="static/img/docker_letters_500px.png">
+
+                        <h2>The Linux container engine</h2>
+                    </div>
+
+                    <div style="display: block; text-align: center; margin-top: 20px;">
+
+                        <h5>
+                            Docker is an open-source engine which automates the deployment of applications as highly portable, self-sufficient containers which are independent of hardware, language, framework, packaging system and hosting provider.
+                        </h5>
+
+                    </div>
+
+
+                    <div style="display: block; text-align: center; margin-top: 30px;">
+                        <a class="btn btn-custom btn-large" href="gettingstarted/">Let's get started</a>
+                    </div>
+
+                </div>
+
+                <div class="span6" >
                     <div class="js-video" >
-                        <iframe width="640" height="360" src="http://www.youtube.com/embed/wW9CAH9nSLs?feature=player_detailpage&rel=0&modestbranding=1&start=11" frameborder="0" allowfullscreen></iframe>
+                        <iframe width="600" height="360" src="http://www.youtube.com/embed/wW9CAH9nSLs?feature=player_detailpage&rel=0&modestbranding=1&start=11" frameborder="0" allowfullscreen></iframe>
                     </div>
                 </div>
 
-                <div style="text-align: center; padding: 50px 30px 50px 30px;">
-		    <h1>Docker</h1>
-		    <h2>The Linux container runtime</h2>
-                </div>
-
-                <div style="display: block; text-align: center; padding: 10px 30px 50px 30px;">
-			<p>
-			Docker complements LXC with a high-level API which operates at the process level.
-			It runs unix processes with strong guarantees of isolation and repeatability across servers.
-			</p>
-
-			<p>
-			Docker is a great building block for automating distributed systems: large-scale web deployments, database clusters, continuous deployment systems, private PaaS, service-oriented architectures, etc.
-			</p>
-                </div>
-
-
-                <div style="display: block; text-align: center;">
-                    <a class="btn btn-custom btn-large" href="gettingstarted/">Let's get started</a>
-                </div>
-
-
-
                 <br style="clear: both"/>
             </section>
         </div>
@@ -101,31 +116,56 @@
 <div class="container">
     <div class="row">
 
-        <div class="span3">
+        <div class="span6">
             <section class="contentblock">
                 <h4>Heterogeneous payloads</h4>
                 <p>Any combination of binaries, libraries, configuration files, scripts, virtualenvs, jars, gems, tarballs, you name it. No more juggling between domain-specific tools. Docker can deploy and run them all.</p>
-            </section>
-        </div>
-        <div class="span3">
-            <section class="contentblock">
                 <h4>Any server</h4>
                 <p>Docker can run on any x64 machine with a modern linux kernel - whether it's a laptop, a bare metal server or a VM. This makes it perfect for multi-cloud deployments.</p>
-            </section>
-        </div>
-        <div class="span3">
-            <section class="contentblock">
                 <h4>Isolation</h4>
-                <p>docker isolates processes from each other and from the underlying host, using lightweight containers.</p>
+                <p>Docker isolates processes from each other and from the underlying host, using lightweight containers.</p>
+                <h4>Repeatability</h4>
+                <p>Because each container is isolated in its own filesystem, they behave the same regardless of where, when, and alongside what they run.</p>
             </section>
         </div>
-        <div class="span3">
+        <div class="span6">
             <section class="contentblock">
-                <h4>Repeatability</h4>
-                <p>Because containers are isolated in their own filesystem, they behave the same regardless of where, when, and alongside what they run.</p>
+                <h1>New! Docker Index</h1>
+                On the Docker Index you can find and explore pre-made container images. It allows you to share your images and download them.
+
+                <br><br>
+                <a href="https://index.docker.io" target="_blank">
+                    <div class="indexlabel">
+                        DOCKER index
+                    </div>
+                </a>
+                &nbsp;
+                <input type="button" class="searchbutton" type="submit" value="Search images"
+                       onClick="window.open('https://index.docker.io')" />
+
+            </section>
+            <section class="contentblock">
+                <div id="wufoo-z7x3p3">
+                    Fill out my <a href="http://dotclouddocker.wufoo.com/forms/z7x3p3">online form</a>.
+                </div>
+                <script type="text/javascript">var z7x3p3;(function(d, t) {
+                    var s = d.createElement(t), options = {
+                        'userName':'dotclouddocker',
+                        'formHash':'z7x3p3',
+                        'autoResize':true,
+                        'height':'577',
+                        'async':true,
+                        'header':'show'};
+                    s.src = ('https:' == d.location.protocol ? 'https://' : 'http://') + 'wufoo.com/scripts/embed/form.js';
+                    s.onload = s.onreadystatechange = function() {
+                        var rs = this.readyState; if (rs) if (rs != 'complete') if (rs != 'loaded') return;
+                        try { z7x3p3 = new WufooForm();z7x3p3.initialize(options);z7x3p3.display(); } catch (e) {}};
+                    var scr = d.getElementsByTagName(t)[0], par = scr.parentNode; par.insertBefore(s, scr);
+                })(document, 'script');</script>
             </section>
         </div>
     </div>
+
 </div>
 
 <style>
@@ -173,18 +213,12 @@
     </div>
 </div>
 
-<!--         <p>Docker encapsulates heterogeneous payloads in <a href="#container">Standard Containers</a>, and runs them on any server with strong guarantees of isolation and repeatability.</p>
-
-        <p>It is a great building block for automating distributed systems: large-scale web deployments, database clusters, continuous deployment systems, private PaaS, service-oriented architectures, etc.</p>
-
- -->
 <div class="container">
     <div class="row">
         <div class="span6">
 
             <section class="contentblock">
 
-                <!-- <img src="_static/lego_docker.jpg" width="600px" style="float:right; margin-left: 10px"> -->
                 <h2>Notable features</h2>
 
                 <ul>
@@ -208,35 +242,26 @@
                     <li><a href="http://lxc.sourceforge.net/">lxc</a>, a set of convenience scripts to simplify the creation of linux containers.</li>
                 </ul>
 
+                <h2>Who started it</h2>
+                <p>
+                    Docker is an open-source implementation of the deployment engine which powers <a href="http://dotcloud.com">dotCloud</a>, a popular Platform-as-a-Service.</p>
+
+                <p>It benefits directly from the experience accumulated over several years of large-scale operation and support of hundreds of thousands
+                    of applications and databases.
+                </p>
+
             </section>
         </div>
 
         <div class="span6">
-            <section class="contentblock">
-                <div id="wufoo-z7x3p3">
-                    Fill out my <a href="http://dotclouddocker.wufoo.com/forms/z7x3p3">online form</a>.
-                </div>
-                <script type="text/javascript">var z7x3p3;(function(d, t) {
-                    var s = d.createElement(t), options = {
-                        'userName':'dotclouddocker',
-                        'formHash':'z7x3p3',
-                        'autoResize':true,
-                        'height':'577',
-                        'async':true,
-                        'header':'show'};
-                    s.src = ('https:' == d.location.protocol ? 'https://' : 'http://') + 'wufoo.com/scripts/embed/form.js';
-                    s.onload = s.onreadystatechange = function() {
-                        var rs = this.readyState; if (rs) if (rs != 'complete') if (rs != 'loaded') return;
-                        try { z7x3p3 = new WufooForm();z7x3p3.initialize(options);z7x3p3.display(); } catch (e) {}};
-                    var scr = d.getElementsByTagName(t)[0], par = scr.parentNode; par.insertBefore(s, scr);
-                })(document, 'script');</script>
-            </section>
+
 
             <section class="contentblock">
                 <h3 id="twitter">Twitter</h3>
                 <a class="twitter-timeline" href="https://twitter.com/getdocker" data-widget-id="312730839718957056">Tweets by @getdocker</a>
                 <script>!function(d,s,id){var js,fjs=d.getElementsByTagName(s)[0];if(!d.getElementById(id)){js=d.createElement(s);js.id=id;js.src="//platform.twitter.com/widgets.js";fjs.parentNode.insertBefore(js,fjs);}}(document,"script","twitter-wjs");</script>
             </section>
+
         </div>
     </div>
 
@@ -265,7 +290,7 @@
 
 
 <!-- bootstrap javascipts -->
-<script src="_static/js/vendor/bootstrap.min.js" type="text/javascript"></script>
+<script src="static/js/vendor/bootstrap.min.js" type="text/javascript"></script>
 
 <!-- Google analytics -->
 <script type="text/javascript">

docs/website/static

@@ -0,0 +1 @@
+../theme/docker/static

graph.go

@@ -1,20 +1,27 @@
 package docker
 
 import (
+	"encoding/json"
 	"fmt"
 	"io"
 	"io/ioutil"
+	"net/http"
 	"os"
 	"path"
 	"path/filepath"
 	"strings"
+	"sync"
 	"time"
 )
 
 // A Graph is a store for versioned filesystem images and the relationship between them.
 type Graph struct {
-	Root    string
-	idIndex *TruncIndex
+	Root         string
+	idIndex      *TruncIndex
+	httpClient   *http.Client
+	checksumLock map[string]*sync.Mutex
+	lockSumFile  *sync.Mutex
+	lockSumMap   *sync.Mutex
 }
 
 // NewGraph instantiates a new graph at the given root path in the filesystem.
@@ -25,12 +32,15 @@ func NewGraph(root string) (*Graph, error) {
 		return nil, err
 	}
 	// Create the root directory if it doesn't exists
-	if err := os.Mkdir(root, 0700); err != nil && !os.IsExist(err) {
+	if err := os.MkdirAll(root, 0700); err != nil && !os.IsExist(err) {
 		return nil, err
 	}
 	graph := &Graph{
-		Root:    abspath,
-		idIndex: NewTruncIndex(),
+		Root:         abspath,
+		idIndex:      NewTruncIndex(),
+		checksumLock: make(map[string]*sync.Mutex),
+		lockSumFile:  &sync.Mutex{},
+		lockSumMap:   &sync.Mutex{},
 	}
 	if err := graph.restore(); err != nil {
 		return nil, err
@@ -80,6 +90,11 @@ func (graph *Graph) Get(name string) (*Image, error) {
 		return nil, fmt.Errorf("Image stored at '%s' has wrong id '%s'", id, img.Id)
 	}
 	img.graph = graph
+	graph.lockSumMap.Lock()
+	defer graph.lockSumMap.Unlock()
+	if _, exists := graph.checksumLock[img.Id]; !exists {
+		graph.checksumLock[img.Id] = &sync.Mutex{}
+	}
 	return img, nil
 }
 
@@ -97,21 +112,17 @@ func (graph *Graph) Create(layerData Archive, container *Container, comment, aut
 		img.Parent = container.Image
 		img.Container = container.Id
 		img.ContainerConfig = *container.Config
-		if config == nil {
-			if parentImage, err := graph.Get(container.Image); err == nil && parentImage != nil {
-				img.Config = parentImage.Config
-			}
-		}
 	}
-	if err := graph.Register(layerData, img); err != nil {
+	if err := graph.Register(layerData, true, img); err != nil {
 		return nil, err
 	}
+	go img.Checksum()
 	return img, nil
 }
 
 // Register imports a pre-existing image into the graph.
 // FIXME: pass img as first argument
-func (graph *Graph) Register(layerData Archive, img *Image) error {
+func (graph *Graph) Register(layerData Archive, store bool, img *Image) error {
 	if err := ValidateId(img.Id); err != nil {
 		return err
 	}
@@ -124,7 +135,7 @@ func (graph *Graph) Register(layerData Archive, img *Image) error {
 	if err != nil {
 		return fmt.Errorf("Mktemp failed: %s", err)
 	}
-	if err := StoreImage(img, layerData, tmp); err != nil {
+	if err := StoreImage(img, layerData, tmp, store); err != nil {
 		return err
 	}
 	// Commit
@@ -133,6 +144,7 @@ func (graph *Graph) Register(layerData Archive, img *Image) error {
 	}
 	img.graph = graph
 	graph.idIndex.Add(img.Id)
+	graph.checksumLock[img.Id] = &sync.Mutex{}
 	return nil
 }
 
@@ -255,14 +267,14 @@ func (graph *Graph) WalkAll(handler func(*Image)) error {
 func (graph *Graph) ByParent() (map[string][]*Image, error) {
 	byParent := make(map[string][]*Image)
 	err := graph.WalkAll(func(image *Image) {
-		image, err := graph.Get(image.Parent)
+		parent, err := graph.Get(image.Parent)
 		if err != nil {
 			return
 		}
-		if children, exists := byParent[image.Parent]; exists {
-			byParent[image.Parent] = []*Image{image}
+		if children, exists := byParent[parent.Id]; exists {
+			byParent[parent.Id] = []*Image{image}
 		} else {
-			byParent[image.Parent] = append(children, image)
+			byParent[parent.Id] = append(children, image)
 		}
 	})
 	return byParent, err
@@ -289,3 +301,26 @@ func (graph *Graph) Heads() (map[string]*Image, error) {
 func (graph *Graph) imageRoot(id string) string {
 	return path.Join(graph.Root, id)
 }
+
+func (graph *Graph) getStoredChecksums() (map[string]string, error) {
+	checksums := make(map[string]string)
+	// FIXME: Store the checksum in memory
+
+	if checksumDict, err := ioutil.ReadFile(path.Join(graph.Root, "checksums")); err == nil {
+		if err := json.Unmarshal(checksumDict, &checksums); err != nil {
+			return nil, err
+		}
+	}
+	return checksums, nil
+}
+
+func (graph *Graph) storeChecksums(checksums map[string]string) error {
+	checksumJson, err := json.Marshal(checksums)
+	if err != nil {
+		return err
+	}
+	if err := ioutil.WriteFile(path.Join(graph.Root, "checksums"), checksumJson, 0600); err != nil {
+		return err
+	}
+	return nil
+}

image.go

@@ -2,6 +2,7 @@ package docker
 
 import (
 	"crypto/rand"
+	"crypto/sha256"
 	"encoding/hex"
 	"encoding/json"
 	"fmt"
@@ -34,8 +35,9 @@ func LoadImage(root string) (*Image, error) {
 	if err != nil {
 		return nil, err
 	}
-	var img Image
-	if err := json.Unmarshal(jsonData, &img); err != nil {
+	img := &Image{}
+
+	if err := json.Unmarshal(jsonData, img); err != nil {
 		return nil, err
 	}
 	if err := ValidateId(img.Id); err != nil {
@@ -51,10 +53,10 @@ func LoadImage(root string) (*Image, error) {
 	} else if !stat.IsDir() {
 		return nil, fmt.Errorf("Couldn't load image %s: %s is not a directory", img.Id, layerPath(root))
 	}
-	return &img, nil
+	return img, nil
 }
 
-func StoreImage(img *Image, layerData Archive, root string) error {
+func StoreImage(img *Image, layerData Archive, root string, store bool) error {
 	// Check that root doesn't already exist
 	if _, err := os.Stat(root); err == nil {
 		return fmt.Errorf("Image %s already exists", img.Id)
@@ -66,6 +68,28 @@ func StoreImage(img *Image, layerData Archive, root string) error {
 	if err := os.MkdirAll(layer, 0700); err != nil {
 		return err
 	}
+
+	if store {
+		layerArchive := layerArchivePath(root)
+		file, err := os.OpenFile(layerArchive, os.O_WRONLY|os.O_CREATE, 0600)
+		if err != nil {
+			return err
+		}
+		// FIXME: Retrieve the image layer size from here?
+		if _, err := io.Copy(file, layerData); err != nil {
+			return err
+		}
+		// FIXME: Don't close/open, read/write instead of Copy
+		file.Close()
+
+		file, err = os.Open(layerArchive)
+		if err != nil {
+			return err
+		}
+		defer file.Close()
+		layerData = file
+	}
+
 	if err := Untar(layerData, layer); err != nil {
 		return err
 	}
@@ -84,6 +108,10 @@ func layerPath(root string) string {
 	return path.Join(root, "layer")
 }
 
+func layerArchivePath(root string) string {
+	return path.Join(root, "layer.tar.xz")
+}
+
 func jsonPath(root string) string {
 	return path.Join(root, "json")
 }
@@ -257,3 +285,77 @@ func (img *Image) layer() (string, error) {
 	}
 	return layerPath(root), nil
 }
+
+func (img *Image) Checksum() (string, error) {
+	img.graph.checksumLock[img.Id].Lock()
+	defer img.graph.checksumLock[img.Id].Unlock()
+
+	root, err := img.root()
+	if err != nil {
+		return "", err
+	}
+
+	checksums, err := img.graph.getStoredChecksums()
+	if err != nil {
+		return "", err
+	}
+	if checksum, ok := checksums[img.Id]; ok {
+		return checksum, nil
+	}
+
+	layer, err := img.layer()
+	if err != nil {
+		return "", err
+	}
+	jsonData, err := ioutil.ReadFile(jsonPath(root))
+	if err != nil {
+		return "", err
+	}
+
+	var layerData io.Reader
+
+	if file, err := os.Open(layerArchivePath(root)); err != nil {
+		if os.IsNotExist(err) {
+			layerData, err = Tar(layer, Xz)
+			if err != nil {
+				return "", err
+			}
+		} else {
+			return "", err
+		}
+	} else {
+		defer file.Close()
+		layerData = file
+	}
+
+	h := sha256.New()
+	if _, err := h.Write(jsonData); err != nil {
+		return "", err
+	}
+	if _, err := h.Write([]byte("\n")); err != nil {
+		return "", err
+	}
+
+	if _, err := io.Copy(h, layerData); err != nil {
+		return "", err
+	}
+	hash := "sha256:" + hex.EncodeToString(h.Sum(nil))
+
+	// Reload the json file to make sure not to overwrite faster sums
+	img.graph.lockSumFile.Lock()
+	defer img.graph.lockSumFile.Unlock()
+
+	checksums, err = img.graph.getStoredChecksums()
+	if err != nil {
+		return "", err
+	}
+
+	checksums[img.Id] = hash
+
+	// Dump the checksums to disc
+	if err := img.graph.storeChecksums(checksums); err != nil {
+		return hash, err
+	}
+
+	return hash, nil
+}

packaging/ubuntu/Vagrantfile

@@ -1,12 +1,15 @@
-BUILDBOT_IP = '192.168.33.32'
+GOPHERS_KEY = "308C15A29AD198E9"
 
 Vagrant::Config.run do |config|
   config.vm.box = 'precise64'
   config.vm.box_url = 'http://files.vagrantup.com/precise64.box'
   config.vm.share_folder 'v-data', '/data/docker', "#{File.dirname(__FILE__)}/../.."
-  config.vm.network :hostonly,BUILDBOT_IP
 
+  # Add docker PPA key to the local repository and install docker
+  pkg_cmd = "apt-key adv --keyserver hkp://keyserver.ubuntu.com:80 --recv-keys #{GOPHERS_KEY}; " \
+    "echo 'deb http://ppa.launchpad.net/gophers/go/ubuntu precise main' >/etc/apt/sources.list.d/gophers-go.list; " \
   # Install ubuntu packaging dependencies and create ubuntu packages
-  config.vm.provision :shell, :inline => 'export DEBIAN_FRONTEND=noninteractive; apt-get -qq update; apt-get install -qq -y git debhelper autotools-dev devscripts golang'
-  config.vm.provision :shell, :inline => "export GPG_KEY='#{ENV['GPG_KEY']}'; cd /data/docker/packaging/ubuntu; make ubuntu"
+  pkg_cmd << "export DEBIAN_FRONTEND=noninteractive; apt-get -qq update; apt-get install -qq -y git debhelper autotools-dev devscripts golang-stable; " \
+    "export GPG_KEY='#{ENV['GPG_KEY']}'; cd /data/docker/packaging/ubuntu; make ubuntu"
+  config.vm.provision :shell, :inline => pkg_cmd
 end

packaging/ubuntu/changelog

@@ -1,16 +1,66 @@
+lxc-docker (0.3.2-1) precise; urgency=low
+ - Runtime: Store the actual archive on commit
+ - Registry: Improve the checksum process
+ - Registry: Use the size to have a good progress bar while pushing
+ - Registry: Use the actual archive if it exists in order to speed up the push
+ - Registry: Fix error 400 on push
+
+ -- dotCloud <ops@dotcloud.com>  Fri, 9 May 2013 00:00:00 -0700
+
+
+lxc-docker (0.3.1-1) precise; urgency=low
+ - Builder: Implement the autorun capability within docker builder
+ - Builder: Add caching to docker builder
+ - Builder: Add support for docker builder with native API as top level command
+ - Runtime: Add go version to debug infos
+ - Builder: Implement ENV within docker builder
+ - Registry: Add docker search top level command in order to search a repository
+ - Images: output graph of images to dot (graphviz)
+ - Documentation: new introduction and high-level overview
+ - Documentation: Add the documentation for docker builder
+ - Website: new high-level overview
+ - Makefile: Swap "go get" for "go get -d", especially to compile on go1.1rc
+ - Images: fix ByParent function
+ - Builder: Check the command existance prior create and add Unit tests for the case
+ - Registry: Fix pull for official images with specific tag
+ - Registry: Fix issue when login in with a different user and trying to push
+ - Documentation: CSS fix for docker documentation to make REST API docs look better.
+ - Documentation: Fixed CouchDB example page header mistake
+ - Documentation: fixed README formatting
+ - Registry: Improve checksum - async calculation
+ - Runtime: kernel version - don't show the dash if flavor is empty
+ - Documentation: updated www.docker.io website.
+ - Builder: use any whitespaces instead of tabs
+ - Packaging: packaging ubuntu; issue #510: Use goland-stable PPA package to build docker
+
+ -- dotCloud <ops@dotcloud.com>  Fri, 8 May 2013 00:00:00 -0700
+
+
+lxc-docker (0.3.0-1) precise; urgency=low
+  - Registry: Implement the new registry
+  - Documentation: new example: sharing data between 2 couchdb databases
+  - Runtime: Fix the command existance check
+  - Runtime: strings.Split may return an empty string on no match
+  - Runtime: Fix an index out of range crash if cgroup memory is not
+  - Documentation: Various improvments
+  - Vagrant: Use only one deb line in /etc/apt
+
+ -- dotCloud <ops@dotcloud.com>  Fri, 5 May 2013 00:00:00 -0700
+
+
 lxc-docker (0.2.2-1) precise; urgency=low
- - Support for data volumes ('docker run -v=PATH')
- - Share data volumes between containers ('docker run -volumes-from')
- - Improved documentation
- - Upgrade to Go 1.0.3
- - Various upgrades to the dev environment for contributors
+  - Support for data volumes ('docker run -v=PATH')
+  - Share data volumes between containers ('docker run -volumes-from')
+  - Improved documentation
+  - Upgrade to Go 1.0.3
+  - Various upgrades to the dev environment for contributors
 
  -- dotCloud <ops@dotcloud.com>  Fri, 3 May 2013 00:00:00 -0700
 
 
 lxc-docker (0.2.1-1) precise; urgency=low
 
-  - 'docker commit -run' bundles a layer with default runtime options: command, ports etc. 
+  - 'docker commit -run' bundles a layer with default runtime options: command, ports etc.
   - Improve install process on Vagrant
   - New Dockerfile operation: "maintainer"
   - New Dockerfile operation: "expose"

packaging/ubuntu/control

@@ -2,7 +2,7 @@ Source: lxc-docker
 Section: misc
 Priority: extra
 Maintainer: Daniel Mizyrycki <daniel@dotcloud.com>
-Build-Depends: debhelper,autotools-dev,devscripts,golang
+Build-Depends: debhelper,autotools-dev,devscripts,golang-stable
 Standards-Version: 3.9.3
 Homepage: http://github.com/dotcloud/docker
 

registry.go

@@ -1,20 +1,22 @@
 package docker
 
 import (
+	"bytes"
 	"encoding/json"
 	"fmt"
 	"github.com/dotcloud/docker/auth"
+	"github.com/shin-/cookiejar"
 	"io"
 	"io/ioutil"
 	"net/http"
+	"net/url"
 	"os"
 	"path"
 	"strings"
 )
 
 //FIXME: Set the endpoint in a conf file or via commandline
-//const REGISTRY_ENDPOINT = "http://registry-creack.dotcloud.com/v1"
-const REGISTRY_ENDPOINT = auth.REGISTRY_SERVER + "/v1"
+const INDEX_ENDPOINT = auth.INDEX_SERVER + "/v1"
 
 // Build an Image object from raw json data
 func NewImgJson(src []byte) (*Image, error) {
@@ -28,34 +30,23 @@ func NewImgJson(src []byte) (*Image, error) {
 	return ret, nil
 }
 
-// Build an Image object list from a raw json data
-// FIXME: Do this in "stream" mode
-func NewMultipleImgJson(src []byte) ([]*Image, error) {
-	ret := []*Image{}
-
-	dec := json.NewDecoder(strings.NewReader(string(src)))
-	for {
-		m := &Image{}
-		if err := dec.Decode(m); err == io.EOF {
-			break
-		} else if err != nil {
-			return nil, err
-		}
-		ret = append(ret, m)
+func doWithCookies(c *http.Client, req *http.Request) (*http.Response, error) {
+	for _, cookie := range c.Jar.Cookies(req.URL) {
+		req.AddCookie(cookie)
 	}
-	return ret, nil
+	return c.Do(req)
 }
 
 // Retrieve the history of a given image from the Registry.
 // Return a list of the parent's json (requested image included)
-func (graph *Graph) getRemoteHistory(imgId string, authConfig *auth.AuthConfig) ([]*Image, error) {
-	client := &http.Client{}
+func (graph *Graph) getRemoteHistory(imgId, registry string, token []string) ([]string, error) {
+	client := graph.getHttpClient()
 
-	req, err := http.NewRequest("GET", REGISTRY_ENDPOINT+"/images/"+imgId+"/history", nil)
+	req, err := http.NewRequest("GET", registry+"/images/"+imgId+"/ancestry", nil)
 	if err != nil {
 		return nil, err
 	}
-	req.SetBasicAuth(authConfig.Username, authConfig.Password)
+	req.Header.Set("Authorization", "Token "+strings.Join(token, ", "))
 	res, err := client.Do(req)
 	if err != nil || res.StatusCode != 200 {
 		if res != nil {
@@ -70,41 +61,83 @@ func (graph *Graph) getRemoteHistory(imgId string, authConfig *auth.AuthConfig)
 		return nil, fmt.Errorf("Error while reading the http response: %s\n", err)
 	}
 
-	history, err := NewMultipleImgJson(jsonString)
-	if err != nil {
-		return nil, fmt.Errorf("Error while parsing the json: %s\n", err)
+	Debugf("Ancestry: %s", jsonString)
+	history := new([]string)
+	if err := json.Unmarshal(jsonString, history); err != nil {
+		return nil, err
 	}
-	return history, nil
+	return *history, nil
+}
+
+func (graph *Graph) getHttpClient() *http.Client {
+	if graph.httpClient == nil {
+		graph.httpClient = &http.Client{}
+		graph.httpClient.Jar = cookiejar.NewCookieJar()
+	}
+	return graph.httpClient
 }
 
 // Check if an image exists in the Registry
-func (graph *Graph) LookupRemoteImage(imgId string, authConfig *auth.AuthConfig) bool {
+func (graph *Graph) LookupRemoteImage(imgId, registry string, authConfig *auth.AuthConfig) bool {
 	rt := &http.Transport{Proxy: http.ProxyFromEnvironment}
 
-	req, err := http.NewRequest("GET", REGISTRY_ENDPOINT+"/images/"+imgId+"/json", nil)
+	req, err := http.NewRequest("GET", registry+"/images/"+imgId+"/json", nil)
 	if err != nil {
 		return false
 	}
 	req.SetBasicAuth(authConfig.Username, authConfig.Password)
 	res, err := rt.RoundTrip(req)
-	if err != nil || res.StatusCode != 307 {
-		return false
+	return err == nil && res.StatusCode == 307
+}
+
+func (graph *Graph) getImagesInRepository(repository string, authConfig *auth.AuthConfig) ([]map[string]string, error) {
+	u := INDEX_ENDPOINT + "/repositories/" + repository + "/images"
+	req, err := http.NewRequest("GET", u, nil)
+	if err != nil {
+		return nil, err
 	}
-	return res.StatusCode == 307
+	if authConfig != nil && len(authConfig.Username) > 0 {
+		req.SetBasicAuth(authConfig.Username, authConfig.Password)
+	}
+	res, err := graph.getHttpClient().Do(req)
+	if err != nil {
+		return nil, err
+	}
+	defer res.Body.Close()
+
+	// Repository doesn't exist yet
+	if res.StatusCode == 404 {
+		return nil, nil
+	}
+
+	jsonData, err := ioutil.ReadAll(res.Body)
+	if err != nil {
+		return nil, err
+	}
+
+	imageList := []map[string]string{}
+
+	err = json.Unmarshal(jsonData, &imageList)
+	if err != nil {
+		Debugf("Body: %s (%s)\n", res.Body, u)
+		return nil, err
+	}
+
+	return imageList, nil
 }
 
 // Retrieve an image from the Registry.
 // Returns the Image object as well as the layer as an Archive (io.Reader)
-func (graph *Graph) getRemoteImage(stdout io.Writer, imgId string, authConfig *auth.AuthConfig) (*Image, Archive, error) {
-	client := &http.Client{}
+func (graph *Graph) getRemoteImage(stdout io.Writer, imgId, registry string, token []string) (*Image, Archive, error) {
+	client := graph.getHttpClient()
 
 	fmt.Fprintf(stdout, "Pulling %s metadata\r\n", imgId)
 	// Get the Json
-	req, err := http.NewRequest("GET", REGISTRY_ENDPOINT+"/images/"+imgId+"/json", nil)
+	req, err := http.NewRequest("GET", registry+"/images/"+imgId+"/json", nil)
 	if err != nil {
 		return nil, nil, fmt.Errorf("Failed to download json: %s", err)
 	}
-	req.SetBasicAuth(authConfig.Username, authConfig.Password)
+	req.Header.Set("Authorization", "Token "+strings.Join(token, ", "))
 	res, err := client.Do(req)
 	if err != nil {
 		return nil, nil, fmt.Errorf("Failed to download json: %s", err)
@@ -127,11 +160,11 @@ func (graph *Graph) getRemoteImage(stdout io.Writer, imgId string, authConfig *a
 
 	// Get the layer
 	fmt.Fprintf(stdout, "Pulling %s fs layer\r\n", imgId)
-	req, err = http.NewRequest("GET", REGISTRY_ENDPOINT+"/images/"+imgId+"/layer", nil)
+	req, err = http.NewRequest("GET", registry+"/images/"+imgId+"/layer", nil)
 	if err != nil {
 		return nil, nil, fmt.Errorf("Error while getting from the server: %s\n", err)
 	}
-	req.SetBasicAuth(authConfig.Username, authConfig.Password)
+	req.Header.Set("Authorization", "Token "+strings.Join(token, ", "))
 	res, err = client.Do(req)
 	if err != nil {
 		return nil, nil, err
@@ -139,21 +172,95 @@ func (graph *Graph) getRemoteImage(stdout io.Writer, imgId string, authConfig *a
 	return img, ProgressReader(res.Body, int(res.ContentLength), stdout, "Downloading %v/%v (%v)"), nil
 }
 
-func (graph *Graph) PullImage(stdout io.Writer, imgId string, authConfig *auth.AuthConfig) error {
-	history, err := graph.getRemoteHistory(imgId, authConfig)
+func (graph *Graph) getRemoteTags(stdout io.Writer, registries []string, repository string, token []string) (map[string]string, error) {
+	client := graph.getHttpClient()
+	if strings.Count(repository, "/") == 0 {
+		// This will be removed once the Registry supports auto-resolution on
+		// the "library" namespace
+		repository = "library/" + repository
+	}
+	for _, host := range registries {
+		endpoint := fmt.Sprintf("https://%s/v1/repositories/%s/tags", host, repository)
+		req, err := http.NewRequest("GET", endpoint, nil)
+		if err != nil {
+			return nil, err
+		}
+		req.Header.Set("Authorization", "Token "+strings.Join(token, ", "))
+		res, err := client.Do(req)
+		defer res.Body.Close()
+		Debugf("Got status code %d from %s", res.StatusCode, endpoint)
+		if err != nil || (res.StatusCode != 200 && res.StatusCode != 404) {
+			continue
+		} else if res.StatusCode == 404 {
+			return nil, fmt.Errorf("Repository not found")
+		}
+
+		result := make(map[string]string)
+
+		rawJson, err := ioutil.ReadAll(res.Body)
+		if err != nil {
+			return nil, err
+		}
+		if err = json.Unmarshal(rawJson, &result); err != nil {
+			return nil, err
+		}
+		return result, nil
+	}
+	return nil, fmt.Errorf("Could not reach any registry endpoint")
+}
+
+func (graph *Graph) getImageForTag(stdout io.Writer, tag, remote, registry string, token []string) (string, error) {
+	client := graph.getHttpClient()
+
+	if !strings.Contains(remote, "/") {
+		remote = "library/" + remote
+	}
+
+	registryEndpoint := "https://" + registry + "/v1"
+	repositoryTarget := registryEndpoint + "/repositories/" + remote + "/tags/" + tag
+
+	req, err := http.NewRequest("GET", repositoryTarget, nil)
+	if err != nil {
+		return "", err
+	}
+	req.Header.Set("Authorization", "Token "+strings.Join(token, ", "))
+	res, err := client.Do(req)
+	if err != nil {
+		return "", fmt.Errorf("Error while retrieving repository info: %v", err)
+	}
+	defer res.Body.Close()
+	if res.StatusCode == 403 {
+		return "", fmt.Errorf("You aren't authorized to access this resource")
+	} else if res.StatusCode != 200 {
+		return "", fmt.Errorf("HTTP code: %d", res.StatusCode)
+	}
+
+	var imgId string
+	rawJson, err := ioutil.ReadAll(res.Body)
+	if err != nil {
+		return "", err
+	}
+	if err = json.Unmarshal(rawJson, &imgId); err != nil {
+		return "", err
+	}
+	return imgId, nil
+}
+
+func (graph *Graph) PullImage(stdout io.Writer, imgId, registry string, token []string) error {
+	history, err := graph.getRemoteHistory(imgId, registry, token)
 	if err != nil {
 		return err
 	}
 	// FIXME: Try to stream the images?
-	// FIXME: Lunch the getRemoteImage() in goroutines
-	for _, j := range history {
-		if !graph.Exists(j.Id) {
-			img, layer, err := graph.getRemoteImage(stdout, j.Id, authConfig)
+	// FIXME: Launch the getRemoteImage() in goroutines
+	for _, id := range history {
+		if !graph.Exists(id) {
+			img, layer, err := graph.getRemoteImage(stdout, id, registry, token)
 			if err != nil {
 				// FIXME: Keep goging in case of error?
 				return err
 			}
-			if err = graph.Register(layer, img); err != nil {
+			if err = graph.Register(layer, false, img); err != nil {
 				return err
 			}
 		}
@@ -161,165 +268,254 @@ func (graph *Graph) PullImage(stdout io.Writer, imgId string, authConfig *auth.A
 	return nil
 }
 
-// FIXME: Handle the askedTag parameter
 func (graph *Graph) PullRepository(stdout io.Writer, remote, askedTag string, repositories *TagStore, authConfig *auth.AuthConfig) error {
-	client := &http.Client{}
+	client := graph.getHttpClient()
 
-	fmt.Fprintf(stdout, "Pulling repository %s\r\n", remote)
-
-	var repositoryTarget string
-	// If we are asking for 'root' repository, lookup on the Library's registry
-	if strings.Index(remote, "/") == -1 {
-		repositoryTarget = REGISTRY_ENDPOINT + "/library/" + remote
-	} else {
-		repositoryTarget = REGISTRY_ENDPOINT + "/users/" + remote
-	}
+	fmt.Fprintf(stdout, "Pulling repository %s from %s\r\n", remote, INDEX_ENDPOINT)
+	repositoryTarget := INDEX_ENDPOINT + "/repositories/" + remote + "/images"
 
 	req, err := http.NewRequest("GET", repositoryTarget, nil)
 	if err != nil {
 		return err
 	}
-	req.SetBasicAuth(authConfig.Username, authConfig.Password)
+	if authConfig != nil && len(authConfig.Username) > 0 {
+		req.SetBasicAuth(authConfig.Username, authConfig.Password)
+	}
+	req.Header.Set("X-Docker-Token", "true")
+
 	res, err := client.Do(req)
 	if err != nil {
 		return err
 	}
 	defer res.Body.Close()
+	if res.StatusCode == 401 {
+		return fmt.Errorf("Please login first (HTTP code %d)", res.StatusCode)
+	}
+	// TODO: Right now we're ignoring checksums in the response body.
+	// In the future, we need to use them to check image validity.
 	if res.StatusCode != 200 {
 		return fmt.Errorf("HTTP code: %d", res.StatusCode)
 	}
-	rawJson, err := ioutil.ReadAll(res.Body)
+
+	var token, endpoints []string
+	if res.Header.Get("X-Docker-Token") != "" {
+		token = res.Header["X-Docker-Token"]
+	}
+	if res.Header.Get("X-Docker-Endpoints") != "" {
+		endpoints = res.Header["X-Docker-Endpoints"]
+	} else {
+		return fmt.Errorf("Index response didn't contain any endpoints")
+	}
+
+	checksumsJson, err := ioutil.ReadAll(res.Body)
 	if err != nil {
 		return err
 	}
-	t := map[string]string{}
-	if err = json.Unmarshal(rawJson, &t); err != nil {
-		return err
-	}
-	for tag, rev := range t {
-		fmt.Fprintf(stdout, "Pulling tag %s:%s\r\n", remote, tag)
-		if err = graph.PullImage(stdout, rev, authConfig); err != nil {
+
+	// Reload the json file to make sure not to overwrite faster sums
+	err = func() error {
+		localChecksums := make(map[string]string)
+		remoteChecksums := []ImgListJson{}
+		checksumDictPth := path.Join(graph.Root, "..", "checksums")
+
+		if err := json.Unmarshal(checksumsJson, &remoteChecksums); err != nil {
 			return err
 		}
-		if err = repositories.Set(remote, tag, rev, true); err != nil {
-			return err
-		}
-	}
-	if err = repositories.Save(); err != nil {
-		return err
-	}
-	return nil
-}
 
-// Push a local image to the registry with its history if needed
-func (graph *Graph) PushImage(stdout io.Writer, imgOrig *Image, authConfig *auth.AuthConfig) error {
-	client := &http.Client{}
+		graph.lockSumFile.Lock()
+		defer graph.lockSumFile.Unlock()
 
-	// FIXME: Factorize the code
-	// FIXME: Do the puts in goroutines
-	if err := imgOrig.WalkHistory(func(img *Image) error {
-
-		jsonRaw, err := ioutil.ReadFile(path.Join(graph.Root, img.Id, "json"))
-		if err != nil {
-			return fmt.Errorf("Error while retreiving the path for {%s}: %s", img.Id, err)
-		}
-
-		fmt.Fprintf(stdout, "Pushing %s metadata\r\n", img.Id)
-
-		// FIXME: try json with UTF8
-		jsonData := strings.NewReader(string(jsonRaw))
-		req, err := http.NewRequest("PUT", REGISTRY_ENDPOINT+"/images/"+img.Id+"/json", jsonData)
-		if err != nil {
-			return err
-		}
-		req.Header.Add("Content-type", "application/json")
-		req.SetBasicAuth(authConfig.Username, authConfig.Password)
-		res, err := client.Do(req)
-		if err != nil {
-			return fmt.Errorf("Failed to upload metadata: %s", err)
-		}
-		defer res.Body.Close()
-		if res.StatusCode != 200 {
-			switch res.StatusCode {
-			case 204:
-				// Case where the image is already on the Registry
-				// FIXME: Do not be silent?
-				return nil
-			default:
-				errBody, err := ioutil.ReadAll(res.Body)
-				if err != nil {
-					errBody = []byte(err.Error())
-				}
-				return fmt.Errorf("HTTP code %d while uploading metadata: %s", res.StatusCode, errBody)
+		if checksumDict, err := ioutil.ReadFile(checksumDictPth); err == nil {
+			if err := json.Unmarshal(checksumDict, &localChecksums); err != nil {
+				return err
 			}
 		}
 
-		fmt.Fprintf(stdout, "Pushing %s fs layer\r\n", img.Id)
-		req2, err := http.NewRequest("PUT", REGISTRY_ENDPOINT+"/images/"+img.Id+"/layer", nil)
-		req2.SetBasicAuth(authConfig.Username, authConfig.Password)
-		res2, err := client.Do(req2)
-		if err != nil {
-			return fmt.Errorf("Registry returned error: %s", err)
-		}
-		res2.Body.Close()
-		if res2.StatusCode != 307 {
-			return fmt.Errorf("Registry returned unexpected HTTP status code %d, expected 307", res2.StatusCode)
-		}
-		url, err := res2.Location()
-		if err != nil || url == nil {
-			return fmt.Errorf("Failed to retrieve layer upload location: %s", err)
+		for _, elem := range remoteChecksums {
+			localChecksums[elem.Id] = elem.Checksum
 		}
 
-		// FIXME: stream the archive directly to the registry instead of buffering it on disk. This requires either:
-		//	a) Implementing S3's proprietary streaming logic, or
-		//	b) Stream directly to the registry instead of S3.
-		// I prefer option b. because it doesn't lock us into a proprietary cloud service.
-		tmpLayer, err := graph.TempLayerArchive(img.Id, Xz, stdout)
+		checksumsJson, err = json.Marshal(localChecksums)
 		if err != nil {
 			return err
 		}
-		defer os.Remove(tmpLayer.Name())
-		req3, err := http.NewRequest("PUT", url.String(), ProgressReader(tmpLayer, int(tmpLayer.Size), stdout, "Uploading %v/%v (%v)"))
-		if err != nil {
+		if err := ioutil.WriteFile(checksumDictPth, checksumsJson, 0600); err != nil {
 			return err
 		}
-		req3.ContentLength = int64(tmpLayer.Size)
-
-		req3.TransferEncoding = []string{"none"}
-		res3, err := client.Do(req3)
-		if err != nil {
-			return fmt.Errorf("Failed to upload layer: %s", err)
-		}
-		res3.Body.Close()
-		if res3.StatusCode != 200 {
-			return fmt.Errorf("Received HTTP code %d while uploading layer", res3.StatusCode)
-		}
 		return nil
-	}); err != nil {
+	}()
+	if err != nil {
 		return err
 	}
+
+	var tagsList map[string]string
+	if askedTag == "" {
+		tagsList, err = graph.getRemoteTags(stdout, endpoints, remote, token)
+		if err != nil {
+			return err
+		}
+	} else {
+		tagsList = map[string]string{askedTag: ""}
+	}
+
+	for askedTag, imgId := range tagsList {
+		fmt.Fprintf(stdout, "Resolving tag \"%s:%s\" from %s\n", remote, askedTag, endpoints)
+		success := false
+		for _, registry := range endpoints {
+			if imgId == "" {
+				imgId, err = graph.getImageForTag(stdout, askedTag, remote, registry, token)
+				if err != nil {
+					fmt.Fprintf(stdout, "Error while retrieving image for tag: %v (%v) ; "+
+						"checking next endpoint", askedTag, err)
+					continue
+				}
+			}
+
+			if err := graph.PullImage(stdout, imgId, "https://"+registry+"/v1", token); err != nil {
+				return err
+			}
+
+			if err = repositories.Set(remote, askedTag, imgId, true); err != nil {
+				return err
+			}
+			success = true
+		}
+
+		if !success {
+			return fmt.Errorf("Could not find repository on any of the indexed registries.")
+		}
+	}
+
+	if err = repositories.Save(); err != nil {
+		return err
+	}
+
+	return nil
+}
+
+// Push a local image to the registry
+func (graph *Graph) PushImage(stdout io.Writer, img *Image, registry string, token []string) error {
+	registry = "https://" + registry + "/v1"
+
+	client := graph.getHttpClient()
+	jsonRaw, err := ioutil.ReadFile(path.Join(graph.Root, img.Id, "json"))
+	if err != nil {
+		return fmt.Errorf("Error while retreiving the path for {%s}: %s", img.Id, err)
+	}
+
+	fmt.Fprintf(stdout, "Pushing %s metadata\r\n", img.Id)
+
+	// FIXME: try json with UTF8
+	jsonData := strings.NewReader(string(jsonRaw))
+	req, err := http.NewRequest("PUT", registry+"/images/"+img.Id+"/json", jsonData)
+	if err != nil {
+		return err
+	}
+	req.Header.Add("Content-type", "application/json")
+	req.Header.Set("Authorization", "Token "+strings.Join(token, ","))
+
+	checksum, err := img.Checksum()
+	if err != nil {
+		return fmt.Errorf("Error while retrieving checksum for %s: %v", img.Id, err)
+	}
+	req.Header.Set("X-Docker-Checksum", checksum)
+	Debugf("Setting checksum for %s: %s", img.ShortId(), checksum)
+	res, err := doWithCookies(client, req)
+	if err != nil {
+		return fmt.Errorf("Failed to upload metadata: %s", err)
+	}
+	defer res.Body.Close()
+	if len(res.Cookies()) > 0 {
+		client.Jar.SetCookies(req.URL, res.Cookies())
+	}
+	if res.StatusCode != 200 {
+		errBody, err := ioutil.ReadAll(res.Body)
+		if err != nil {
+			return fmt.Errorf("HTTP code %d while uploading metadata and error when"+
+				" trying to parse response body: %v", res.StatusCode, err)
+		}
+		var jsonBody map[string]string
+		if err := json.Unmarshal(errBody, &jsonBody); err != nil {
+			errBody = []byte(err.Error())
+		} else if jsonBody["error"] == "Image already exists" {
+			fmt.Fprintf(stdout, "Image %v already uploaded ; skipping\n", img.Id)
+			return nil
+		}
+		return fmt.Errorf("HTTP code %d while uploading metadata: %s", res.StatusCode, errBody)
+	}
+
+	fmt.Fprintf(stdout, "Pushing %s fs layer\r\n", img.Id)
+	root, err := img.root()
+	if err != nil {
+		return err
+	}
+
+	var layerData *TempArchive
+	// If the archive exists, use it
+	file, err := os.Open(layerArchivePath(root))
+	if err != nil {
+		if os.IsNotExist(err) {
+			// If the archive does not exist, create one from the layer
+			layerData, err = graph.TempLayerArchive(img.Id, Xz, stdout)
+			if err != nil {
+				return fmt.Errorf("Failed to generate layer archive: %s", err)
+			}
+		} else {
+			return err
+		}
+	} else {
+		defer file.Close()
+		st, err := file.Stat()
+		if err != nil {
+			return err
+		}
+		layerData = &TempArchive{file, st.Size()}
+	}
+
+	req3, err := http.NewRequest("PUT", registry+"/images/"+img.Id+"/layer",
+		ProgressReader(layerData, int(layerData.Size), stdout, ""))
+	if err != nil {
+		return err
+	}
+
+	req3.ContentLength = -1
+	req3.TransferEncoding = []string{"chunked"}
+	req3.Header.Set("Authorization", "Token "+strings.Join(token, ","))
+	res3, err := doWithCookies(client, req3)
+	if err != nil {
+		return fmt.Errorf("Failed to upload layer: %s", err)
+	}
+	defer res3.Body.Close()
+
+	if res3.StatusCode != 200 {
+		errBody, err := ioutil.ReadAll(res3.Body)
+		if err != nil {
+			return fmt.Errorf("HTTP code %d while uploading metadata and error when"+
+				" trying to parse response body: %v", res.StatusCode, err)
+		}
+		return fmt.Errorf("Received HTTP code %d while uploading layer: %s", res3.StatusCode, errBody)
+	}
 	return nil
 }
 
 // push a tag on the registry.
 // Remote has the format '<user>/<repo>
-func (graph *Graph) pushTag(remote, revision, tag string, authConfig *auth.AuthConfig) error {
-
-	// Keep this for backward compatibility
-	if tag == "" {
-		tag = "lastest"
-	}
-
+func (graph *Graph) pushTag(remote, revision, tag, registry string, token []string) error {
 	// "jsonify" the string
 	revision = "\"" + revision + "\""
+	registry = "https://" + registry + "/v1"
 
-	Debugf("Pushing tags for rev [%s] on {%s}\n", revision, REGISTRY_ENDPOINT+"/users/"+remote+"/"+tag)
+	Debugf("Pushing tags for rev [%s] on {%s}\n", revision, registry+"/users/"+remote+"/"+tag)
 
-	client := &http.Client{}
-	req, err := http.NewRequest("PUT", REGISTRY_ENDPOINT+"/users/"+remote+"/"+tag, strings.NewReader(revision))
+	client := graph.getHttpClient()
+	req, err := http.NewRequest("PUT", registry+"/repositories/"+remote+"/tags/"+tag, strings.NewReader(revision))
+	if err != nil {
+		return err
+	}
 	req.Header.Add("Content-type", "application/json")
-	req.SetBasicAuth(authConfig.Username, authConfig.Password)
-	res, err := client.Do(req)
+	req.Header.Set("Authorization", "Token "+strings.Join(token, ","))
+	req.ContentLength = int64(len(revision))
+	res, err := doWithCookies(client, req)
 	if err != nil {
 		return err
 	}
@@ -327,82 +523,226 @@ func (graph *Graph) pushTag(remote, revision, tag string, authConfig *auth.AuthC
 	if res.StatusCode != 200 && res.StatusCode != 201 {
 		return fmt.Errorf("Internal server error: %d trying to push tag %s on %s", res.StatusCode, tag, remote)
 	}
-	Debugf("Result of push tag: %d\n", res.StatusCode)
-	switch res.StatusCode {
-	default:
-		return fmt.Errorf("Error %d\n", res.StatusCode)
-	case 200:
-	case 201:
-	}
 	return nil
 }
 
-func (graph *Graph) LookupRemoteRepository(remote string, authConfig *auth.AuthConfig) bool {
-	rt := &http.Transport{Proxy: http.ProxyFromEnvironment}
-
-	var repositoryTarget string
-	// If we are asking for 'root' repository, lookup on the Library's registry
-	if strings.Index(remote, "/") == -1 {
-		repositoryTarget = REGISTRY_ENDPOINT + "/library/" + remote + "/lookup"
-	} else {
-		repositoryTarget = REGISTRY_ENDPOINT + "/users/" + remote + "/lookup"
-	}
-	Debugf("Checking for permissions on: %s", repositoryTarget)
-	req, err := http.NewRequest("PUT", repositoryTarget, strings.NewReader("\"\""))
-	if err != nil {
-		Debugf("%s\n", err)
-		return false
-	}
-	req.SetBasicAuth(authConfig.Username, authConfig.Password)
-	req.Header.Add("Content-type", "application/json")
-	res, err := rt.RoundTrip(req)
-	if err != nil || res.StatusCode != 404 {
-		errBody, err := ioutil.ReadAll(res.Body)
-		if err != nil {
-			errBody = []byte(err.Error())
-		}
-		Debugf("Lookup status code: %d (body: %s)", res.StatusCode, errBody)
-		return false
-	}
-	return true
-}
-
 // FIXME: this should really be PushTag
-func (graph *Graph) pushPrimitive(stdout io.Writer, remote, tag, imgId string, authConfig *auth.AuthConfig) error {
+func (graph *Graph) pushPrimitive(stdout io.Writer, remote, tag, imgId, registry string, token []string) error {
 	// Check if the local impage exists
 	img, err := graph.Get(imgId)
 	if err != nil {
 		fmt.Fprintf(stdout, "Skipping tag %s:%s: %s does not exist\r\n", remote, tag, imgId)
 		return nil
 	}
-	fmt.Fprintf(stdout, "Pushing tag %s:%s\r\n", remote, tag)
+	fmt.Fprintf(stdout, "Pushing image %s:%s\r\n", remote, tag)
 	// Push the image
-	if err = graph.PushImage(stdout, img, authConfig); err != nil {
+	if err = graph.PushImage(stdout, img, registry, token); err != nil {
 		return err
 	}
 	fmt.Fprintf(stdout, "Registering tag %s:%s\r\n", remote, tag)
 	// And then the tag
-	if err = graph.pushTag(remote, imgId, tag, authConfig); err != nil {
+	if err = graph.pushTag(remote, imgId, tag, registry, token); err != nil {
 		return err
 	}
 	return nil
 }
 
+// Retrieve the checksum of an image
+// Priority:
+// - Check on the stored checksums
+// - Check if the archive exists, if it does not, ask the registry
+// - If the archive does exists, process the checksum from it
+// - If the archive does not exists and not found on registry, process checksum from layer
+func (graph *Graph) getChecksum(imageId string) (string, error) {
+	// FIXME: Use in-memory map instead of reading the file each time
+	if sums, err := graph.getStoredChecksums(); err != nil {
+		return "", err
+	} else if checksum, exists := sums[imageId]; exists {
+		return checksum, nil
+	}
+
+	img, err := graph.Get(imageId)
+	if err != nil {
+		return "", err
+	}
+
+	if _, err := os.Stat(layerArchivePath(graph.imageRoot(imageId))); err != nil {
+		if os.IsNotExist(err) {
+			// TODO: Ask the registry for the checksum
+			//       As the archive is not there, it is supposed to come from a pull.
+		} else {
+			return "", err
+		}
+	}
+
+	checksum, err := img.Checksum()
+	if err != nil {
+		return "", err
+	}
+	return checksum, nil
+}
+
+type ImgListJson struct {
+	Id       string `json:"id"`
+	Checksum string `json:"checksum,omitempty"`
+	tag      string
+}
+
 // Push a repository to the registry.
 // Remote has the format '<user>/<repo>
 func (graph *Graph) PushRepository(stdout io.Writer, remote string, localRepo Repository, authConfig *auth.AuthConfig) error {
-	// Check if the remote repository exists/if we have the permission
-	if !graph.LookupRemoteRepository(remote, authConfig) {
-		return fmt.Errorf("Permission denied on repository %s\n", remote)
-	}
+	client := graph.getHttpClient()
+	// FIXME: Do not reset the cookie each time? (need to reset it in case updating latest of a repo and repushing)
+	client.Jar = cookiejar.NewCookieJar()
+	var imgList []*ImgListJson
 
-	fmt.Fprintf(stdout, "Pushing repository %s (%d tags)\r\n", remote, len(localRepo))
-	// For each image within the repo, push them
-	for tag, imgId := range localRepo {
-		if err := graph.pushPrimitive(stdout, remote, tag, imgId, authConfig); err != nil {
-			// FIXME: Continue on error?
+	fmt.Fprintf(stdout, "Processing checksums\n")
+	imageSet := make(map[string]struct{})
+
+	for tag, id := range localRepo {
+		img, err := graph.Get(id)
+		if err != nil {
 			return err
 		}
+		img.WalkHistory(func(img *Image) error {
+			if _, exists := imageSet[img.Id]; exists {
+				return nil
+			}
+			imageSet[img.Id] = struct{}{}
+			checksum, err := graph.getChecksum(img.Id)
+			if err != nil {
+				return err
+			}
+			imgList = append([]*ImgListJson{{
+				Id:       img.Id,
+				Checksum: checksum,
+				tag:      tag,
+			}}, imgList...)
+			return nil
+		})
 	}
+
+	imgListJson, err := json.Marshal(imgList)
+	if err != nil {
+		return err
+	}
+
+	Debugf("json sent: %s\n", imgListJson)
+
+	fmt.Fprintf(stdout, "Sending image list\n")
+	req, err := http.NewRequest("PUT", INDEX_ENDPOINT+"/repositories/"+remote+"/", bytes.NewReader(imgListJson))
+	if err != nil {
+		return err
+	}
+	req.SetBasicAuth(authConfig.Username, authConfig.Password)
+	req.ContentLength = int64(len(imgListJson))
+	req.Header.Set("X-Docker-Token", "true")
+
+	res, err := client.Do(req)
+	if err != nil {
+		return err
+	}
+	defer res.Body.Close()
+
+	for res.StatusCode >= 300 && res.StatusCode < 400 {
+		Debugf("Redirected to %s\n", res.Header.Get("Location"))
+		req, err = http.NewRequest("PUT", res.Header.Get("Location"), bytes.NewReader(imgListJson))
+		if err != nil {
+			return err
+		}
+		req.SetBasicAuth(authConfig.Username, authConfig.Password)
+		req.ContentLength = int64(len(imgListJson))
+		req.Header.Set("X-Docker-Token", "true")
+
+		res, err = client.Do(req)
+		if err != nil {
+			return err
+		}
+		defer res.Body.Close()
+	}
+
+	if res.StatusCode != 200 && res.StatusCode != 201 {
+		errBody, err := ioutil.ReadAll(res.Body)
+		if err != nil {
+			return err
+		}
+		return fmt.Errorf("Error: Status %d trying to push repository %s: %s", res.StatusCode, remote, errBody)
+	}
+
+	var token, endpoints []string
+	if res.Header.Get("X-Docker-Token") != "" {
+		token = res.Header["X-Docker-Token"]
+		Debugf("Auth token: %v", token)
+	} else {
+		return fmt.Errorf("Index response didn't contain an access token")
+	}
+	if res.Header.Get("X-Docker-Endpoints") != "" {
+		endpoints = res.Header["X-Docker-Endpoints"]
+	} else {
+		return fmt.Errorf("Index response didn't contain any endpoints")
+	}
+
+	// FIXME: Send only needed images
+	for _, registry := range endpoints {
+		fmt.Fprintf(stdout, "Pushing repository %s to %s (%d tags)\r\n", remote, registry, len(localRepo))
+		// For each image within the repo, push them
+		for _, elem := range imgList {
+			if err := graph.pushPrimitive(stdout, remote, elem.tag, elem.Id, registry, token); err != nil {
+				// FIXME: Continue on error?
+				return err
+			}
+		}
+	}
+
+	req2, err := http.NewRequest("PUT", INDEX_ENDPOINT+"/repositories/"+remote+"/images", bytes.NewReader(imgListJson))
+	if err != nil {
+		return err
+	}
+	req2.SetBasicAuth(authConfig.Username, authConfig.Password)
+	req2.Header["X-Docker-Endpoints"] = endpoints
+	req2.ContentLength = int64(len(imgListJson))
+	res2, err := client.Do(req2)
+	if err != nil {
+		return err
+	}
+	defer res2.Body.Close()
+	if res2.StatusCode != 204 {
+		if errBody, err := ioutil.ReadAll(res2.Body); err != nil {
+			return err
+		} else {
+			return fmt.Errorf("Error: Status %d trying to push checksums %s: %s", res2.StatusCode, remote, errBody)
+		}
+	}
+
 	return nil
 }
+
+type SearchResults struct {
+	Query      string              `json:"query"`
+	NumResults int                 `json:"num_results"`
+	Results    []map[string]string `json:"results"`
+}
+
+func (graph *Graph) SearchRepositories(stdout io.Writer, term string) (*SearchResults, error) {
+	client := graph.getHttpClient()
+	u := INDEX_ENDPOINT + "/search?q=" + url.QueryEscape(term)
+	req, err := http.NewRequest("GET", u, nil)
+	if err != nil {
+		return nil, err
+	}
+	res, err := client.Do(req)
+	if err != nil {
+		return nil, err
+	}
+	defer res.Body.Close()
+	if res.StatusCode != 200 {
+		return nil, fmt.Errorf("Unexepected status code %d", res.StatusCode)
+	}
+	rawData, err := ioutil.ReadAll(res.Body)
+	if err != nil {
+		return nil, err
+	}
+	result := new(SearchResults)
+	err = json.Unmarshal(rawData, result)
+	return result, err
+}

runtime.go

@@ -12,7 +12,6 @@ import (
 	"path"
 	"sort"
 	"strings"
-	"time"
 )
 
 type Capabilities struct {
@@ -79,114 +78,6 @@ func (runtime *Runtime) containerRoot(id string) string {
 	return path.Join(runtime.repository, id)
 }
 
-func (runtime *Runtime) mergeConfig(userConf, imageConf *Config) {
-	if userConf.Hostname != "" {
-		userConf.Hostname = imageConf.Hostname
-	}
-	if userConf.User != "" {
-		userConf.User = imageConf.User
-	}
-	if userConf.Memory == 0 {
-		userConf.Memory = imageConf.Memory
-	}
-	if userConf.MemorySwap == 0 {
-		userConf.MemorySwap = imageConf.MemorySwap
-	}
-	if userConf.PortSpecs == nil || len(userConf.PortSpecs) == 0 {
-		userConf.PortSpecs = imageConf.PortSpecs
-	}
-	if !userConf.Tty {
-		userConf.Tty = userConf.Tty
-	}
-	if !userConf.OpenStdin {
-		userConf.OpenStdin = imageConf.OpenStdin
-	}
-	if !userConf.StdinOnce {
-		userConf.StdinOnce = imageConf.StdinOnce
-	}
-	if userConf.Env == nil || len(userConf.Env) == 0 {
-		userConf.Env = imageConf.Env
-	}
-	if userConf.Cmd == nil || len(userConf.Cmd) == 0 {
-		userConf.Cmd = imageConf.Cmd
-	}
-	if userConf.Dns == nil || len(userConf.Dns) == 0 {
-		userConf.Dns = imageConf.Dns
-	}
-}
-
-func (runtime *Runtime) Create(config *Config) (*Container, error) {
-
-	// Lookup image
-	img, err := runtime.repositories.LookupImage(config.Image)
-	if err != nil {
-		return nil, err
-	}
-
-	if img.Config != nil {
-		runtime.mergeConfig(config, img.Config)
-	}
-
-	if config.Cmd == nil {
-		return nil, fmt.Errorf("No command specified")
-	}
-
-	// Generate id
-	id := GenerateId()
-	// Generate default hostname
-	// FIXME: the lxc template no longer needs to set a default hostname
-	if config.Hostname == "" {
-		config.Hostname = id[:12]
-	}
-
-	container := &Container{
-		// FIXME: we should generate the ID here instead of receiving it as an argument
-		Id:              id,
-		Created:         time.Now(),
-		Path:            config.Cmd[0],
-		Args:            config.Cmd[1:], //FIXME: de-duplicate from config
-		Config:          config,
-		Image:           img.Id, // Always use the resolved image id
-		NetworkSettings: &NetworkSettings{},
-		// FIXME: do we need to store this in the container?
-		SysInitPath: sysInitPath,
-	}
-
-	container.root = runtime.containerRoot(container.Id)
-	// Step 1: create the container directory.
-	// This doubles as a barrier to avoid race conditions.
-	if err := os.Mkdir(container.root, 0700); err != nil {
-		return nil, err
-	}
-
-	// If custom dns exists, then create a resolv.conf for the container
-	if len(config.Dns) > 0 {
-		container.ResolvConfPath = path.Join(container.root, "resolv.conf")
-		f, err := os.Create(container.ResolvConfPath)
-		if err != nil {
-			return nil, err
-		}
-		defer f.Close()
-		for _, dns := range config.Dns {
-			if _, err := f.Write([]byte("nameserver " + dns + "\n")); err != nil {
-				return nil, err
-			}
-		}
-	} else {
-		container.ResolvConfPath = "/etc/resolv.conf"
-	}
-
-	// Step 2: save the container json
-	if err := container.ToDisk(); err != nil {
-		return nil, err
-	}
-	// Step 3: register the container
-	if err := runtime.Register(container); err != nil {
-		return nil, err
-	}
-	return container, nil
-}
-
 func (runtime *Runtime) Load(id string) (*Container, error) {
 	container := &Container{root: runtime.containerRoot(id)}
 	if err := container.FromDisk(); err != nil {
@@ -287,6 +178,10 @@ func (runtime *Runtime) LogToDisk(src *writeBroadcaster, dst string) error {
 }
 
 func (runtime *Runtime) Destroy(container *Container) error {
+	if container == nil {
+		return fmt.Errorf("The given container is <nil>")
+	}
+
 	element := runtime.getContainerElement(container.Id)
 	if element == nil {
 		return fmt.Errorf("Container %v not found - maybe it was already destroyed?", container.Id)
@@ -311,33 +206,6 @@ func (runtime *Runtime) Destroy(container *Container) error {
 	return nil
 }
 
-// Commit creates a new filesystem image from the current state of a container.
-// The image can optionally be tagged into a repository
-func (runtime *Runtime) Commit(id, repository, tag, comment, author string, config *Config) (*Image, error) {
-	container := runtime.Get(id)
-	if container == nil {
-		return nil, fmt.Errorf("No such container: %s", id)
-	}
-	// FIXME: freeze the container before copying it to avoid data corruption?
-	// FIXME: this shouldn't be in commands.
-	rwTar, err := container.ExportRw()
-	if err != nil {
-		return nil, err
-	}
-	// Create a new image from the container's base layers + a new layer from container changes
-	img, err := runtime.graph.Create(rwTar, container, comment, author, config)
-	if err != nil {
-		return nil, err
-	}
-	// Register the image if needed
-	if repository != "" {
-		if err := runtime.repositories.Set(repository, tag, img.Id, true); err != nil {
-			return img, err
-		}
-	}
-	return img, nil
-}
-
 func (runtime *Runtime) restore() error {
 	dir, err := ioutil.ReadDir(runtime.repository)
 	if err != nil {

runtime_test.go

@@ -118,7 +118,10 @@ func TestRuntimeCreate(t *testing.T) {
 	if len(runtime.List()) != 0 {
 		t.Errorf("Expected 0 containers, %v found", len(runtime.List()))
 	}
-	container, err := runtime.Create(&Config{
+
+	builder := NewBuilder(runtime)
+
+	container, err := builder.Create(&Config{
 		Image: GetTestImage(runtime).Id,
 		Cmd:   []string{"ls", "-al"},
 	},
@@ -157,6 +160,26 @@ func TestRuntimeCreate(t *testing.T) {
 	if !runtime.Exists(container.Id) {
 		t.Errorf("Exists() returned false for a newly created container")
 	}
+
+	// Make sure crete with bad parameters returns an error
+	_, err = builder.Create(
+		&Config{
+			Image: GetTestImage(runtime).Id,
+		},
+	)
+	if err == nil {
+		t.Fatal("Builder.Create should throw an error when Cmd is missing")
+	}
+
+	_, err = builder.Create(
+		&Config{
+			Image: GetTestImage(runtime).Id,
+			Cmd:   []string{},
+		},
+	)
+	if err == nil {
+		t.Fatal("Builder.Create should throw an error when Cmd is empty")
+	}
 }
 
 func TestDestroy(t *testing.T) {
@@ -165,7 +188,7 @@ func TestDestroy(t *testing.T) {
 		t.Fatal(err)
 	}
 	defer nuke(runtime)
-	container, err := runtime.Create(&Config{
+	container, err := NewBuilder(runtime).Create(&Config{
 		Image: GetTestImage(runtime).Id,
 		Cmd:   []string{"ls", "-al"},
 	},
@@ -212,7 +235,10 @@ func TestGet(t *testing.T) {
 		t.Fatal(err)
 	}
 	defer nuke(runtime)
-	container1, err := runtime.Create(&Config{
+
+	builder := NewBuilder(runtime)
+
+	container1, err := builder.Create(&Config{
 		Image: GetTestImage(runtime).Id,
 		Cmd:   []string{"ls", "-al"},
 	},
@@ -222,7 +248,7 @@ func TestGet(t *testing.T) {
 	}
 	defer runtime.Destroy(container1)
 
-	container2, err := runtime.Create(&Config{
+	container2, err := builder.Create(&Config{
 		Image: GetTestImage(runtime).Id,
 		Cmd:   []string{"ls", "-al"},
 	},
@@ -232,7 +258,7 @@ func TestGet(t *testing.T) {
 	}
 	defer runtime.Destroy(container2)
 
-	container3, err := runtime.Create(&Config{
+	container3, err := builder.Create(&Config{
 		Image: GetTestImage(runtime).Id,
 		Cmd:   []string{"ls", "-al"},
 	},
@@ -262,7 +288,7 @@ func TestAllocatePortLocalhost(t *testing.T) {
 	if err != nil {
 		t.Fatal(err)
 	}
-	container, err := runtime.Create(&Config{
+	container, err := NewBuilder(runtime).Create(&Config{
 		Image:     GetTestImage(runtime).Id,
 		Cmd:       []string{"sh", "-c", "echo well hello there | nc -l -p 5555"},
 		PortSpecs: []string{"5555"},
@@ -325,8 +351,10 @@ func TestRestore(t *testing.T) {
 		t.Fatal(err)
 	}
 
+	builder := NewBuilder(runtime1)
+
 	// Create a container with one instance of docker
-	container1, err := runtime1.Create(&Config{
+	container1, err := builder.Create(&Config{
 		Image: GetTestImage(runtime1).Id,
 		Cmd:   []string{"ls", "-al"},
 	},
@@ -337,7 +365,7 @@ func TestRestore(t *testing.T) {
 	defer runtime1.Destroy(container1)
 
 	// Create a second container meant to be killed
-	container2, err := runtime1.Create(&Config{
+	container2, err := builder.Create(&Config{
 		Image:     GetTestImage(runtime1).Id,
 		Cmd:       []string{"/bin/cat"},
 		OpenStdin: true,

utils.go

@@ -2,6 +2,8 @@ package docker
 
 import (
 	"bytes"
+	"crypto/sha256"
+	"encoding/hex"
 	"errors"
 	"fmt"
 	"github.com/dotcloud/docker/rcli"
@@ -153,6 +155,13 @@ func SelfPath() string {
 	return path
 }
 
+type nopWriter struct {
+}
+
+func (w *nopWriter) Write(buf []byte) (int, error) {
+	return len(buf), nil
+}
+
 type nopWriteCloser struct {
 	io.Writer
 }
@@ -395,6 +404,14 @@ func CopyEscapable(dst io.Writer, src io.ReadCloser) (written int64, err error)
 	return written, err
 }
 
+func HashData(src io.Reader) (string, error) {
+	h := sha256.New()
+	if _, err := io.Copy(h, src); err != nil {
+		return "", err
+	}
+	return "sha256:" + hex.EncodeToString(h.Sum(nil)), nil
+}
+
 type KernelVersionInfo struct {
 	Kernel int
 	Major  int
@@ -408,7 +425,11 @@ func GetKernelVersion() (*KernelVersionInfo, error) {
 }
 
 func (k *KernelVersionInfo) String() string {
-	return fmt.Sprintf("%d.%d.%d-%s", k.Kernel, k.Major, k.Minor, k.Flavor)
+	flavor := ""
+	if len(k.Flavor) > 0 {
+		flavor = fmt.Sprintf("-%s", k.Flavor)
+	}
+	return fmt.Sprintf("%d.%d.%d%s", k.Kernel, k.Major, k.Minor, flavor)
 }
 
 // Compare two KernelVersionInfo struct.
@@ -445,7 +466,7 @@ func FindCgroupMountpoint(cgroupType string) (string, error) {
 	// cgroup /sys/fs/cgroup/devices cgroup rw,relatime,devices 0 0
 	for _, line := range strings.Split(string(output), "\n") {
 		parts := strings.Split(line, " ")
-		if parts[2] == "cgroup" {
+		if len(parts) == 6 && parts[2] == "cgroup" {
 			for _, opt := range strings.Split(parts[3], ",") {
 				if opt == cgroupType {
 					return parts[1], nil
@@ -456,3 +477,50 @@ func FindCgroupMountpoint(cgroupType string) (string, error) {
 
 	return "", fmt.Errorf("cgroup mountpoint not found for %s", cgroupType)
 }
+
+// Compare two Config struct. Do not compare the "Image" nor "Hostname" fields
+// If OpenStdin is set, then it differs
+func CompareConfig(a, b *Config) bool {
+	if a == nil || b == nil ||
+		a.OpenStdin || b.OpenStdin {
+		return false
+	}
+	if a.AttachStdout != b.AttachStdout ||
+		a.AttachStderr != b.AttachStderr ||
+		a.User != b.User ||
+		a.Memory != b.Memory ||
+		a.MemorySwap != b.MemorySwap ||
+		a.OpenStdin != b.OpenStdin ||
+		a.Tty != b.Tty {
+		return false
+	}
+	if len(a.Cmd) != len(b.Cmd) ||
+		len(a.Dns) != len(b.Dns) ||
+		len(a.Env) != len(b.Env) ||
+		len(a.PortSpecs) != len(b.PortSpecs) {
+		return false
+	}
+
+	for i := 0; i < len(a.Cmd); i++ {
+		if a.Cmd[i] != b.Cmd[i] {
+			return false
+		}
+	}
+	for i := 0; i < len(a.Dns); i++ {
+		if a.Dns[i] != b.Dns[i] {
+			return false
+		}
+	}
+	for i := 0; i < len(a.Env); i++ {
+		if a.Env[i] != b.Env[i] {
+			return false
+		}
+	}
+	for i := 0; i < len(a.PortSpecs); i++ {
+		if a.PortSpecs[i] != b.PortSpecs[i] {
+			return false
+		}
+	}
+
+	return true
+}