1
0
Fork 0
mirror of https://github.com/moby/moby.git synced 2022-11-09 12:21:53 -05:00
moby--moby/api/client/build.go
Anusha Ragunathan 14215ed5a1 Make daemonbuilder.Docker leaner.
Currently builder.Backend is implemented by daemonbuilder.Docker{} for
the daemon. This registration happens in the API/server code. However,
this is too implementation specific. Ideally we should be able to specify
that docker daemon (or any other) is implementing the Backend and abstract
the implementation details. So we should remove package daemonbuilder
dependency in build_routes.go

With this change, daemonbuilder.Docker is nothing more than the daemon.
A follow on change will remove the daemonbuilder package and move relevant
methods under daemon, so that API only knows about the backend.

Also cleanup code in api/client/build.go. docker cli always performs build
context tar download for remoteURLs and sends an empty remoteContext. So
remove relevant dead code.

Signed-off-by: Anusha Ragunathan <anusha@docker.com>
2016-01-18 09:16:11 -08:00

652 lines
22 KiB
Go

package client
import (
"archive/tar"
"bufio"
"bytes"
"fmt"
"io"
"io/ioutil"
"os"
"os/exec"
"path/filepath"
"regexp"
"runtime"
"strings"
"github.com/docker/docker/api"
"github.com/docker/docker/builder/dockerignore"
Cli "github.com/docker/docker/cli"
"github.com/docker/docker/opts"
"github.com/docker/docker/pkg/archive"
"github.com/docker/docker/pkg/fileutils"
"github.com/docker/docker/pkg/gitutils"
"github.com/docker/docker/pkg/httputils"
"github.com/docker/docker/pkg/ioutils"
"github.com/docker/docker/pkg/jsonmessage"
flag "github.com/docker/docker/pkg/mflag"
"github.com/docker/docker/pkg/progress"
"github.com/docker/docker/pkg/streamformatter"
"github.com/docker/docker/pkg/urlutil"
"github.com/docker/docker/reference"
runconfigopts "github.com/docker/docker/runconfig/opts"
"github.com/docker/engine-api/types"
"github.com/docker/engine-api/types/container"
"github.com/docker/go-units"
)
type translatorFunc func(reference.NamedTagged) (reference.Canonical, error)
// CmdBuild builds a new image from the source code at a given path.
//
// If '-' is provided instead of a path or URL, Docker will build an image from either a Dockerfile or tar archive read from STDIN.
//
// Usage: docker build [OPTIONS] PATH | URL | -
func (cli *DockerCli) CmdBuild(args ...string) error {
cmd := Cli.Subcmd("build", []string{"PATH | URL | -"}, Cli.DockerCommands["build"].Description, true)
flTags := opts.NewListOpts(validateTag)
cmd.Var(&flTags, []string{"t", "-tag"}, "Name and optionally a tag in the 'name:tag' format")
suppressOutput := cmd.Bool([]string{"q", "-quiet"}, false, "Suppress the build output and print image ID on success")
noCache := cmd.Bool([]string{"-no-cache"}, false, "Do not use cache when building the image")
rm := cmd.Bool([]string{"-rm"}, true, "Remove intermediate containers after a successful build")
forceRm := cmd.Bool([]string{"-force-rm"}, false, "Always remove intermediate containers")
pull := cmd.Bool([]string{"-pull"}, false, "Always attempt to pull a newer version of the image")
dockerfileName := cmd.String([]string{"f", "-file"}, "", "Name of the Dockerfile (Default is 'PATH/Dockerfile')")
flMemoryString := cmd.String([]string{"m", "-memory"}, "", "Memory limit")
flMemorySwap := cmd.String([]string{"-memory-swap"}, "", "Swap limit equal to memory plus swap: '-1' to enable unlimited swap")
flShmSize := cmd.String([]string{"-shm-size"}, "", "Size of /dev/shm, default value is 64MB")
flCPUShares := cmd.Int64([]string{"#c", "-cpu-shares"}, 0, "CPU shares (relative weight)")
flCPUPeriod := cmd.Int64([]string{"-cpu-period"}, 0, "Limit the CPU CFS (Completely Fair Scheduler) period")
flCPUQuota := cmd.Int64([]string{"-cpu-quota"}, 0, "Limit the CPU CFS (Completely Fair Scheduler) quota")
flCPUSetCpus := cmd.String([]string{"-cpuset-cpus"}, "", "CPUs in which to allow execution (0-3, 0,1)")
flCPUSetMems := cmd.String([]string{"-cpuset-mems"}, "", "MEMs in which to allow execution (0-3, 0,1)")
flCgroupParent := cmd.String([]string{"-cgroup-parent"}, "", "Optional parent cgroup for the container")
flBuildArg := opts.NewListOpts(runconfigopts.ValidateEnv)
cmd.Var(&flBuildArg, []string{"-build-arg"}, "Set build-time variables")
isolation := cmd.String([]string{"-isolation"}, "", "Container isolation level")
ulimits := make(map[string]*units.Ulimit)
flUlimits := runconfigopts.NewUlimitOpt(&ulimits)
cmd.Var(flUlimits, []string{"-ulimit"}, "Ulimit options")
cmd.Require(flag.Exact, 1)
// For trusted pull on "FROM <image>" instruction.
addTrustedFlags(cmd, true)
cmd.ParseFlags(args, true)
var (
context io.ReadCloser
err error
)
_, err = exec.LookPath("git")
hasGit := err == nil
specifiedContext := cmd.Arg(0)
var (
contextDir string
tempDir string
relDockerfile string
progBuff io.Writer
buildBuff io.Writer
)
progBuff = cli.out
buildBuff = cli.out
if *suppressOutput {
progBuff = bytes.NewBuffer(nil)
buildBuff = bytes.NewBuffer(nil)
}
switch {
case specifiedContext == "-":
context, relDockerfile, err = getContextFromReader(cli.in, *dockerfileName)
case urlutil.IsGitURL(specifiedContext) && hasGit:
tempDir, relDockerfile, err = getContextFromGitURL(specifiedContext, *dockerfileName)
case urlutil.IsURL(specifiedContext):
context, relDockerfile, err = getContextFromURL(progBuff, specifiedContext, *dockerfileName)
default:
contextDir, relDockerfile, err = getContextFromLocalDir(specifiedContext, *dockerfileName)
}
if err != nil {
if *suppressOutput && urlutil.IsURL(specifiedContext) {
fmt.Fprintln(cli.err, progBuff)
}
return fmt.Errorf("unable to prepare context: %s", err)
}
if tempDir != "" {
defer os.RemoveAll(tempDir)
contextDir = tempDir
}
if context == nil {
// And canonicalize dockerfile name to a platform-independent one
relDockerfile, err = archive.CanonicalTarNameForPath(relDockerfile)
if err != nil {
return fmt.Errorf("cannot canonicalize dockerfile path %s: %v", relDockerfile, err)
}
f, err := os.Open(filepath.Join(contextDir, ".dockerignore"))
if err != nil && !os.IsNotExist(err) {
return err
}
var excludes []string
if err == nil {
excludes, err = dockerignore.ReadAll(f)
if err != nil {
return err
}
}
if err := validateContextDirectory(contextDir, excludes); err != nil {
return fmt.Errorf("Error checking context: '%s'.", err)
}
// If .dockerignore mentions .dockerignore or the Dockerfile
// then make sure we send both files over to the daemon
// because Dockerfile is, obviously, needed no matter what, and
// .dockerignore is needed to know if either one needs to be
// removed. The daemon will remove them for us, if needed, after it
// parses the Dockerfile. Ignore errors here, as they will have been
// caught by validateContextDirectory above.
var includes = []string{"."}
keepThem1, _ := fileutils.Matches(".dockerignore", excludes)
keepThem2, _ := fileutils.Matches(relDockerfile, excludes)
if keepThem1 || keepThem2 {
includes = append(includes, ".dockerignore", relDockerfile)
}
context, err = archive.TarWithOptions(contextDir, &archive.TarOptions{
Compression: archive.Uncompressed,
ExcludePatterns: excludes,
IncludeFiles: includes,
})
if err != nil {
return err
}
}
var resolvedTags []*resolvedTag
if isTrusted() {
// Wrap the tar archive to replace the Dockerfile entry with the rewritten
// Dockerfile which uses trusted pulls.
context = replaceDockerfileTarWrapper(context, relDockerfile, cli.trustedReference, &resolvedTags)
}
// Setup an upload progress bar
progressOutput := streamformatter.NewStreamFormatter().NewProgressOutput(progBuff, true)
var body io.Reader = progress.NewProgressReader(context, progressOutput, 0, "", "Sending build context to Docker daemon")
var memory int64
if *flMemoryString != "" {
parsedMemory, err := units.RAMInBytes(*flMemoryString)
if err != nil {
return err
}
memory = parsedMemory
}
var memorySwap int64
if *flMemorySwap != "" {
if *flMemorySwap == "-1" {
memorySwap = -1
} else {
parsedMemorySwap, err := units.RAMInBytes(*flMemorySwap)
if err != nil {
return err
}
memorySwap = parsedMemorySwap
}
}
var shmSize int64
if *flShmSize != "" {
shmSize, err = units.RAMInBytes(*flShmSize)
if err != nil {
return err
}
}
options := types.ImageBuildOptions{
Context: body,
Memory: memory,
MemorySwap: memorySwap,
Tags: flTags.GetAll(),
SuppressOutput: *suppressOutput,
NoCache: *noCache,
Remove: *rm,
ForceRemove: *forceRm,
PullParent: *pull,
IsolationLevel: container.IsolationLevel(*isolation),
CPUSetCPUs: *flCPUSetCpus,
CPUSetMems: *flCPUSetMems,
CPUShares: *flCPUShares,
CPUQuota: *flCPUQuota,
CPUPeriod: *flCPUPeriod,
CgroupParent: *flCgroupParent,
Dockerfile: relDockerfile,
ShmSize: shmSize,
Ulimits: flUlimits.GetList(),
BuildArgs: runconfigopts.ConvertKVStringsToMap(flBuildArg.GetAll()),
AuthConfigs: cli.configFile.AuthConfigs,
}
response, err := cli.client.ImageBuild(options)
if err != nil {
return err
}
err = jsonmessage.DisplayJSONMessagesStream(response.Body, buildBuff, cli.outFd, cli.isTerminalOut, nil)
if err != nil {
if jerr, ok := err.(*jsonmessage.JSONError); ok {
// If no error code is set, default to 1
if jerr.Code == 0 {
jerr.Code = 1
}
if *suppressOutput {
fmt.Fprintf(cli.err, "%s%s", progBuff, buildBuff)
}
return Cli.StatusError{Status: jerr.Message, StatusCode: jerr.Code}
}
}
// Windows: show error message about modified file permissions if the
// daemon isn't running Windows.
if response.OSType != "windows" && runtime.GOOS == "windows" {
fmt.Fprintln(cli.err, `SECURITY WARNING: You are building a Docker image from Windows against a non-Windows Docker host. All files and directories added to build context will have '-rwxr-xr-x' permissions. It is recommended to double check and reset permissions for sensitive files and directories.`)
}
// Everything worked so if -q was provided the output from the daemon
// should be just the image ID and we'll print that to stdout.
if *suppressOutput {
fmt.Fprintf(cli.out, "%s", buildBuff)
}
if isTrusted() {
// Since the build was successful, now we must tag any of the resolved
// images from the above Dockerfile rewrite.
for _, resolved := range resolvedTags {
if err := cli.tagTrusted(resolved.digestRef, resolved.tagRef); err != nil {
return err
}
}
}
return nil
}
// validateContextDirectory checks if all the contents of the directory
// can be read and returns an error if some files can't be read
// symlinks which point to non-existing files don't trigger an error
func validateContextDirectory(srcPath string, excludes []string) error {
contextRoot, err := getContextRoot(srcPath)
if err != nil {
return err
}
return filepath.Walk(contextRoot, func(filePath string, f os.FileInfo, err error) error {
// skip this directory/file if it's not in the path, it won't get added to the context
if relFilePath, err := filepath.Rel(contextRoot, filePath); err != nil {
return err
} else if skip, err := fileutils.Matches(relFilePath, excludes); err != nil {
return err
} else if skip {
if f.IsDir() {
return filepath.SkipDir
}
return nil
}
if err != nil {
if os.IsPermission(err) {
return fmt.Errorf("can't stat '%s'", filePath)
}
if os.IsNotExist(err) {
return nil
}
return err
}
// skip checking if symlinks point to non-existing files, such symlinks can be useful
// also skip named pipes, because they hanging on open
if f.Mode()&(os.ModeSymlink|os.ModeNamedPipe) != 0 {
return nil
}
if !f.IsDir() {
currentFile, err := os.Open(filePath)
if err != nil && os.IsPermission(err) {
return fmt.Errorf("no permission to read from '%s'", filePath)
}
currentFile.Close()
}
return nil
})
}
// validateTag checks if the given image name can be resolved.
func validateTag(rawRepo string) (string, error) {
_, err := reference.ParseNamed(rawRepo)
if err != nil {
return "", err
}
return rawRepo, nil
}
// isUNC returns true if the path is UNC (one starting \\). It always returns
// false on Linux.
func isUNC(path string) bool {
return runtime.GOOS == "windows" && strings.HasPrefix(path, `\\`)
}
// getDockerfileRelPath uses the given context directory for a `docker build`
// and returns the absolute path to the context directory, the relative path of
// the dockerfile in that context directory, and a non-nil error on success.
func getDockerfileRelPath(givenContextDir, givenDockerfile string) (absContextDir, relDockerfile string, err error) {
if absContextDir, err = filepath.Abs(givenContextDir); err != nil {
return "", "", fmt.Errorf("unable to get absolute context directory: %v", err)
}
// The context dir might be a symbolic link, so follow it to the actual
// target directory.
//
// FIXME. We use isUNC (always false on non-Windows platforms) to workaround
// an issue in golang. On Windows, EvalSymLinks does not work on UNC file
// paths (those starting with \\). This hack means that when using links
// on UNC paths, they will not be followed.
if !isUNC(absContextDir) {
absContextDir, err = filepath.EvalSymlinks(absContextDir)
if err != nil {
return "", "", fmt.Errorf("unable to evaluate symlinks in context path: %v", err)
}
}
stat, err := os.Lstat(absContextDir)
if err != nil {
return "", "", fmt.Errorf("unable to stat context directory %q: %v", absContextDir, err)
}
if !stat.IsDir() {
return "", "", fmt.Errorf("context must be a directory: %s", absContextDir)
}
absDockerfile := givenDockerfile
if absDockerfile == "" {
// No -f/--file was specified so use the default relative to the
// context directory.
absDockerfile = filepath.Join(absContextDir, api.DefaultDockerfileName)
// Just to be nice ;-) look for 'dockerfile' too but only
// use it if we found it, otherwise ignore this check
if _, err = os.Lstat(absDockerfile); os.IsNotExist(err) {
altPath := filepath.Join(absContextDir, strings.ToLower(api.DefaultDockerfileName))
if _, err = os.Lstat(altPath); err == nil {
absDockerfile = altPath
}
}
}
// If not already an absolute path, the Dockerfile path should be joined to
// the base directory.
if !filepath.IsAbs(absDockerfile) {
absDockerfile = filepath.Join(absContextDir, absDockerfile)
}
// Evaluate symlinks in the path to the Dockerfile too.
//
// FIXME. We use isUNC (always false on non-Windows platforms) to workaround
// an issue in golang. On Windows, EvalSymLinks does not work on UNC file
// paths (those starting with \\). This hack means that when using links
// on UNC paths, they will not be followed.
if !isUNC(absDockerfile) {
absDockerfile, err = filepath.EvalSymlinks(absDockerfile)
if err != nil {
return "", "", fmt.Errorf("unable to evaluate symlinks in Dockerfile path: %v", err)
}
}
if _, err := os.Lstat(absDockerfile); err != nil {
if os.IsNotExist(err) {
return "", "", fmt.Errorf("Cannot locate Dockerfile: %q", absDockerfile)
}
return "", "", fmt.Errorf("unable to stat Dockerfile: %v", err)
}
if relDockerfile, err = filepath.Rel(absContextDir, absDockerfile); err != nil {
return "", "", fmt.Errorf("unable to get relative Dockerfile path: %v", err)
}
if strings.HasPrefix(relDockerfile, ".."+string(filepath.Separator)) {
return "", "", fmt.Errorf("The Dockerfile (%s) must be within the build context (%s)", givenDockerfile, givenContextDir)
}
return absContextDir, relDockerfile, nil
}
// writeToFile copies from the given reader and writes it to a file with the
// given filename.
func writeToFile(r io.Reader, filename string) error {
file, err := os.OpenFile(filename, os.O_CREATE|os.O_WRONLY|os.O_TRUNC, os.FileMode(0600))
if err != nil {
return fmt.Errorf("unable to create file: %v", err)
}
defer file.Close()
if _, err := io.Copy(file, r); err != nil {
return fmt.Errorf("unable to write file: %v", err)
}
return nil
}
// getContextFromReader will read the contents of the given reader as either a
// Dockerfile or tar archive. Returns a tar archive used as a context and a
// path to the Dockerfile inside the tar.
func getContextFromReader(r io.ReadCloser, dockerfileName string) (out io.ReadCloser, relDockerfile string, err error) {
buf := bufio.NewReader(r)
magic, err := buf.Peek(archive.HeaderSize)
if err != nil && err != io.EOF {
return nil, "", fmt.Errorf("failed to peek context header from STDIN: %v", err)
}
if archive.IsArchive(magic) {
return ioutils.NewReadCloserWrapper(buf, func() error { return r.Close() }), dockerfileName, nil
}
// Input should be read as a Dockerfile.
tmpDir, err := ioutil.TempDir("", "docker-build-context-")
if err != nil {
return nil, "", fmt.Errorf("unbale to create temporary context directory: %v", err)
}
f, err := os.Create(filepath.Join(tmpDir, api.DefaultDockerfileName))
if err != nil {
return nil, "", err
}
_, err = io.Copy(f, buf)
if err != nil {
f.Close()
return nil, "", err
}
if err := f.Close(); err != nil {
return nil, "", err
}
if err := r.Close(); err != nil {
return nil, "", err
}
tar, err := archive.Tar(tmpDir, archive.Uncompressed)
if err != nil {
return nil, "", err
}
return ioutils.NewReadCloserWrapper(tar, func() error {
err := tar.Close()
os.RemoveAll(tmpDir)
return err
}), api.DefaultDockerfileName, nil
}
// getContextFromGitURL uses a Git URL as context for a `docker build`. The
// git repo is cloned into a temporary directory used as the context directory.
// Returns the absolute path to the temporary context directory, the relative
// path of the dockerfile in that context directory, and a non-nil error on
// success.
func getContextFromGitURL(gitURL, dockerfileName string) (absContextDir, relDockerfile string, err error) {
if absContextDir, err = gitutils.Clone(gitURL); err != nil {
return "", "", fmt.Errorf("unable to 'git clone' to temporary context directory: %v", err)
}
return getDockerfileRelPath(absContextDir, dockerfileName)
}
// getContextFromURL uses a remote URL as context for a `docker build`. The
// remote resource is downloaded as either a Dockerfile or a tar archive.
// Returns the tar archive used for the context and a path of the
// dockerfile inside the tar.
func getContextFromURL(out io.Writer, remoteURL, dockerfileName string) (io.ReadCloser, string, error) {
response, err := httputils.Download(remoteURL)
if err != nil {
return nil, "", fmt.Errorf("unable to download remote context %s: %v", remoteURL, err)
}
progressOutput := streamformatter.NewStreamFormatter().NewProgressOutput(out, true)
// Pass the response body through a progress reader.
progReader := progress.NewProgressReader(response.Body, progressOutput, response.ContentLength, "", fmt.Sprintf("Downloading build context from remote url: %s", remoteURL))
return getContextFromReader(ioutils.NewReadCloserWrapper(progReader, func() error { return response.Body.Close() }), dockerfileName)
}
// getContextFromLocalDir uses the given local directory as context for a
// `docker build`. Returns the absolute path to the local context directory,
// the relative path of the dockerfile in that context directory, and a non-nil
// error on success.
func getContextFromLocalDir(localDir, dockerfileName string) (absContextDir, relDockerfile string, err error) {
// When using a local context directory, when the Dockerfile is specified
// with the `-f/--file` option then it is considered relative to the
// current directory and not the context directory.
if dockerfileName != "" {
if dockerfileName, err = filepath.Abs(dockerfileName); err != nil {
return "", "", fmt.Errorf("unable to get absolute path to Dockerfile: %v", err)
}
}
return getDockerfileRelPath(localDir, dockerfileName)
}
var dockerfileFromLinePattern = regexp.MustCompile(`(?i)^[\s]*FROM[ \f\r\t\v]+(?P<image>[^ \f\r\t\v\n#]+)`)
// resolvedTag records the repository, tag, and resolved digest reference
// from a Dockerfile rewrite.
type resolvedTag struct {
digestRef reference.Canonical
tagRef reference.NamedTagged
}
// rewriteDockerfileFrom rewrites the given Dockerfile by resolving images in
// "FROM <image>" instructions to a digest reference. `translator` is a
// function that takes a repository name and tag reference and returns a
// trusted digest reference.
func rewriteDockerfileFrom(dockerfile io.Reader, translator translatorFunc) (newDockerfile []byte, resolvedTags []*resolvedTag, err error) {
scanner := bufio.NewScanner(dockerfile)
buf := bytes.NewBuffer(nil)
// Scan the lines of the Dockerfile, looking for a "FROM" line.
for scanner.Scan() {
line := scanner.Text()
matches := dockerfileFromLinePattern.FindStringSubmatch(line)
if matches != nil && matches[1] != api.NoBaseImageSpecifier {
// Replace the line with a resolved "FROM repo@digest"
ref, err := reference.ParseNamed(matches[1])
if err != nil {
return nil, nil, err
}
ref = reference.WithDefaultTag(ref)
if ref, ok := ref.(reference.NamedTagged); ok && isTrusted() {
trustedRef, err := translator(ref)
if err != nil {
return nil, nil, err
}
line = dockerfileFromLinePattern.ReplaceAllLiteralString(line, fmt.Sprintf("FROM %s", trustedRef.String()))
resolvedTags = append(resolvedTags, &resolvedTag{
digestRef: trustedRef,
tagRef: ref,
})
}
}
_, err := fmt.Fprintln(buf, line)
if err != nil {
return nil, nil, err
}
}
return buf.Bytes(), resolvedTags, scanner.Err()
}
// replaceDockerfileTarWrapper wraps the given input tar archive stream and
// replaces the entry with the given Dockerfile name with the contents of the
// new Dockerfile. Returns a new tar archive stream with the replaced
// Dockerfile.
func replaceDockerfileTarWrapper(inputTarStream io.ReadCloser, dockerfileName string, translator translatorFunc, resolvedTags *[]*resolvedTag) io.ReadCloser {
pipeReader, pipeWriter := io.Pipe()
go func() {
tarReader := tar.NewReader(inputTarStream)
tarWriter := tar.NewWriter(pipeWriter)
defer inputTarStream.Close()
for {
hdr, err := tarReader.Next()
if err == io.EOF {
// Signals end of archive.
tarWriter.Close()
pipeWriter.Close()
return
}
if err != nil {
pipeWriter.CloseWithError(err)
return
}
var content io.Reader = tarReader
if hdr.Name == dockerfileName {
// This entry is the Dockerfile. Since the tar archive was
// generated from a directory on the local filesystem, the
// Dockerfile will only appear once in the archive.
var newDockerfile []byte
newDockerfile, *resolvedTags, err = rewriteDockerfileFrom(content, translator)
if err != nil {
pipeWriter.CloseWithError(err)
return
}
hdr.Size = int64(len(newDockerfile))
content = bytes.NewBuffer(newDockerfile)
}
if err := tarWriter.WriteHeader(hdr); err != nil {
pipeWriter.CloseWithError(err)
return
}
if _, err := io.Copy(tarWriter, content); err != nil {
pipeWriter.CloseWithError(err)
return
}
}
}()
return pipeReader
}