docker: Take a list of directives instead of a list of symlinks.
[jackhill/guix/guix.git] / guix / docker.scm
1 ;;; GNU Guix --- Functional package management for GNU
2 ;;; Copyright © 2017 Ricardo Wurmus <rekado@elephly.net>
3 ;;; Copyright © 2017, 2018, 2019 Ludovic Courtès <ludo@gnu.org>
4 ;;; Copyright © 2018 Chris Marusich <cmmarusich@gmail.com>
5 ;;;
6 ;;; This file is part of GNU Guix.
7 ;;;
8 ;;; GNU Guix is free software; you can redistribute it and/or modify it
9 ;;; under the terms of the GNU General Public License as published by
10 ;;; the Free Software Foundation; either version 3 of the License, or (at
11 ;;; your option) any later version.
12 ;;;
13 ;;; GNU Guix is distributed in the hope that it will be useful, but
14 ;;; WITHOUT ANY WARRANTY; without even the implied warranty of
15 ;;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
16 ;;; GNU General Public License for more details.
17 ;;;
18 ;;; You should have received a copy of the GNU General Public License
19 ;;; along with GNU Guix. If not, see <http://www.gnu.org/licenses/>.
20
21 (define-module (guix docker)
22 #:use-module (gcrypt hash)
23 #:use-module (guix base16)
24 #:use-module ((guix build utils)
25 #:select (mkdir-p
26 delete-file-recursively
27 with-directory-excursion
28 invoke))
29 #:use-module (gnu build install)
30 #:use-module (json) ;guile-json
31 #:use-module (srfi srfi-1)
32 #:use-module (srfi srfi-19)
33 #:use-module (srfi srfi-26)
34 #:use-module ((texinfo string-utils)
35 #:select (escape-special-chars))
36 #:use-module (rnrs bytevectors)
37 #:use-module (ice-9 ftw)
38 #:use-module (ice-9 match)
39 #:export (build-docker-image))
40
41 ;; Generate a 256-bit identifier in hexadecimal encoding for the Docker image.
42 (define docker-id
43 (compose bytevector->base16-string sha256 string->utf8))
44
45 (define (layer-diff-id layer)
46 "Generate a layer DiffID for the given LAYER archive."
47 (string-append "sha256:" (bytevector->base16-string (file-sha256 layer))))
48
49 ;; This is the semantic version of the JSON metadata schema according to
50 ;; https://github.com/docker/docker/blob/master/image/spec/v1.2.md
51 ;; It is NOT the version of the image specification.
52 (define schema-version "1.0")
53
54 (define (image-description id time)
55 "Generate a simple image description."
56 `((id . ,id)
57 (created . ,time)
58 (container_config . #nil)))
59
60 (define (generate-tag path)
61 "Generate an image tag for the given PATH."
62 (match (string-split (basename path) #\-)
63 ((hash name . rest) (string-append name ":" hash))))
64
65 (define (manifest path id)
66 "Generate a simple image manifest."
67 `#(((Config . "config.json")
68 (RepoTags . #(,(generate-tag path)))
69 (Layers . #(,(string-append id "/layer.tar"))))))
70
71 ;; According to the specifications this is required for backwards
72 ;; compatibility. It duplicates information provided by the manifest.
73 (define (repositories path id)
74 "Generate a repositories file referencing PATH and the image ID."
75 `((,(generate-tag path) . ((latest . ,id)))))
76
77 ;; See https://github.com/opencontainers/image-spec/blob/master/config.md
78 (define* (config layer time arch #:key entry-point (environment '()))
79 "Generate a minimal image configuration for the given LAYER file."
80 ;; "architecture" must be values matching "platform.arch" in the
81 ;; runtime-spec at
82 ;; https://github.com/opencontainers/runtime-spec/blob/v1.0.0-rc2/config.md#platform
83 `((architecture . ,arch)
84 (comment . "Generated by GNU Guix")
85 (created . ,time)
86 (config . ,`((env . ,(list->vector
87 (map (match-lambda
88 ((name . value)
89 (string-append name "=" value)))
90 environment)))
91 ,@(if entry-point
92 `((entrypoint . ,(list->vector entry-point)))
93 '())))
94 (container_config . #nil)
95 (os . "linux")
96 (rootfs . ((type . "layers")
97 (diff_ids . #(,(layer-diff-id layer)))))))
98
99 (define %tar-determinism-options
100 ;; GNU tar options to produce archives deterministically.
101 '("--sort=name" "--mtime=@1"
102 "--owner=root:0" "--group=root:0"))
103
104 (define directive-file
105 ;; Return the file or directory created by a 'evaluate-populate-directive'
106 ;; directive.
107 (match-lambda
108 ((source '-> target)
109 (string-trim source #\/))
110 (('directory name _ ...)
111 (string-trim name #\/))))
112
113 (define* (build-docker-image image paths prefix
114 #:key
115 (extra-files '())
116 (transformations '())
117 (system (utsname:machine (uname)))
118 database
119 entry-point
120 (environment '())
121 compressor
122 (creation-time (current-time time-utc)))
123 "Write to IMAGE a Docker image archive containing the given PATHS. PREFIX
124 must be a store path that is a prefix of any store paths in PATHS.
125
126 When DATABASE is true, copy it to /var/guix/db in the image and create
127 /var/guix/gcroots and friends.
128
129 When ENTRY-POINT is true, it must be a list of strings; it is stored as the
130 entry point in the Docker image JSON structure.
131
132 ENVIRONMENT must be a list of name/value pairs. It specifies the environment
133 variables that must be defined in the resulting image.
134
135 EXTRA-FILES must be a list of directives for 'evaluate-populate-directive'
136 describing non-store files that must be created in the image.
137
138 TRANSFORMATIONS must be a list of (OLD -> NEW) tuples describing how to
139 transform the PATHS. Any path in PATHS that begins with OLD will be rewritten
140 in the Docker image so that it begins with NEW instead. If a path is a
141 non-empty directory, then its contents will be recursively added, as well.
142
143 SYSTEM is a GNU triplet (or prefix thereof) of the system the binaries in
144 PATHS are for; it is used to produce metadata in the image. Use COMPRESSOR, a
145 command such as '(\"gzip\" \"-9n\"), to compress IMAGE. Use CREATION-TIME, a
146 SRFI-19 time-utc object, as the creation time in metadata."
147 (define (sanitize path-fragment)
148 (escape-special-chars
149 ;; GNU tar strips the leading slash off of absolute paths before applying
150 ;; the transformations, so we need to do the same, or else our
151 ;; replacements won't match any paths.
152 (string-trim path-fragment #\/)
153 ;; Escape the basic regexp special characters (see: "(sed) BRE syntax").
154 ;; We also need to escape "/" because we use it as a delimiter.
155 "/*.^$[]\\"
156 #\\))
157 (define transformation->replacement
158 (match-lambda
159 ((old '-> new)
160 ;; See "(tar) transform" for details on the expression syntax.
161 (string-append "s/^" (sanitize old) "/" (sanitize new) "/"))))
162 (define (transformations->expression transformations)
163 (let ((replacements (map transformation->replacement transformations)))
164 (string-append
165 ;; Avoid transforming link targets, since that would break some links
166 ;; (e.g., symlinks that point to an absolute store path).
167 "flags=rSH;"
168 (string-join replacements ";")
169 ;; Some paths might still have a leading path delimiter even after tar
170 ;; transforms them (e.g., "/a/b" might be transformed into "/b"), so
171 ;; strip any leading path delimiters that remain.
172 ";s,^//*,,")))
173 (define transformation-options
174 (if (eq? '() transformations)
175 '()
176 `("--transform" ,(transformations->expression transformations))))
177 (let* ((directory "/tmp/docker-image") ;temporary working directory
178 (id (docker-id prefix))
179 (time (date->string (time-utc->date creation-time) "~4"))
180 (arch (let-syntax ((cond* (syntax-rules ()
181 ((_ (pattern clause) ...)
182 (cond ((string-prefix? pattern system)
183 clause)
184 ...
185 (else
186 (error "unsupported system"
187 system)))))))
188 (cond* ("x86_64" "amd64")
189 ("i686" "386")
190 ("arm" "arm")
191 ("mips64" "mips64le")))))
192 ;; Make sure we start with a fresh, empty working directory.
193 (mkdir directory)
194 (with-directory-excursion directory
195 (mkdir id)
196 (with-directory-excursion id
197 (with-output-to-file "VERSION"
198 (lambda () (display schema-version)))
199 (with-output-to-file "json"
200 (lambda () (scm->json (image-description id time))))
201
202 ;; Create a directory for the non-store files that need to go into the
203 ;; archive.
204 (mkdir "extra")
205
206 (with-directory-excursion "extra"
207 ;; Create non-store files.
208 (for-each (cut evaluate-populate-directive <> "./")
209 extra-files)
210
211 (when database
212 ;; Initialize /var/guix, assuming PREFIX points to a profile.
213 (install-database-and-gc-roots "." database prefix))
214
215 (apply invoke "tar" "-cf" "../layer.tar"
216 `(,@transformation-options
217 ,@%tar-determinism-options
218 ,@paths
219 ,@(scandir "."
220 (lambda (file)
221 (not (member file '("." ".."))))))))
222
223 ;; It is possible for "/" to show up in the archive, especially when
224 ;; applying transformations. For example, the transformation
225 ;; "s,^/a,," will (perhaps surprisingly) cause GNU tar to transform
226 ;; the path "/a" into "/". The presence of "/" in the archive is
227 ;; probably benign, but it is definitely safe to remove it, so let's
228 ;; do that. This fails when "/" is not in the archive, so use system*
229 ;; instead of invoke to avoid an exception in that case, and redirect
230 ;; stderr to the bit bucket to avoid "Exiting with failure status"
231 ;; error messages.
232 (with-error-to-port (%make-void-port "w")
233 (lambda ()
234 (system* "tar" "--delete" "/" "-f" "layer.tar")))
235
236 (delete-file-recursively "extra"))
237
238 (with-output-to-file "config.json"
239 (lambda ()
240 (scm->json (config (string-append id "/layer.tar")
241 time arch
242 #:environment environment
243 #:entry-point entry-point))))
244 (with-output-to-file "manifest.json"
245 (lambda ()
246 (scm->json (manifest prefix id))))
247 (with-output-to-file "repositories"
248 (lambda ()
249 (scm->json (repositories prefix id)))))
250
251 (apply invoke "tar" "-cf" image "-C" directory
252 `(,@%tar-determinism-options
253 ,@(if compressor
254 (list "-I" (string-join compressor))
255 '())
256 "."))
257 (delete-file-recursively directory)))