1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
| #!/bin/sh
set -u
DOCKER_ROOT="${DOCKER_ROOT:-/var/lib/docker}"
OUTPUT_BASE="${1:-/tmp}"
REPO_JSON="$DOCKER_ROOT/image/overlay2/repositories.json"
IMAGE_DB_DIR="$DOCKER_ROOT/image/overlay2/imagedb/content/sha256"
LAYERDB_DIR="$DOCKER_ROOT/image/overlay2/layerdb/sha256"
OVERLAY2_DIR="$DOCKER_ROOT/overlay2"
TMP_FILES=""
MAP_FILE=""
DIFF_FILE=""
HIST_FILE=""
MERGED_FILE=""
TOTAL_IMAGES=0
HIST_COUNT=0
CHOSEN=""
RESOLVED_CHAIN_ID=""
RESOLVED_CACHE_ID=""
info() { printf '[INFO] %s\n' "$*"; }
warn() { printf '[WARN] %s\n' "$*" >&2; }
err() { printf '[ERROR] %s\n' "$*" >&2; }
cleanup() {
for f in $TMP_FILES; do
[ -n "$f" ] && [ -e "$f" ] && rm -f "$f"
done
}
trap 'cleanup' EXIT HUP INT TERM
require_cmd() {
if ! command -v "$1" >/dev/null 2>&1; then
err "未检测到命令: $1"
return 1
fi
return 0
}
is_number() {
case "$1" in
''|*[!0-9]*) return 1 ;;
*) return 0 ;;
esac
}
sanitize_name() {
printf '%s' "$1" | sed 's/[^A-Za-z0-9._-]/_/g'
}
init_tmp_files() {
MAP_FILE="$(mktemp /tmp/image_map.XXXXXX)" || return 1
DIFF_FILE="$(mktemp /tmp/image_diff.XXXXXX)" || return 1
HIST_FILE="$(mktemp /tmp/image_hist.XXXXXX)" || return 1
MERGED_FILE="$(mktemp /tmp/image_merged.XXXXXX)" || return 1
TMP_FILES="$MAP_FILE $DIFF_FILE $HIST_FILE $MERGED_FILE"
return 0
}
load_image_mappings() {
if [ ! -f "$REPO_JSON" ]; then
err "未找到 repositories.json: $REPO_JSON"
return 1
fi
if ! jq -r '.Repositories | to_entries[] as $repo | $repo.value | to_entries[] | [.key, .value] | @tsv' "$REPO_JSON" | awk -F '\t' '!seen[$2]++' > "$MAP_FILE"; then
err "解析 repositories.json 失败"
return 1
fi
TOTAL_IMAGES="$(awk 'END{print NR+0}' "$MAP_FILE")"
if [ "$TOTAL_IMAGES" -le 0 ]; then
err "repositories.json 中未解析到镜像映射"
return 1
fi
return 0
}
show_image_mappings() {
printf '\n==== 镜像列表 ====\n'
awk -F '\t' '{printf "%3d) %s => %s\n", NR, $1, $2}' "$MAP_FILE"
}
load_image_metadata() {
image_id="$1"
image_hash="$(printf '%s' "$image_id" | sed 's/^sha256://')"
image_cfg="$IMAGE_DB_DIR/$image_hash"
if [ ! -f "$image_cfg" ]; then
err "未找到镜像配置文件: $image_cfg"
return 1
fi
if ! jq -r '.rootfs.diff_ids[]?' "$image_cfg" > "$DIFF_FILE"; then
err "读取 diff_ids 失败"
return 1
fi
if ! jq -r '.history[] | [(.created // ""), (.created_by // ""), ((.empty_layer // false)|tostring)] | @tsv' "$image_cfg" > "$HIST_FILE"; then
err "读取 history 失败"
return 1
fi
awk -F '\t' '
NR==FNR { d[++n]=$0; next }
{
created=$1; cmd=$2; empty=$3;
if (empty=="true") { diff=""; type="empty" }
else { i++; diff=d[i]; type="layer" }
printf "%d\t%s\t%s\t%s\t%s\t%s\n", FNR, type, diff, created, cmd, empty
}
' "$DIFF_FILE" "$HIST_FILE" > "$MERGED_FILE"
HIST_COUNT="$(awk 'END{print NR+0}' "$MERGED_FILE")"
if [ "$HIST_COUNT" -le 0 ]; then
err "该镜像没有可用 history"
return 1
fi
return 0
}
show_history_layers() {
printf '\n==== 历史构建命令 ====\n'
awk -F '\t' '
{
idx=$1; type=$2; diff=$3; created=$4; cmd=$5;
if (type=="empty") diff="N/A";
printf "%3s) [%s] diff_id=%s\n", idx, type, diff;
printf " created=%s\n", created;
printf " cmd=%s\n\n", cmd;
}
' "$MERGED_FILE"
}
add_choice() {
idx="$1"
max="$2"
if ! is_number "$idx"; then
return 1
fi
if [ "$idx" -lt 1 ] || [ "$idx" -gt "$max" ]; then
return 1
fi
case " $CHOSEN " in
*" $idx "*) ;;
*) CHOSEN="$CHOSEN $idx" ;;
esac
return 0
}
parse_selection() {
selection="$1"
max="$2"
CHOSEN=""
selection="$(printf '%s' "$selection" \
| tr -d '\r' \
| tr '，' ',' \
| sed 's/[[:space:]]*-[[:space:]]*/-/g; s/^[[:space:]]*//; s/[[:space:]]*$//')"
for token in $(printf '%s' "$selection" | tr ',' ' '); do
token="$(printf '%s' "$token" | sed 's/^[[:space:]]*//; s/[[:space:]]*$//')"
[ -z "$token" ] && continue
case "$token" in
*-*)
start="${token%-*}"
end="${token#*-}"
if ! is_number "$start" || ! is_number "$end"; then
return 1
fi
if [ "$start" -gt "$end" ]; then
return 1
fi
i="$start"
while [ "$i" -le "$end" ]; do
add_choice "$i" "$max" || return 1
i=$((i + 1))
done
;;
*)
add_choice "$token" "$max" || return 1
;;
esac
done
[ -n "$CHOSEN" ]
}
resolve_layer_ids() {
diff_id="$1"
RESOLVED_CHAIN_ID=""
RESOLVED_CACHE_ID=""
[ -n "$diff_id" ] || return 1
diff_path="$(grep -Rsl --fixed-strings -- "$diff_id" "$LAYERDB_DIR" 2>/dev/null | grep '/diff$' | head -n 1)"
[ -n "$diff_path" ] || return 1
RESOLVED_CHAIN_ID="$(basename "$(dirname "$diff_path")")"
cache_file="$LAYERDB_DIR/$RESOLVED_CHAIN_ID/cache-id"
[ -f "$cache_file" ] || return 1
RESOLVED_CACHE_ID="$(cat "$cache_file" 2>/dev/null)"
[ -n "$RESOLVED_CACHE_ID" ] || return 1
return 0
}
extract_copy_destination() {
cmd="$1"
dest=""
case "$cmd" in
*"#(nop) COPY "*" in "*)
dest="${cmd##* in }"
;;
*COPY*)
body="$(printf '%s\n' "$cmd" | sed 's/^.*COPY[[:space:]][[:space:]]*//; s/[[:space:]]# buildkit.*$//')"
# shellcheck disable=SC2086
set -- $body
while [ "$#" -gt 0 ]; do
case "$1" in
--*) shift ;;
*) break ;;
esac
done
if [ "$#" -gt 0 ]; then
last=""
for arg in "$@"; do
last="$arg"
done
dest="$last"
fi
;;
esac
dest="$(printf '%s' "$dest" | sed "s/,$//; s/^['\"]//; s/['\"]$//")"
printf '%s' "$dest"
}
extract_from_diff() {
diff_id="$1"
created_by="$2"
target_dir="$3"
case "$created_by" in
*COPY*) ;;
*) return 2 ;;
esac
dest_path="$(extract_copy_destination "$created_by")"
if [ -z "$dest_path" ]; then
warn "COPY 命令解析失败，跳过: $created_by"
return 1
fi
case "$dest_path" in
/*) ;;
*)
warn "COPY 目标路径不是绝对路径，跳过: $dest_path"
return 1
;;
esac
if ! resolve_layer_ids "$diff_id"; then
warn "无法通过 diff_id 找到 cache-id: $diff_id"
return 1
fi
diff_dir="$OVERLAY2_DIR/$RESOLVED_CACHE_ID/diff"
if [ ! -d "$diff_dir" ]; then
warn "diff 目录不存在: $diff_dir"
return 1
fi
rel="$(printf '%s' "$dest_path" | sed 's#^/##; s#/$##')"
if [ -z "$rel" ]; then
warn "目标路径为空，跳过: $dest_path"
return 1
fi
src="$diff_dir/$rel"
if [ -d "$src" ]; then
mkdir -p "$target_dir/$rel" || return 1
cp -a "$src"/. "$target_dir/$rel"/ || return 1
info "已提取目录: $dest_path"
return 0
fi
if [ -e "$src" ]; then
parent="$(dirname "$rel")"
mkdir -p "$target_dir/$parent" || return 1
cp -a "$src" "$target_dir/$rel" || return 1
info "已提取文件: $dest_path"
return 0
fi
warn "在 diff 中未找到目标路径，跳过: $dest_path"
return 1
}
main() {
require_cmd jq || {
err "请先安装 jq，例如: sudo apt-get install -y jq"
exit 1
}
require_cmd awk || exit 1
require_cmd grep || exit 1
require_cmd sed || exit 1
require_cmd cp || exit 1
require_cmd mktemp || exit 1
if [ ! -d "$DOCKER_ROOT" ]; then
err "DOCKER_ROOT 不存在: $DOCKER_ROOT"
exit 1
fi
mkdir -p "$OUTPUT_BASE" || {
err "无法创建输出目录: $OUTPUT_BASE"
exit 1
}
init_tmp_files || {
err "初始化临时文件失败"
exit 1
}
info "DOCKER_ROOT=$DOCKER_ROOT"
info "OUTPUT_BASE=$OUTPUT_BASE"
while :; do
load_image_mappings || exit 1
show_image_mappings
printf '\n输入镜像编号继续；输入 q 退出。\n'
printf '请选择镜像编号: '
IFS= read -r image_choice || break
case "$image_choice" in
q|Q) info "退出。"; break ;;
esac
if ! is_number "$image_choice" || [ "$image_choice" -lt 1 ] || [ "$image_choice" -gt "$TOTAL_IMAGES" ]; then
warn "镜像编号无效，请重试。"
continue
fi
image_ref="$(awk -F '\t' -v n="$image_choice" 'NR==n{print $1; exit}' "$MAP_FILE")"
image_id="$(awk -F '\t' -v n="$image_choice" 'NR==n{print $2; exit}' "$MAP_FILE")"
image_safe="$(sanitize_name "$image_ref")"
load_image_metadata "$image_id" || continue
show_history_layers
printf '\n输入层编号（支持 3,5-7），仅非 empty 层可选；输入 b 返回镜像列表。\n'
printf '请选择层编号: '
IFS= read -r layer_selection || break
case "$layer_selection" in
b|B) continue ;;
esac
if ! parse_selection "$layer_selection" "$HIST_COUNT"; then
warn "层编号无效，请输入 1-$HIST_COUNT 范围内，例如: 3,5-7"
continue
fi
invalid=0
for idx in $CHOSEN; do
type="$(awk -F '\t' -v n="$idx" 'NR==n{print $2; exit}' "$MERGED_FILE")"
if [ "$type" = "empty" ]; then
warn "第 $idx 层是 empty_layer，不能提取。"
invalid=1
fi
done
[ "$invalid" -eq 1 ] && continue
final_target="$OUTPUT_BASE/$image_safe"
mkdir -p "$final_target" || {
warn "无法创建目录: $final_target"
continue
}
printf '\n==== 选择结果 ====\n'
printf '镜像: %s\n' "$image_ref"
printf '镜像ID: %s\n' "$image_id"
printf '输出目录: %s\n' "$final_target"
printf '层详情:\n'
for idx in $CHOSEN; do
diff_id="$(awk -F '\t' -v n="$idx" 'NR==n{print $3; exit}' "$MERGED_FILE")"
cmd_text="$(awk -F '\t' -v n="$idx" 'NR==n{print $5; exit}' "$MERGED_FILE")"
chain_id="not-found"
cache_id="not-found"
if resolve_layer_ids "$diff_id"; then
chain_id="$RESOLVED_CHAIN_ID"
cache_id="$RESOLVED_CACHE_ID"
fi
printf ' - 层编号=%s\n' "$idx"
printf ' diff_id=%s\n' "$diff_id"
printf ' chain_id=%s\n' "$chain_id"
printf ' cache_id(layer id)=%s\n' "$cache_id"
printf ' history_cmd=%s\n' "$cmd_text"
done
extracted_ok=0
extracted_fail=0
for idx in $CHOSEN; do
diff_id="$(awk -F '\t' -v n="$idx" 'NR==n{print $3; exit}' "$MERGED_FILE")"
cmd_text="$(awk -F '\t' -v n="$idx" 'NR==n{print $5; exit}' "$MERGED_FILE")"
case "$cmd_text" in
*COPY*)
if extract_from_diff "$diff_id" "$cmd_text" "$final_target"; then
extracted_ok=$((extracted_ok + 1))
else
extracted_fail=$((extracted_fail + 1))
fi
;;
*)
info "第 $idx 层不是 COPY 命令，跳过提取。"
;;
esac
done
printf '\n==== 执行完成 ====\n'
printf '镜像: %s\n' "$image_ref"
printf '选中层数: %s\n' "$(printf '%s\n' "$CHOSEN" | awk '{print NF}')"
printf 'COPY 提取成功: %s\n' "$extracted_ok"
printf 'COPY 提取失败: %s\n' "$extracted_fail"
printf '提取目录: %s\n' "$final_target"
printf '%s\n' '-----------------------------'
done
}
main "$@"
|
